PDA

View Full Version : AVCHD high bitrate tests- MUD FREE footage



Pages : [1] 2 3

jobless
06-09-2010, 04:37 AM
______________UPDATE 13/06/2010_________________


See second post for FAQ for newbies and new database for settings:
If you have any questions of any kind feel free to PM me.

______________UPDATE 12/06/2010_________________


USEFUL SETTINGS AND EXPLANATIONS FOR AVCHD:

In Ptool there are 3 suggested settings A, B, and C
I should say that C settings are the famous D settings that you can find on various posts here.
Video buffer doesn't seem to have any effect, so you don't need to check this option.
Avg. bitrate for C settings is 44Mbps.

http://nikolicnemanja.com/pics/settings.jpg

IMPORTANT:

As for 'Video Bitrate Adjustment Simplified' and 'Video Bitrate Adjustment' , you need to check just one of these. However 'Video Bitrate Adjustment Simplified' is not yet tested..

If you shoot in 720 60p mode, D settings probably won’t work with high detailed scenes and camera will interrupt recoding after few seconds. For 720 MODE use the following settings that will work without problems:

Video bitrate adjustment 38000000
Overall bitrate adjustment 40000000
Limiting bitrate adjustment 42000000

For these settings avg. bitrate in 1080 MODE is 34.4 Mbps and for 720 34.5 Mbps.
There are no problems with recording of any kind either with 1080 or 720.
No MUD with this setting also.
I should say that these are stable settings and I would recommend them for normal users that want to shoot with 1080 and 720.

For 720 MODE do not use shutter speed 50 or slower ( bitrate drops to 2MBps resulting in horrible picture).
USE ONLY 60 AND FASTER.

Here you can find some examples for these settings:
http://www.nikolicnemanja.com/test6/1080samurai.MTS
http://www.nikolicnemanja.com/test6/720samurai.MTS


ABOUT FPN and NOISE IN COLOR GRADIENTS:

-Fpn is not a codec problem and probably cannot be resolved...

- Noise in color gradients is still present (because in Low detail scenes such as walls, backdrops, etc, bitrate drops down). So for now this problem is not solved, infect IQ in static low detailed shots with enhanced codec is identical of IQ of original codec.

Tripod test:
http://www.nikolicnemanja.com/test6/...1080normal.MTS
http://www.nikolicnemanja.com/test6/.../720normal.MTS

http://www.nikolicnemanja.com/test6/...-Dsettings.MTS
http://www.nikolicnemanja.com/test6/...-Dsettings.MTS

Other examples:
http://www.sendspace.com/file/amiv1h
http://www.sendspace.com/file/af64gf
http://vimeo.com/12477730
http://www.nikolicnemanja.com/test4.rar

jobless
06-09-2010, 04:41 AM
FAQ for newbies and new database for settings:

1. imagebuff has created simple datebase of PTools patch settings using Google docs / forms.
When you have a set of patch settings worth sharing with the rest of the members simply go to:
http://bit.ly/PToolsForm
and quickly fill in the form listing your settings. You get to name them for easy reference in the forum threads.
To locate a specific named patch set or a user's settings simply go to the list at:
http://bit.ly/PToolsList

If everyone will start contributing their settings to this simple database I believe those that are tracking testing (tester13 and others) will be able to compare and contrast settings and results much easier. Regular GH1 enthusiasts like me will also have an up-to-date, complete, and easy reference for setting up PTool.

If anyone has any suggestions for changes to the form or anything else please feel free to contact me directly.

Thanks
Chris - imagebuff



2. When first time installing Ptool please visit
http://www.dvxuser.com/V6/showthread.php?t=212664

3. To install Ptool go to:
http://www.dvxuser.com/V6/showthread.php?t=206788
and read carrefully first and second post.



4. When shooting with enhanced firmware your bitrate will be higher only in situations whit Hidetail or clips with lot of motion involved ( Wide angles, panning camera, fast shutter speed, footage with lot of trees and grass, etc). In Lowdetailed scene ( Shallow depth of field scenes, tripod footage, indoor shooting, etc) bitrate drops down and it's perfectly normal ( AVCHD is variable codec) so you'll probably get around 12-18 mbps. Knowing that we can assume that different settings will gave us different results based on scene in front of camera. For example, Shooting with 'C' settings will work fine, except you find your self doing Bungee jumping and want to shoot the fall with wide angle at shutter speed 200...

Ozpeter
06-09-2010, 04:49 AM
Downloading now...

But is this a true solution - to get mud free results from the GH1 we have to use bitrates way above normal, or to put it another way, way higher than other devices use for equivalent quality? (Just asking, not knocking, this does seem like a breakthrough).

Ozpeter
06-09-2010, 04:55 AM
Also... with these higher-than-intended bitrates (MJPEG or AVCHD) is there a possibly hidden overheat downside? Maybe the whole GH1 body isn't (or is??) getting hot, but perhaps certain components are being thrashed too hard and may fail prematurely? There has to be a catch!

Vitaliy Kiselev
06-09-2010, 05:00 AM
Also... with these higher-than-intended bitrates (MJPEG or AVCHD) is there a possibly hidden overheat downside? Maybe the whole GH1 body isn't (or is??) getting hot, but perhaps certain components are being thrashed too hard and may fail prematurely? There has to be a catch!

No, it won't be hot.
Sensor is not affected.
And in reality most blocks in 1080p panning are almost all intraframe.
So you have just different levels of compression for same blocks.
But LSI load must be the same. Only transfer load on SD controller is higher, but it must not cause any serious heat.

Martti Ekstrand
06-09-2010, 05:24 AM
Wow! Not only doesn't the encoding break badly in the pans, the near black areas looks quite a bit better as well when staying still and there's very little of the "half-seconds flutter" in lowlight areas.

The catch is simple: this is not for Panasonic's target consumer, a rich man going on vacation wanting to have both good stills and nice video in a simple fashion. With this kind of bitrate the SDHC card will fill up more quickly and one needs to be somewhat more careful in choosing what kind of card to buy. I can also imagine that Panasonic's monitor with built-in SDHD slots might have problems with straight-from-the-card playback.

Now for the ultimate bitrate test: THE FOLIAGE OF DEATH PAN!

spamrakuen
06-09-2010, 05:27 AM
Now for the ultimate bitrate test: THE FOLIAGE OF DEATH PAN!

LOOOOOOOL :cheesy:

bumkicho
06-09-2010, 06:05 AM
I tried with this settings and it's working.
I record 4 minutes long clip ( I stopped the camera)
Video buffer 40.000.000
Video bitrate adjustment 40000000
Overall bitrate adjustment 46000000
Limiting bitrate adjustment 50000000

Overall bit rate : 34.8 Mbps
Maximum Overall bit rate : 46.0 Mbps

I tested tester13's B setting this morning and I was able to record and play back using class 4 card. With higher setting you have here, were you able to play it back on the camera? what card do you use?

Vitaliy Kiselev
06-09-2010, 06:09 AM
It is not necessary to have such large difference between Overall and Video bitrates.
It looks like 2Mbit is enought, but more testing required.

jobless
06-09-2010, 06:14 AM
New tests:
Video buffer 40.000.000
Video bitrate adjustment - c)40000000 d)50000000 e)60000000
Overall bitrate adjustment - c)46000000 d)56000000 e)66000000
Limiting bitrate adjustment - c)50000000 d)60000000 e)70000000
c) Gives you Overall bit rate : 34.2 Mbps
Maximum Overall bit rate : 46.0 Mbps
http://www.nikolicnemanja.com/Csettings.MTS
d) Gives you Overall bit rate : 37.4 Mbps
Maximum Overall bit rate : 56.0 Mbps
http://www.nikolicnemanja.com/Dsettings.MTS
e) Gives you Overall bit rate : 4 755 Kbps
Maximum Overall bit rate : 66.0 Mbps
http://www.nikolicnemanja.com/Esettings.MTS

Unpatched Version:
http://www.nikolicnemanja.com/Unpatched.MTS

With E setting bitrate drops! I double checked all parameters and everything was ok...
Also I noted that on C and D first frame is with MUD, but others are not.. And, there is difference in sharpens between C,D and unpatched .. :)

AKED
06-09-2010, 06:27 AM
Hi have made some stills from the panning. Quite interesting:
First the pictures. Left original, right Hack. Below is a zoomed part of the pictures.

I do not see too much of a difference.
The Pan comparison, see next post.

AKED
06-09-2010, 06:30 AM
Now the panning comparison:
Now one can see, that the original is full of mud, the hacked version is absolutely mud fee.
See the full pictures and the zoomed closeups in comparison.

Wow tester13, great.

jobless
06-09-2010, 06:30 AM
Video buffer 40.000.000
Video bitrate adjustment - c)40000000 d)50000000 e)60000000 F)55000000
Overall bitrate adjustment - c)46000000 d)56000000 e)66000000 F)57000000
Limiting bitrate adjustment - c)50000000 d)60000000 e)70000000 F)63000000

F settings:
Overall bit rate : 38.3 Mbps
Maximum Overall bit rate : 57.7 Mbps

http://www.nikolicnemanja.com/Csettings.MTS
http://www.nikolicnemanja.com/Dsettings.MTS
http://www.nikolicnemanja.com/Esettings.MTS

http://www.nikolicnemanja.com/Fsettings.MTS

http://www.nikolicnemanja.com/Unpatched.MTS

alignment1
06-09-2010, 06:34 AM
Hi have made some stills from the panning. Quite interesting:
First the pictures. Left original, right Hack. Below is a zoomed part of the pictures.

I do not see too much of a difference.
The Pan comparison, see next post.

nice work guys but is nobody using a tripod?--- especially for stills, can we have tests that have identical framing?? it's a pain to try to go back and forth between shots that are framed off:cry:

pans look great!

thanks for the tests none the less:2vrolijk_08:

Vitaliy Kiselev
06-09-2010, 06:47 AM
As for test results using low level analysys:

Only D setting is ok at low level.
All others contains large waste of space as your bitrate is consumed in garbage large size spike frame.

Plus low level show it as native 23,97, but MediaInfo show as 47,952, and player also tries to play such.

bumkicho
06-09-2010, 07:02 AM
As for test results using low level analysys:

Only D setting is ok at low level.
All others contains large waste of space as your bitrate is consumed in garbage large size spike frame.

That is good to know. Man! I cannot wait to go home and test this setting out. Thanks again, tester13!

e-steve
06-09-2010, 07:04 AM
Amazing work guys!

Has anyone confirmed:

1) Length limit with these new settings?

2) With a higher mts bitrate is transcoding to ProRes going to yield larger file sizes than before, or the same sizes? Just wondering for workflow and storage issues.

This is some exciting news nonetheless!!

Martti Ekstrand
06-09-2010, 07:04 AM
I do not see too much of a difference.

Look at the window blind behind the green marquis. In the unpatched version there's a half-second flutter all over it when the I-frames update, in the patched version there's almost none. Also in motion there's much less flickering in the levels just above black, you don't see this in a still grab. Not so important in a shoot like this but for night footage it will be quite an improvement I think.

All in all when I compare this to the 5DmkII footage I edited with a couple of weeks ago it's almost on par already in jobless first test.

tsupka
06-09-2010, 07:16 AM
2) With a higher mts bitrate is transcoding to ProRes going to yield larger file sizes than before, or the same sizes? Just wondering for workflow and storage issues.



size of prores file generally depends only from type of prores codec - i.e. HQ or LT or other, fps and resolution

svecher
06-09-2010, 07:30 AM
Amazing work guys!
2) With a higher mts bitrate is transcoding to ProRes going to yield larger file sizes than before, or the same sizes? Just wondering for workflow and storage issues.

This is some exciting news nonetheless!!
Indeed...
The differences in editing (ProRes) storage requirements should be minimal. Your acquisition requirements will now change, however, as overall bitrates being tested easily double the "standard" ones. I'm curious as to how well does this new AVCHD footage stack up to 80 Mbps MJPEG results achieved earlier. More efficient storage, as well as lack of 2/4 Gb file size limit make AVCHD a more attractive option.

Impressive work, Tester13!

jobless
06-09-2010, 07:38 AM
As for test results using low level analysys:

Only D setting is ok at low level.
All others contains large waste of space as your bitrate is consumed in garbage large size spike frame.

Plus low level show it as native 23,97, but MediaInfo show as 47,952, and player also tries to play such.

Does that mean that optimal values are around D settings?
After effects shows 23,976 fps...

e-steve
06-09-2010, 07:40 AM
size of prores file generally depends only from type of prores codec - i.e. HQ or LT or other, fps and resolution

That's what I was assuming - that file sizes would stay about the same from the ProRes transcode.

It's the conversion to PRORES that typically eats up all my storage (for documentary work I have hundreds of hours). I normally put the original MTS files offline on cheaper storage, so size of the original footage isn't as much as a concern for me.

This is actually more exciting for me at the moment, since there seems to be less length limitations than the MJPEG mode!

The GH1 game is about to change. :nads:

jobless
06-09-2010, 07:41 AM
Amazing work guys!

Has anyone confirmed:

1) Length limit with these new settings?

2) With a higher mts bitrate is transcoding to ProRes going to yield larger file sizes than before, or the same sizes? Just wondering for workflow and storage issues.

This is some exciting news nonetheless!!

I recorded 2 minutes for C and D settings and then I stopped the camera...

e-steve
06-09-2010, 08:01 AM
I recorded 2 minutes for C and D settings and then I stopped the camera...

A 10 or 15 minute test would be great if someone has the time. If not, I will try to do some tests with it this weekend!

Thanks for the great testing!

Adventsam
06-09-2010, 08:29 AM
A 10 or 15 minute test would be great if someone has the time. If not, I will try to do some tests with it this weekend!

Thanks for the great testing!

It will run and run.

IanB
06-09-2010, 08:30 AM
I haven't been able to follow close enough to learn how to get the 'playback' functionality to work.

Right now, my experience is that using the Real 24p setting, and the High bitrate AVHCD settings does not allow playback for me. Others can set me straight though, because I would really love playback!

Adventsam
06-09-2010, 08:39 AM
Just tick version change(remember to increment every time), time limit(for pal people) and the settings as tester13 has indicated on pg1, that's it, use the settings above that jobless initially gave(the ones that are gooood) and everything is good to go, do not tick anything else, not native 24p if you want to playback in-camera. The quality is looking truly awesome imo, but I am a complete amateur. After that you can go with jobless d, or go straight to jobless d if you want. Tester13 has turned a huge corner, Panasonic should buy the code from him or pay him handsomely to implement, this is huge r&d cost to Pana, they could pay him as freelance imo and release as official f/w

IanB
06-09-2010, 08:42 AM
I have made an update and am using the 'real' 24p setup. I have changed the bitrate to match Jobless "C" settings and here are the results.

Steadicam Merlin, GH1, Schneider Xenon 28mm F2 (@F11/F8) and one Really nice field of flowers, this is straight out of the camera, no curves (slightly over-exposed) no contrast, nothing....

SUPER stats tells me this file is Overall Bitrate 36.7 Mbps, Maximum Overall Bitrate 46 Mbps, 23.971 FPS, 1920x1080, 4:2:0 color, Bits (Pixel*Frame) .705, Duration is 26s 958ms.

http://vimeo.com/12428914

12428914

Thanks ATL Media Group (http://www.dvxuser.com/V6/member.php?u=44593)

GMC
06-09-2010, 08:43 AM
Tester finally did it!
Mud free AVCHD....
These are settings that tester used:
Video buffer - a) 33.500.000 b) 40.000.000
Video bitrate adjustment - a) 22.000.000 b) 32.000.000
Overall bitrate adjustment - a) 24.000.000 b) 35.000.000
Limiting bitrate adjustment - a) 26.000.000 b) 40.000.000

http://www.nikolicnemanja.com/test4.rar
I started panning after 5 second of the footage...

Hi jobless, I had a look at your footage, quite good! What I noticed tough is that you did not use the native 24p option, right? Are you going to use it in the next sample footages as tester13 suggested?

Great job!!

Thanks a lot, Gunther

ATL Media Group
06-09-2010, 08:44 AM
12428914

Here.. Just put the numbers from the vimeo link between the vimeo tags in the post.
:Drogar-BigGrin(DBG)

andrzejkra
06-09-2010, 09:09 AM
What is the level of risk in using this new patch? I have a very important shoot for a short that I have been working on for a long time and the extra quality would be appreciated, but I would hate to brick my camera for unseen reasons. I know you can't give us a guarantee, but maybe a clear statement of the risk involved in applying just the high bitrate AVCHD patch so I can make a good decision.

Vitaliy Kiselev
06-09-2010, 09:12 AM
What is the level of risk in using this new patch? I have a very important shoot for a short that I have been working on for a long time and the extra quality would be appreciated, but I would hate to brick my camera for unseen reasons. I know you can't give us a guarantee, but maybe a clear statement of the risk involved in applying just the high bitrate AVCHD patch so I can make a good decision.

Download ptool and see hints for each patch.
All green is very safe.
All yellow is also tested, but not in good working form.
Very risky patches is marked in red, currently we don't have such patches.

jobless
06-09-2010, 09:44 AM
Hi jobless, I had a look at your footage, quite good! What I noticed tough is that you did not use the native 24p option, right? Are you going to use it in the next sample footages as tester13 suggested?

Great job!!

Thanks a lot, Gunther

It was native 24p footage...

DrDave
06-09-2010, 09:55 AM
Holy cow! Tester13, you totally rock! Epic hack.

GMC
06-09-2010, 09:56 AM
It was native 24p footage...

Thanks for your fast answer. I dropped it into a 23.976 1080p timeline and it showed the same 24p-in-a-60i wrapper behaviour like the original GH1 footage.

I downloaded some native 24p footage from Car3o as well and this was "real" 24p.

Strange ... Anyway, things like that will get sorted out. Thanks for testing.

Svart
06-09-2010, 10:03 AM
Here.. Just put the numbers from the vimeo link between the vimeo tags in the post.
:Drogar-BigGrin(DBG)

I just downloaded this and it looks grand. I still see very slight mud in the darker areas between the flowers but it's a whole lot better than you'd normally get!

jobless
06-09-2010, 10:06 AM
Strange... Because in after effects shows 23,986 and I don't have any problems...
No double speed play in players as tester13 reported...
I'll check again...

Vitaliy Kiselev
06-09-2010, 10:08 AM
I downloaded some native 24p footage from Car3o as well and this was "real" 24p.


Can you provide link to Car3o foorage with 23,97 footage.

Park Edwards
06-09-2010, 10:14 AM
http://www.bokehfilms.com/Gh1/

Svart
06-09-2010, 10:20 AM
Strange... Because in after effects shows 23,986 and I don't have any problems...
No double speed play in players as tester13 reported...
I'll check again...

In VLC the video seems to play at normal speed but it's reported as 47.9fps.

alignment1
06-09-2010, 10:29 AM
vimeo link shows me still muddy in the flowers- I don't see any huge improvements.

I'm thinking vimeo compression here?

Ken Steadman
06-09-2010, 10:36 AM
vimeo link shows me still muddy in the flowers- I don't see any huge improvements.

I'm thinking vimeo compression here?


I thought the same thing so then i downloaded his MTS file.

Svart
06-09-2010, 10:37 AM
yeah it's vimeo. I thought it looked funny myself so I downloaded the real file. The downloaded file looks *much* better.

I find this interesting because some files that are uploaded to vimeo do not exhibit this mud while others do even though the source material looks fine either way. I think vimeo doesn't do an equal job compressing different files that are uploaded to it.

IanB
06-09-2010, 10:37 AM
Vimeo is going to suck ass at showing high detail, Vimeo isn't any good. I use it to host the 'original file' and give you a quick peak at the content (maybe you don't think the test I did was any good because the scene wasn't high enough detail. That is why I show the vimeo, it's sort of like a preview)

To get the full effect you must download the original .MTS on the right side of the page at the bottom.

New shot will be up soon too, walking and talking interview style in the woods. Just a new perspective/scene with the same settings, but totally different than the flowers.

Good luck,
IB

svecher
06-09-2010, 10:44 AM
vimeo link shows me still muddy in the flowers- I don't see any huge improvements.

I'm thinking vimeo compression here?
Embedded Vimeo clips do not have HD button turned on. When played directly on Vimeo website IQ is much better. For careful examination you would need to download the MTS files as others have suggested.

Vitaliy Kiselev
06-09-2010, 10:49 AM
http://www.bokehfilms.com/Gh1/

Your video is the same as myn and jobeless, also 47,92fps reported by MediaInfo.
And as some players (I use ZoomPlayer) play footage at this framerate it can be quite high probability that camera is affected by same problem.
As 720p24 footage (cropped from 1080p24) show good framerate and plays fine in camera.

cowpunk52
06-09-2010, 11:02 AM
Holy mother of mercy... I just tried out the D settings, shot in FHD with kit lens at f14 - f/22, 1/50, auto-focus, waving it around like mad.

I just transcoded into prores and put it in FCP. To say the least, I'm nothing short of completely shocked, flabbergasted and amazed. Clean footage, no mud, crisp details, flawless motion blur. My my my my my - today is a good day. I see my GH1 as a completely new camera now.

Thanks to all who made this possible, including everyone who donated & tested, and especially to Tester13

cowpunk52
06-09-2010, 11:27 AM
Your video is the same as myn and jobeless, also 47,92fps reported by MediaInfo.
And as some players (I use ZoomPlayer) play footage at this framerate it can be quite high probability that camera is affected by same problem.
As 720p24 footage (cropped from 1080p24) show good framerate and plays fine in camera.

Perhaps this means something, perhaps not - but I've noticed that with native 24p clips out of the gh1 that is imported into final cut pro, FCP lists the field dominance as "Upper (Odd)" even though frame rate is "23.98 fps." For progressive clips in FCP, the field dominance column is typically marked as "None." Field dominance is usually associated with interlaced material.

billy fattey
06-09-2010, 11:28 AM
I patched using the jobless "C" settings. I am not seeing any bitrate increase. Twice I have patched and reset the camera and then shot 90 seconds of hand-held walking-around video only to find the files are only 150-160 MB. That's about 14 mbps. I downloaded the jobless flowers .mts and can see how great it looks and that it has a file size of 108MB for 25 seconds.

Am I missing something?

IanB
06-09-2010, 11:32 AM
Same settings as before guys, this time a walk & talk shot with me holding the Steadicam and giving you some skin-tones (and stubble probably!) to pixel peek.

12432686

jobless
06-09-2010, 11:37 AM
I patched using the jobless "C" settings. I am not seeing any bitrate increase. Twice I have patched and reset the camera and then shot 90 seconds of hand-held walking-around video only to find the files are only 150-160 MB. That's about 14 mbps. I downloaded the jobless flowers .mts and can see how great it looks and that it has a file size of 108MB for 25 seconds.

Am I missing something?

Avchd is variable codec. You need hi-detailed scene in order to increase the bitrate.
Actually I was experimenting with D setting in low detail scene and I got 9 426 Kbps.
I wanted to see how codec dealing with gradient color banding. It seems that is still there...
I'll post clips leter....

kiguar
06-09-2010, 11:37 AM
1080 with newest AVCHD patch.
There is no comparision with not-patched firmware but improvement is clearly visible.
Footage is not representative.
Kit lens, ISO 200...ladies and gents I have a brand new GH1-i where "i" stand for improved

I suggest downloading bigger file.

http://www.athlante.com/gh1/

Thanks to tester for his ingenious work and deditation to project.
Bravo!

alignment1
06-09-2010, 11:56 AM
Same settings as before guys, this time a walk & talk shot with me holding the Steadicam and giving you some skin-tones (and stubble probably!) to pixel peek.

12432686


lol ..."doesn't have to look like...muddy....piece o' crap...muddy...film.."

alignment1
06-09-2010, 11:59 AM
1080 with newest AVCHD patch.
There is no comparision with not-patched firmware but improvement is clearly visible.
Footage is not representative.
Kitlens, ISO 200...ladies and gents I have a brand new GH1-i where "i" stand for improved

I suggest downloading bigger file.

http://www.athlante.com/gh1/

Thanks to tester for his ingenious work and deditation to project.
Bravo!

still seeing slight mud in pavement and lots of rolling shutter (click back and forth frame by frame @ 7:02 on second file).

But the color looks vibrant- great hi's/lows./

This is great progress

alignment1
06-09-2010, 12:09 PM
1080 with newest AVCHD patch.
There is no comparision with not-patched firmware but improvement is clearly visible.
Footage is not representative.
Kitlens, ISO 200...ladies and gents I have a brand new GH1-i where "i" stand for improved

I suggest downloading bigger file.

http://www.athlante.com/gh1/

Thanks to tester for his ingenious work and deditation to project.
Bravo!

Are you using Autofocus? On the big file?

When you zoom in to leaves on wall-- I go frame for frame and every 2 frames or so---the leaves in the shadows/ or BG of the front leaves goes out of focus, in focus, out of focus, etc, etc.

kiguar
06-09-2010, 12:17 PM
Are you using Autofocus? On the big file?

yes, I used autofocus. and maybe also pressed to de-focus while recording.
rolling shutter is different story. overall I am really impressed with result.

Oedipax
06-09-2010, 12:19 PM
Same settings as before guys, this time a walk & talk shot with me holding the Steadicam and giving you some skin-tones (and stubble probably!) to pixel peek.

Great stuff, IanB! I was a bit underwhelmed by the Vimeo version but the MTS original looks spot-on. I'm going to patch my camera in a few minutes and go shoot some on my own.

One problem - when I load the MTS file into Cineform NeoScene, it rejects the file. Do you know of a way to transcode from MTS to another format like ProRes? When I've got some test footage I'll try FCP's Log and Transfer as well.

dbwolfe
06-09-2010, 12:19 PM
so my GH1 has 1.32 already installed from Panasonic. I downloaded the ptool3.exe and moded the 1.32 to increase the AVCHD bitrate and saved the moded firmware as GH1__133.bin, copied it to the SD card but I can't get the camera to update once the SD card is installed. I've tried over and over, even using different version numbers for the moded firmware. nothing works. what am I doing wrong?

bumkicho
06-09-2010, 12:23 PM
One problem - when I load the MTS file into Cineform NeoScene, it rejects the file. Do you know of a way to transcode from MTS to another format like ProRes? When I've got some test footage I'll try FCP's Log and Transfer as well.

Hmm.. I use NeoScene to transcode MTS to cineform avi all the time. It is in window version and it may be different in mac version.

Oedipax
06-09-2010, 12:29 PM
Oh I do too, but Cineform didn't work for me just now on the MTS file IanB uploaded. Wondering if it's something to do with it being from the modified firmware, or just a coincidence or something got messed up on the Vimeo side. It should be identical to the original, but maybe not.

I'm about to patch my camera and go shoot some tests so I'll have other footage to try and will report back.

alignment1
06-09-2010, 12:32 PM
Great stuff, IanB! I was a bit underwhelmed by the Vimeo version but the MTS original looks spot-on. I'm going to patch my camera in a few minutes and go shoot some on my own.

One problem - when I load the MTS file into Cineform NeoScene, it rejects the file. Do you know of a way to transcode from MTS to another format like ProRes? When I've got some test footage I'll try FCP's Log and Transfer as well.

I use clipwrap

http://clipwrap.com/

Park Edwards
06-09-2010, 12:38 PM
why are you using cineform? just to have an intermediate codec? the pulldown is already done in camera.

jobless
06-09-2010, 12:45 PM
I did some tripod tests.
I wanted to see resolution and color banding in gradients.
Also I included Mjpeg at higher bitrate

--Quality settings (E1 to E4) - 720, 412, 352, 300
--Table settings (E1 to E4) - 24, 24, 24, 24

--For AVCHD 1080 and 720 I used D settings

For Avchd 1080 D settings I got 9734 Kbps in this scene... It isn't sufficient to banding artefacts.It seems that resolution is the same and banding is still present
However I found that Mjpeg has some problems with blacks.. In Ae it has some blue tint in shadows and in player mjpeg too contrasty and seems it has less dynamic range compared to AVCHD 720 or 1080.

Download link:


http://www.nikolicnemanja.com/test6/Normal/1080normal.MTS
http://www.nikolicnemanja.com/test6/Normal/720normal.MTS
http://www.nikolicnemanja.com/test6/Normal/Mjpegnormal.MOV

http://www.nikolicnemanja.com/test6/Dsettings/1080-Dsettings.MTS
http://www.nikolicnemanja.com/test6/Dsettings/720-Dsettings.MTS
http://www.nikolicnemanja.com/test6/Dsettings/Mjpeghi.MOV

alignment1
06-09-2010, 12:48 PM
Same settings as before guys, this time a walk & talk shot with me holding the Steadicam and giving you some skin-tones (and stubble probably!) to pixel peek.

Hey Ian- this is awesome- just downloaded mts. file--looks fantastic I couldn'r find any real classic mud especially when you were making hard pivots.

btw- great 'real-world' run-n-gun example.

It gives you a good variety of texures, i.e. trees, skin tones, lens flares, whites/blacks, motion, pans, hand-held, you name it- it's in there.


It looks like you just shaved....this codec is getting so dang good- it looks like you have to shave again!!! :grin::grin:

Thanks for the efforts

Svart
06-09-2010, 01:29 PM
I did some tripod tests.
I wanted to see resolution and color banding in gradients.
Also I included Mjpeg at higher bitrate

--Quality settings (E1 to E4) - 720, 412, 352, 300
--Table settings (E1 to E4) - 24, 24, 24, 24

--For AVCHD 1080 and 720 I used D settings

For Avchd 1080 D settings I got 9734 Kbps in this scene... It isn't sufficient to banding artefacts.It seems that resolution is the same and banding is still present
However I found that Mjpeg has some problems with blacks.. In Ae it has some blue tint in shadows and in player mjpeg too contrasty and seems it has less dynamic range compared to AVCHD 720 or 1080.

Download link:


http://www.nikolicnemanja.com/test6/Normal/1080normal.MTS
http://www.nikolicnemanja.com/test6/Normal/720normal.MTS
http://www.nikolicnemanja.com/test6/Normal/Mjpegnormal.MOV

http://www.nikolicnemanja.com/test6/Dsettings/1080-Dsettings.MTS
http://www.nikolicnemanja.com/test6/Dsettings/720-Dsettings.MTS
http://www.nikolicnemanja.com/test6/Dsettings/Mjpeghi.MOV


Didn't Tester13 mention that there was something strange with the D settings or am I mistaken?

PappasArts
06-09-2010, 01:49 PM
These are settings that tester used:
Video buffer - a) 33.500.000 b) 40.000.000
Video bitrate adjustment - a) 22.000.000 b) 32.000.000
Overall bitrate adjustment - a) 24.000.000 b) 35.000.000
Limiting bitrate adjustment - a) 26.000.000 b) 40.000.000



Do any of these have an effect on mjpeg mode? How do we adjust buffer etc for mjpeg?


.

semajha
06-09-2010, 02:02 PM
what exactly is this "mud" that everyone is talking about?

jobless
06-09-2010, 02:03 PM
As for test results using low level analysys:

Only D setting is ok at low level.
All others contains large waste of space as your bitrate is consumed in garbage large size spike frame.
As Tester said...

Park Edwards
06-09-2010, 02:05 PM
what exactly is this "mud" that everyone is talking about?

search the threads or web, plenty of info out there. in short, it's where the codec breaks up when high detail images are displayed or when panning too fast.

Ian-T
06-09-2010, 02:10 PM
what exactly is this "mud" that everyone is talking about?Exactly. :laugh:

jobless
06-09-2010, 02:15 PM
Do any of these have an effect on mjpeg mode? How do we adjust buffer etc for mjpeg?


.
No they not affecting Mjpeg mode...

Adventsam
06-09-2010, 02:19 PM
I tried with this settings and it's working.
I record 4 minutes long clip ( I stopped the camera)
Video buffer 40.000.000
Video bitrate adjustment 40000000
Overall bitrate adjustment 46000000
Limiting bitrate adjustment 50000000

Overall bit rate : 34.8 Mbps
Maximum Overall bit rate : 46.0 Mbps

Other settings, incl. d(suggested as possibly a good set by tetser13) are as follows;
Video buffer 40.000.000
Video bitrate adjustment - c)40000000 d)50000000 e)60000000 F)55000000
Overall bitrate adjustment - c)46000000 d)56000000 e)66000000 F)57000000
Limiting bitrate adjustment - c)50000000 d)60000000 e)70000000 F)63000000

jobless
06-09-2010, 02:25 PM
I tried with this settings and it's working.
I record 4 minutes long clip ( I stopped the camera)
Video buffer 40.000.000
Video bitrate adjustment 40000000
Overall bitrate adjustment 46000000
Limiting bitrate adjustment 50000000

Overall bit rate : 34.8 Mbps
Maximum Overall bit rate : 46.0 Mbps

Other settings, incl. d(suggested as possibly a good set by tetser13) are as follows;
Video buffer 40.000.000
Video bitrate adjustment - c)40000000 d)50000000 e)60000000 F)55000000
Overall bitrate adjustment - c)46000000 d)56000000 e)66000000 F)57000000
Limiting bitrate adjustment - c)50000000 d)60000000 e)70000000 F)63000000

In order to see if settings are good you need to analyse with elecard software. Then you should see bitrate at low level.
As for settings c , e and f they all have spikes and they are not good( they are good but your wasting your card capacity).
Only d is good at low level...

Ozpeter
06-09-2010, 02:25 PM
Before I went to bed last night I checked the originally posted before-and-after example and try as I might, at normal speed or when stepping through frame by frame, I couldn't see any persuasive difference between the two - certainly I wouldn't be able to pick which is which if presented with the two without prior knowledge of their identity. I've just done the same in the clear light of day, and I still can't. Maybe it's not a suitable example, or there's something wrong with my eyes.

Has anything else been posted before-and-after which would be more persuasive? Just 'after' isn't a meaningful test in my book.

Isaac_Brody
06-09-2010, 02:25 PM
One problem - when I load the MTS file into Cineform NeoScene, it rejects the file. Do you know of a way to transcode from MTS to another format like ProRes? When I've got some test footage I'll try FCP's Log and Transfer as well.

Rename your mts files from lower case mts to uppercase and that should fix the neoscene issue.

Drcoffee
06-09-2010, 02:45 PM
I updated my camera with the setting B firmware for a PAL camera (also did the panasonic battery patch and 30min limit patch) and the time remaining for FHD is still 1 hour on an 8 gig card, but MJPEG HD has dropped right done to 8 mins (from the 8gig 33 mins stated in the manual)

New AVCHD footage looks to be MUD free, and at the higher data rate.

What's going on?

alignment1
06-09-2010, 02:52 PM
I updated my camera with the setting B firmware for a PAL camera (also did the panasonic battery patch and 30min limit patch) and the time remaining for FHD is still 1 hour on an 8 gig card, but MJPEG HD has dropped right done to 8 mins (from the 8gig 33 mins stated in the manual)

New AVCHD footage looks to be MUD free, and at the higher data rate.

What's going on?



8 GB.....
8 minutes of true HD .....

sounds about right to me

Drcoffee
06-09-2010, 02:55 PM
But it's the AVCHD FHD, time remaining, which should change? not the MJPEG HD for this firmware change?

svecher
06-09-2010, 02:56 PM
Has anything else been posted before-and-after which would be more persuasive? Just 'after' isn't a meaningful test in my book.
Well, jobless posted some footage in the first post. Looking at frame extracts the difference is quite clear (to my eyes). http://www.dvxuser.com/V6/showpost.php?p=2012714&postcount=12
Unpatched output is full of smeared compression artifacts, while patches has smooth "proper" motion blur. The differences are only evident when motion is present in the scene. Tripod tests (http://www.dvxuser.com/V6/showpost.php?p=2013231&postcount=63) produce identical footage (too my eyes).

Oedipax
06-09-2010, 02:56 PM
why are you using cineform? just to have an intermediate codec? the pulldown is already done in camera.

Yeah, just for transcoding. Easier than launching FCP if I just want to convert a few files (and FCP requires you keeping all the other directories/files from the card, rather than just the MTS).

Shot some footage with Tester's high quality 1080p AVCHD settings and the native 24p patch. Those files worked fine in Cineform Neoscene, so I don't know what the issue was with the other MTS files. EDIT: Just saw Isaac's post, thanks.

Results look great for the most part, I'll upload a few MTS's for you guys to check out in a few minutes.

PappasArts
06-09-2010, 03:03 PM
Before I went to bed last night I checked the originally posted before-and-after example and try as I might, at normal speed or when stepping through frame by frame, I couldn't see any persuasive difference between the two - certainly I wouldn't be able to pick which is which if presented with the two without prior knowledge of their identity. I've just done the same in the clear light of day, and I still can't. Maybe it's not a suitable example, or there's something wrong with my eyes.

Has anything else been posted before-and-after which would be more persuasive? Just 'after' isn't a meaningful test in my book.



That's true for the most part. The issue never was when the GH1 was not moving. The images looked good. However any pans or movement would render mud. This affected the moving frame.

Mjpeg is the route to be doing. Hopefully we can get the Mjpeg in 1080P at the highest bit rate; that is golden for me. I like AVCHD, however mjpeg at higher bit rate delivers a very smooth tonality. The color in mjpeg with my 70+ mbit is gorgeous.

Mjpeg I can pull apart in AE and it's doesn't give in. AVCHD, not the same. Yes 1080P can be sharper, however detail is only one slice of the pie. That's why 720P at 60Mbits is better than 1080P at 60mbits. I would rather have more bits thrown at the smaller frame, than a bigger one. It pays you back in the end with a high bit rate codec rendition that allows for better tonality and color in post and distribution. IMHO


Pappas

Ozpeter
06-09-2010, 03:03 PM
OK, I'm now seeing the problem clearly at 6 seconds in, in the "before" footage, though after that portion of that clip it's not nearly so bad. The "after" footage has no such problem at any point. I guess to be persuaded I'd need to see a much longer comparison - and for that I'd have to do it myself!

GMC
06-09-2010, 03:13 PM
Your video is the same as myn and jobeless, also 47,92fps reported by MediaInfo.
And as some players (I use ZoomPlayer) play footage at this framerate it can be quite high probability that camera is affected by same problem.
As 720p24 footage (cropped from 1080p24) show good framerate and plays fine in camera.

Yeah, funny. I was actually referring to early Car3o footage when he was for the first time showing the native 24p patch. He had cars going back and forth at night. This footage dropped to a 23.976 1080p Edius Neo 2 timeline was "real" 24p. The latest footage of jobless (first post on first page of this thread) clearly is not correct 24p, or at least Edius doesn't interpret it correctly (whatever it is).

Just relying on players (and scrubbing back and forth) to judge if it is 24p or not doesn't work (you run into all these issues with "tolerant" players and less tolerant players, plus the fact that computer screens typically run on 60Hz, thereby possibly hiding some of the issues). You have to go to a timeline in an editing software and check frame by frame in a 23.976p environment. Then you see it.

Anyway, thanks to all the efforts. We are damn close!!


UPDATE: I used the vimeo example on page 3 just now, "C" settings and Field of Flowers from IanB (downloaded the MTS file) and also this was supposed to be native 24p. It is not. Out of curiosity I applied the pulldown I usually have to do with original GH1 footage and now I get "real" 24p. Hence, I think that the native 24p patch doesn't work correctly or somehow is obsolete if high AVCHD bitrates are chosen from ptool.

Ozpeter
06-09-2010, 03:23 PM
I'll ask here rather than clutter the main thread - are we likely to see in-camera switchable bitrates in due course, so that one can go from 'consumer' quality to 'pro' quality as the occasion demands, and secondly, as I understand it the codec should vary its bitrate to meet the demands of the shot, so can the 'curve' involved be altered so that it would significantly ramp up the bitrate for (say) fast pans, but leave it more or less original bitrate for static shots?

Overall I'm thinking in terms of optimising filesize/quality tradeoffs.

Oedipax
06-09-2010, 03:26 PM
http://i45.tinypic.com/250mbk6.jpg

http://www.sendspace.com/file/amiv1h [MTS format, 79.8MB]

Okay, first clip... I shot this using the (outdated?) "B" settings from Tester13, I missed the "D" settings somehow. So the quality will probably be even better next time.

Camera stuff: 50mm Nikon f/1.4 with m4/3 adapter, Kowa 2X anamorphic mounted on front. To display it correctly you need to unsqueeze the image. In FCP, that's a -100 in the "Distort" tab under Motion; in After Effects, just interpret the clip with a 2.0 pixel aspect ratio, or you can reduce the Y scale to 50%. In Quicktime you would change the resolution to 1920x540. I was using a modified "Normal" film mode with -2 Contrast, +1 Sharpening, -2 Saturation, 0 NR. Aperture was f/11, focused near infinity, 1/50 shutter, ISO either 100 or 160. Apologies for the crappy handheld.

One thing this firmware hasn't fixed yet - I'm still getting color shifts during pans. I don't think it shows up in this clip but I'll upload another one where it does. Basically the color green suddenly changes to a different shade while the other colors in the picture seem the same. It's not a white balance thing, I had it manually set. And it seems to happen on both AVCHD and Motion JPEG.

PappasArts
06-09-2010, 03:42 PM
I'll ask here rather than clutter the main thread - are we likely to see in-camera switchable bitrates in due course, so that one can go from 'consumer' quality to 'pro' quality as the occasion demands, and secondly, as I understand it the codec should vary its bitrate to meet the demands of the shot, so can the 'curve' involved be altered so that it would significantly ramp up the bitrate for (say) fast pans, but leave it more or less original bitrate for static shots?

Overall I'm thinking in terms of optimising filesize/quality tradeoffs.

That would be good.

I would like to set a narrow Hi/low zone that the mjpeg codec never went below a certain point. So even if the scene is dark, it would still throw a strong bit rate at it- keeping the low key frame clean.

.

.

Oedipax
06-09-2010, 04:18 PM
Another clip, native 24p 1080p AVCHD, "B" settings, this time showing the color shift in green I mention above (a bug of the GH1 itself, I think, not the firmware patch).

http://www.sendspace.com/file/1bssjl [MTS format, 171mb]

Isaac_Brody
06-09-2010, 04:25 PM
Another clip, native 24p 1080p AVCHD, "B" settings, this time showing the color shift in green I mention above (a bug of the GH1 itself, I think, not the firmware patch).

http://www.sendspace.com/file/1bssjl [MTS format, 171mb]


Are you getting same results with new patch? Be interested to see if new one fixes this.

Kellar42
06-09-2010, 04:39 PM
Not to get greedy, but has anyone done any low-light testing? I'm curious if the better codec might help noise in the shadows and possibly even fixed pattern noise, although that might not be possible.

This is all very cool stuff.

Oedipax
06-09-2010, 04:50 PM
http://www.sendspace.com/file/af64gf [MTS format, 61mb]

Here's a real codec torture test. Stock 14-140 lens @ 14mm, f/11, 1/800 shutter, ISO boosted a bit to 400 or so to compensate. Then just an outrageous whip pan back and forth. No mud spotted :)

Oh and this is using the "D" settings for AVCHD. The first shot I made outdoors with the "D" settings, my camera gave a message saying it stopped recording because of write speed limitations. I had to pull the battery in order to reset. Oddly when I made the same exact shot again, it didn't give any more errors and recorded fine. It only happened on that first shot.

Oedipax
06-09-2010, 04:54 PM
Are you getting same results with new patch? Be interested to see if new one fixes this.

Haven't been able to reproduce the artifact yet, so that's a good sign, but it's always been pretty intermittent for me. I had the same issues on the stock GH1 firmware 9 months ago, so I'm skeptical that the current firmware mods will clear it up completely, but maybe they will go away just because the bitrate is so much higher.

PappasArts
06-09-2010, 04:56 PM
http://www.sendspace.com/file/af64gf [MTS format, 61mb]

Here's a real codec torture test. Stock 14-140 lens @ 14mm, f/11, 1/800 shutter, ISO boosted a bit to 400 or so to compensate. Then just an outrageous whip pan back and forth. No mud spotted :)

Oh and this is using the "D" settings for AVCHD. The first shot I made outdoors with the "D" settings, my camera gave a message saying it stopped recording because of write speed limitations. I had to pull the battery in order to reset. Oddly when I made the same exact shot again, it didn't give any more errors and recorded fine. It only happened on that first shot.


That's weird with the write speeds. What sd card? How does "D" compare to the others?

Oedipax
06-09-2010, 04:57 PM
That's weird with the write speeds. What sd card?

Transcend 16GB, Class 6

Isaac_Brody
06-09-2010, 04:58 PM
http://www.sendspace.com/file/af64gf [MTS format, 61mb]

Here's a real codec torture test. Stock 14-140 lens @ 14mm, f/11, 1/800 shutter, ISO boosted a bit to 400 or so to compensate. Then just an outrageous whip pan back and forth. No mud spotted :)

This is great stuff, it passed the torture test. Looks like "D" settings are the sweet spot. I can't believe it looks this good. Fantastic. :)

Park Edwards
06-09-2010, 05:10 PM
Yeah, funny. I was actually referring to early Car3o footage when he was for the first time showing the native 24p patch. He had cars going back and forth at night. This footage dropped to a 23.976 1080p Edius Neo 2 timeline was "real" 24p. The latest footage of jobless (first post on first page of this thread) clearly is not correct 24p, or at least Edius doesn't interpret it correctly (whatever it is).

Just relying on players (and scrubbing back and forth) to judge if it is 24p or not doesn't work (you run into all these issues with "tolerant" players and less tolerant players, plus the fact that computer screens typically run on 60Hz, thereby possibly hiding some of the issues). You have to go to a timeline in an editing software and check frame by frame in a 23.976p environment. Then you see it.

Anyway, thanks to all the efforts. We are damn close!!


UPDATE: I used the vimeo example on page 3 just now, "C" settings and Field of Flowers from IanB (downloaded the MTS file) and also this was supposed to be native 24p. It is not. Out of curiosity I applied the pulldown I usually have to do with original GH1 footage and now I get "real" 24p. Hence, I think that the native 24p patch doesn't work correctly or somehow is obsolete if high AVCHD bitrates are chosen from ptool.


I uploaded two files. One was a 24p MJPEG file with skipped frames, the other was a .mts file.
I haven't changed or patched my camera any and the link of the girl sitting in the car is from the same patch as the two files of cars passing by. Nothing's changed on my end. So you must be referring to the MJPEG file.

By any measure, the files are 47 fps and change, drop that into a 24p timeline and you'll get 24p.

rambooc1
06-09-2010, 05:13 PM
http://www.sendspace.com/file/af64gf [MTS format, 61mb]

Here's a real codec torture test. Stock 14-140 lens @ 14mm, f/11, 1/800 shutter, ISO boosted a bit to 400 or so to compensate. Then just an outrageous whip pan back and forth. No mud spotted :)

Oh and this is using the "D" settings for AVCHD. The first shot I made outdoors with the "D" settings, my camera gave a message saying it stopped recording because of write speed limitations. I had to pull the battery in order to reset. Oddly when I made the same exact shot again, it didn't give any more errors and recorded fine. It only happened on that first shot.

Wow that seals it for me. I was wondering what the hi shutter speed shots would be like. I shoot at 1/500 + 95% of the time, this is awesome quality and the skew i can handle. 59.94fps would be even better, less rolling shutter.

PappasArts
06-09-2010, 05:15 PM
http://www.sendspace.com/file/af64gf [MTS format, 61mb]

Here's a real codec torture test. Stock 14-140 lens @ 14mm, f/11, 1/800 shutter, ISO boosted a bit to 400 or so to compensate. Then just an outrageous whip pan back and forth. No mud spotted :)

Oh and this is using the "D" settings for AVCHD. The first shot I made outdoors with the "D" settings, my camera gave a message saying it stopped recording because of write speed limitations. I had to pull the battery in order to reset. Oddly when I made the same exact shot again, it didn't give any more errors and recorded fine. It only happened on that first shot.


Just viewed it. Very sharp. Nice! If this was a year ago; the GH1 would have stolen Canon's thunder.... Timing is everything.

Svart
06-09-2010, 05:19 PM
http://www.sendspace.com/file/af64gf [MTS format, 61mb]

Here's a real codec torture test. Stock 14-140 lens @ 14mm, f/11, 1/800 shutter, ISO boosted a bit to 400 or so to compensate. Then just an outrageous whip pan back and forth. No mud spotted :)

Oh and this is using the "D" settings for AVCHD. The first shot I made outdoors with the "D" settings, my camera gave a message saying it stopped recording because of write speed limitations. I had to pull the battery in order to reset. Oddly when I made the same exact shot again, it didn't give any more errors and recorded fine. It only happened on that first shot.

There is mud in the dark areas of the trees but it's very minimal compared to what we would see in these same conditions on an un-modded camera.. :laugh:

seanmcleod
06-09-2010, 05:25 PM
retracted... great job tester, donation coming!

rambooc1
06-09-2010, 05:36 PM
http://www.sendspace.com/file/af64gf [MTS format, 61mb]

Here's a real codec torture test. Stock 14-140 lens @ 14mm, f/11, 1/800 shutter, ISO boosted a bit to 400 or so to compensate. Then just an outrageous whip pan back and forth. No mud spotted :)

Oh and this is using the "D" settings for AVCHD. The first shot I made outdoors with the "D" settings, my camera gave a message saying it stopped recording because of write speed limitations. I had to pull the battery in order to reset. Oddly when I made the same exact shot again, it didn't give any more errors and recorded fine. It only happened on that first shot.

Dropping it in VLC and playing it slowed down at .1 speed is just like watching a bunch of stills in a slide show. I've not seen $50k video cameras capture footage as good as that under those conditions.

Sony Vegas reads the clip as .......
General
Name: GH1 2010.06.09 - Clip 14.MTS
Folder: C:\Users\ian\Desktop\gh1-14
Type: MPEG-2 Transport Stream
Size: 63.85 MB (65,378,304 bytes)
Created: Thursday, 10 June 2010, 10:05:42 AM
Modified: Wednesday, 9 June 2010, 6:59:56 PM
Accessed: Thursday, 10 June 2010, 10:05:42 AM
Attributes: Archive

Streams
Video: 00:00:11.472, 23.971 fps progressive, 1920x1080x12, AVC
Audio: 00:00:11.512, 48,000 Hz, Stereo, Dolby AC-3

dbwolfe
06-09-2010, 05:41 PM
be careful with D settings. I had my camera stop recording multiple times with the "stopped recording because of write speed limitations". I have a 32GB SanDisk extreme III class 10 30MB/s card so that's not the issue. I think the issue is the max bitrate number is larger than the buffer number in the D settings. if you look at the dialog in tester13's ptool recommendations there are guidelines for the numbers. I'm sticking with B for now. also, I noticed that changing the audio bitrate to any other setting besides the default causes the movies to not be playable in camera. I shoot only 720p60 and am getting about 28mbs max. definitely an improvement.

Kholi
06-09-2010, 05:49 PM
Where did you get the card from?

Vitaliy Kiselev
06-09-2010, 05:53 PM
8 GB.....
8 minutes of true HD .....

sounds about right to me

Initial time estimation is still wrong.
But remaining counter reduce by 2-3 each second :-)

dbwolfe
06-09-2010, 05:53 PM
direct from SanDisk

Phil Seastrand
06-09-2010, 06:16 PM
be careful with D settings. I had my camera stop recording multiple times with the "stopped recording because of write speed limitations". I think the issue is the max bitrate number is larger than the buffer number in the D settings. if you look at the dialog in tester13's ptool recommendations there are guidelines for the numbers.
I was going to ask this as well. Are the warnings in the tool correct and so far people are just being lucky?

Ozpeter
06-09-2010, 06:44 PM
The high-speed garden tour is indeed impressive, seen frame by frame (nice garden!). It suggests an interesting photo technique - just "hose" all over a scene and you'll have the whole thing in moments.

Looking at the file size, you'd get about 22 minutes on a 8GB card at that kind of bit rate, yes?

dbwolfe
06-09-2010, 06:54 PM
yes, I think people are just getting lucky with the D settings. I first noticed the failure when I was trying to really stress the codec. very detailed scene with lots of camera movement. once I noticed the failure I tried to repeat the record shutoff and was able to cause the same problem everytime. that's when I went back to the B settings. it makes sense, you can't have a max bitrate that's bigger than the buffer.

cowpunk52
06-09-2010, 07:07 PM
yes, I think people are just getting lucky with the D settings. I first noticed the failure when I was trying to really stress the codec. very detailed scene with lots of camera movement. once I noticed the failure I tried to repeat the record shutoff and was able to cause the same problem everytime. that's when I went back to the B settings. it makes sense, you can't have a max bitrate that's bigger than the buffer.

I must be getting pretty darn lucky, then! I'm using a Sandisk 8 GB Extreme III Class 6 card, just shot with the D setting + native 24p + 448 bps audio for 10 minutes - all handheld, quick pans on high detail at 1/500, f/11, 800 ISO with a Canon FD 50mm lens.

No write error, I just got bored and cut after 10 minutes. Didn't break the codec and didn't stress the card.

Ozpeter
06-09-2010, 07:08 PM
I've been seeing what happens when rendering the 'garden whipping' footage to 'standard' AVCHD 15mbps. There's very little degradation but half the file size. In other words, to get the high bitrate quality you don't actually have to use a high bitrate. So aren't we throwing a high bitrate at some other problem in the GH1 AVCHD codec which should be resolvable without resorting to that solution? Or is this the only practical means now to do so? Or is this not a fair comparison (encoding in an NLE vs encoding in the camera)?

It's interesting to look frame by frame from the start of that clip what happens to the tall ornamental grasses at the base of the large tree which is in the centre of the shot at the outset (it has a double trunk). They start off with clearly defined blades of foliage, but as the pan begins, they "splodge" in a way that looks more like 'mud' to me than motion blur. Maybe I'm being too picky.

PDR
06-09-2010, 07:27 PM
I've been seeing what happens when rendering the 'garden whipping' footage to 'standard' AVCHD 15mbps. There's very little degradation but half the file size. In other words, to get the high bitrate quality you don't actually have to use a high bitrate. So aren't we throwing a high bitrate at some other problem in the GH1 AVCHD codec which should be resolvable without resorting to that solution? Or is this the only practical means now to do so? Or is this not a fair comparison (encoding in an NLE vs encoding in the camera)?

It's interesting to look frame by frame from the start of that clip what happens to the tall ornamental grasses at the base of the large tree which is in the centre of the shot at the outset (it has a double trunk). They start off with clearly defined blades of foliage, but as the pan begins, they "splodge" in a way that looks more like 'mud' to me than motion blur. Maybe I'm being too picky.

Correct, you cannot compare a software encode to an onboard hardware encode. They use very different algorithms and encoding strategies.

To make it more efficient with better compression (i.e. better quality at lower filesize), you would have to implement better features like using b-frames, more reference frames, CABAC, stronger search and motion vector prediction. This is not something firmware mods can do very easily. You need new hardware for these types of changes because the CPU just can't handle it.

rambooc1
06-09-2010, 07:31 PM
I've been seeing what happens when rendering the 'garden whipping' footage to 'standard' AVCHD 15mbps. There's very little degradation but half the file size. In other words, to get the high bitrate quality you don't actually have to use a high bitrate. So aren't we throwing a high bitrate at some other problem in the GH1 AVCHD codec which should be resolvable without resorting to that solution? Or is this the only practical means now to do so? Or is this not a fair comparison (encoding in an NLE vs encoding in the camera)?

It's interesting to look frame by frame from the start of that clip what happens to the tall ornamental grasses at the base of the large tree which is in the centre of the shot at the outset (it has a double trunk). They start off with clearly defined blades of foliage, but as the pan begins, they "splodge" in a way that looks more like 'mud' to me than motion blur. Maybe I'm being too picky.

Totally different thing recording at 15mbps and downconverting higher bitrate footage. Even HD downconverted to SD looks better than footage shot at SD.

alignment1
06-09-2010, 08:07 PM
http://www.sendspace.com/file/af64gf [MTS format, 61mb]

Here's a real codec torture test. Stock 14-140 lens @ 14mm, f/11, 1/800 shutter, ISO boosted a bit to 400 or so to compensate. Then just an outrageous whip pan back and forth. No mud spotted :)

Oh and this is using the "D" settings for AVCHD. The first shot I made outdoors with the "D" settings, my camera gave a message saying it stopped recording because of write speed limitations. I had to pull the battery in order to reset. Oddly when I made the same exact shot again, it didn't give any more errors and recorded fine. It only happened on that first shot.

There is some mud but- nothing in comparison to what we would see on out the box GH1! Frame 23 is when it goes kaplooey....but who's countin'.

I'm working on some snowboard footage from this winter- could've def. used this!! The trees are a mess!!!!

However, the rolling shutter has really come to the forefront- more so than I've ever seen on GH-1.

NTL Good work!!

Svart
06-09-2010, 08:15 PM
I've been seeing what happens when rendering the 'garden whipping' footage to 'standard' AVCHD 15mbps. There's very little degradation but half the file size. In other words, to get the high bitrate quality you don't actually have to use a high bitrate. So aren't we throwing a high bitrate at some other problem in the GH1 AVCHD codec which should be resolvable without resorting to that solution? Or is this the only practical means now to do so? Or is this not a fair comparison (encoding in an NLE vs encoding in the camera)?

It's interesting to look frame by frame from the start of that clip what happens to the tall ornamental grasses at the base of the large tree which is in the centre of the shot at the outset (it has a double trunk). They start off with clearly defined blades of foliage, but as the pan begins, they "splodge" in a way that looks more like 'mud' to me than motion blur. Maybe I'm being too picky.


Yes you are right. right now the only option is to up the bitrate because as Tester13 has uncovered, there is something strange happening with the encoding. Higher bitrate helps get around this issue but does not solve the real issue at hand.

Camera Expert
06-09-2010, 11:06 PM
So what's the scoop if any as far as 720 60p is concerned? I'm shooting something today with a GH1 and an HVX200A and as much as I want to make the picture quality better in the 1080 24p mode, I really don't trust my Transcend class 6 card. The only real issues I had was on the fist big day shooting with it in which the camera stopped recording several times.

The good news is that it have never happened since then. I mean perhaps the GH1 was playing hard to get and after knowing the Transcend for a while, they don't mind being together now but I fear that may change.

I bought that card because my 16GB Sandisk class 6 got somehow lost on the first day of using it. Why the heck did it have to be that card? I have cheaper cards that I wouldn't have mind loosing like the class 2 ones. I spent a couple of weeks deciding on rather or not putting down the money to get the card was worth it and that had to happen. That card would have been perfect for these tests.

Anyway, I've decided to be a rebel and shoot my project in 720 60p.

After saying all that, another question just popped into my head that I've been wondering. Have anybody got better results using the top Panasonic cards for these high bit rate tests?

Ben_B
06-10-2010, 12:07 AM
So at present what are the best settings? The sample from the first post looks pretty darn great.

Adventsam
06-10-2010, 01:18 AM
c, works great, I have a class6 peak card and have had no problems with writing, during fast pans etc. at foliage. I think d is working too but maybe its a little more demanding, c are gooood though. Just going to look at my mjpeg footage with these settings.

sammysammy
06-10-2010, 01:19 AM
i would also love to know the best settings for the latest tester patches ..if someone can make detailed setting inside the latest ptool (what should be marked or checked and what values should be entered) , thanks !

chrisma
06-10-2010, 01:32 AM
Hey all,
while using the a,b,c, or d setting do you also patch the GOP size? if so at what settings?
what other settings do you patch ptools?
Thanks for your input

Chris

Oedipax
06-10-2010, 02:11 AM
I didn't patch the GOP size on mine, just Video Buffer, Video Bitrate, Overall Bitrate & Limiting Bitrate. Aside from those I also did the MJPEG E1-E4 (Quality & Table) and native 24p, 3rd party battery & of course version change.

jobless
06-10-2010, 02:13 AM
All of you could experiment with settings. If D settings causing crashes try to change values a bit, and then upload to someone who can control check at low level. We need to find optimal values. I'll try with new settings later.
As for C settings, They are not good... It's better to use B that it's good at low level even with smaller bitrate...

Adventsam
06-10-2010, 02:20 AM
They work very very well.

jobless
06-10-2010, 02:29 AM
Look your self...
C Settings:
http://www.nikolicnemanja.com/Csettings.jpg

D settings:
http://www.nikolicnemanja.com/Dsettings.jpg

chrisma
06-10-2010, 02:47 AM
I didn't patch the GOP size on mine, just Video Buffer, Video Bitrate, Overall Bitrate & Limiting Bitrate. Aside from those I also did the MJPEG E1-E4 (Quality & Table) and native 24p, 3rd party battery & of course version change.

Funny,

Doing this and using the b settings, and maxing out on the audio patch settings i get only 17261kbits/s.
Am I doing anything wrong?

Adventsam
06-10-2010, 03:04 AM
[quote=jobless;2013935]Look your self...
C Settings:
]

are minimal, needs to be scaled closer? d, does look better though, although I thought you were saying use b?

jobless
06-10-2010, 03:09 AM
I didn't test B, Tester did. I believe that he checked at low level...

jobless
06-10-2010, 03:25 AM
However i Did retest D setting in hidetail scene with lot of zooms and pans and with shutter sped at 500 and strange thing is that I get spikes at low level...
My clip is 43.7 Mbps and Oedipax clip is 45.5 Mbps. So I suppose that camera limit for AVCHD is near by 45 Mbps...
And I didn't have any problems with recording.Maybe my scene is not so detailed.

download link:
http://www.nikolicnemanja.com/Dsettings-Hidetail .MTS


I will try to decrease velues for next test. ( but I can't find a scene with more details so Oedipax could repeat his test with slightly decreased values)

Adventsam
06-10-2010, 03:33 AM
However i Did retest D setting in hidetail scene with lot of zooms and pans and with shutter sped at 500 and strange thing is that I get spikes at low level...
My clip is 43.7 Mbps and Oedipax clip is 45.5 Mbps. So I suppose that camera limit for AVCHD is near by 45 Mbps...
And I didn't have any problems with recording.Maybe my scene is not so detailed.

download link:
http://www.nikolicnemanja.com/Dsettings-Hidetail.MTS
Wait a minut... It's still uploading...

I will try to decrease velues for next test. ( but I can't find a scene with more details so Oedipax could repeat his test with slightly decreased values)

I never go over 1/100, 1/125, why do we need to go so high? wont this cause issues? someone said your footage had high rolling shutter, is this why?

jobless
06-10-2010, 03:40 AM
I never go over 1/100, 1/125, why do we need to go so high? wont this cause issues? someone said your footage had high rolling shutter, is this why?

It's just for test purpose.... Higher shutter speed = more details...
As for rolling shutter, it has nothing with Avchd codec...

jobless
06-10-2010, 08:48 AM
New test:

Video buffer 84000000
Video bitrate adjustment 70000000
Overall bitrate adjustment 72000000
Limiting bitrate adjustment 80000000

1080 - Overall bit rate : 11.0 Mbps
Maximum Overall bit rate : 73.3 Mbps

but interesting 720p - Overall bit rate : 43.0 Mbps
Maximum Overall bit rate : 73.3 Mbps
however camera stops recording after 5 sec when I do fast pans...

jobless
06-10-2010, 09:35 AM
test 2 and 3:
Video buffer 84000000
Video bitrate adjustment 2)55000000 3)60000000
Overall bitrate adjustment 2)57000000 3)62000000
Limiting bitrate adjustment 2)63000000 3)70000000

2) 720 36,5 Mbps
1080 36,8 Mbps
No problems with recording

3) 720 44,1 Mbps (camara stops recording on fast pans after 3-4 sec)
1080 5478 Kbps

Isaac_Brody
06-10-2010, 09:57 AM
Footage?

Paul Shields
06-10-2010, 10:32 AM
Did a quick test with the following settings:


video buffer 40,000,000
video bitrate adj 50,000,000
overall bitrate adj 52,000,000
limiting bitrate adj 60,000,000


All worked very well! Plays back nicely in camera and there is a definite improvement in mud reduction.

One question I have for any Mac user though - how do I extract the information showing bitrate etc? Jobless seems to have the elecard suite on windows for his analysis, but so far haven't come across anything on OS X that does the same thing with AVCHD files...

Park Edwards
06-10-2010, 11:32 AM
test 2 and 3:
Video buffer 84000000
Video bitrate adjustment 2)55000000 3)60000000
Overall bitrate adjustment 2)57000000 3)62000000
Limiting bitrate adjustment 2)63000000 3)70000000

2) 720 36,5 Mbps
1080 36,8 Mbps
No problems with recording

3) 720 44,1 Mbps (camara stops recording on fast pans after 3-4 sec)
1080 5478 Kbps


Are you adjusting ABCDEF? Where are you putting these numbers in?

GMC
06-10-2010, 12:18 PM
Yeah, funny. I was actually referring to early Car3o footage when he was for the first time showing the native 24p patch. He had cars going back and forth at night. This footage dropped to a 23.976 1080p Edius Neo 2 timeline was "real" 24p. The latest footage of jobless (first post on first page of this thread) clearly is not correct 24p, or at least Edius doesn't interpret it correctly (whatever it is).

Just relying on players (and scrubbing back and forth) to judge if it is 24p or not doesn't work (you run into all these issues with "tolerant" players and less tolerant players, plus the fact that computer screens typically run on 60Hz, thereby possibly hiding some of the issues). You have to go to a timeline in an editing software and check frame by frame in a 23.976p environment. Then you see it.

Anyway, thanks to all the efforts. We are damn close!!


UPDATE: I used the vimeo example on page 3 just now, "C" settings and Field of Flowers from IanB (downloaded the MTS file) and also this was supposed to be native 24p. It is not. Out of curiosity I applied the pulldown I usually have to do with original GH1 footage and now I get "real" 24p. Hence, I think that the native 24p patch doesn't work correctly or somehow is obsolete if high AVCHD bitrates are chosen from ptool.

Did some further research. When the MTS (24p native) is directly dropped into an Edius 23.976 1080p timeline, it looks like the original 24p -in-a 60i wrapper. When I convert it first to Canopus HQ and then drop it into the same timeline it is "real" 24p. Hence there must be some room for misinterpretation in the file header (Edius gets it wrong, the Canopus HQ conversion utility gets it right). I thought that I should mention that.

Psynema
06-10-2010, 02:11 PM
Did a quick test with the following settings:


video buffer 40,000,000
video bitrate adj 50,000,000
overall bitrate adj 52,000,000
limiting bitrate adj 60,000,000


All worked very well! Plays back nicely in camera and there is a definite improvement in mud reduction.

One question I have for any Mac user though - how do I extract the information showing bitrate etc? Jobless seems to have the elecard suite on windows for his analysis, but so far haven't come across anything on OS X that does the same thing with AVCHD files...

Wait

so you CAN PLAY THIS BACK IN CAMERA NOW???? Thought it froze the whole thing.

So this is all ready to go then?

Adventsam
06-10-2010, 02:15 PM
not with native 24/25p patch ticked? so if your happy with pal 50i wrapper or 60i wrapper and can pull-down, you are good to go, we are just looking for optimum settings and maximum-minimum iq that removes mud best.

svecher
06-10-2010, 02:18 PM
Wait

so you CAN PLAY THIS BACK IN CAMERA NOW???? Thought it froze the whole thing.

So this is all ready to go then?
I don't think there were ever problems playing back footage with modified bitrates. The only problem that was (and still is) playback of 1080 native progressive footage.

AndersM
06-10-2010, 02:29 PM
Wait

so you CAN PLAY THIS BACK IN CAMERA NOW???? Thought it froze the whole thing.

So this is all ready to go then?

I'm wondering about this too.
I used the B settings, and I get "This motion picture cannot be played".

Should I be able to play this back in camera? I have a PAL camera if it matters(no native 25p patch).

EDIT: I also added the battery, 30 min removal and audio encoding bitrate patch. Maybe it's the bitrate patch that causes the problems?

Thanks!

Anders

Ken Steadman
06-10-2010, 02:55 PM
I'm wondering about this too.
I used the B settings, and I get "This motion picture cannot be played".

Should I be able to play this back in camera? I have a PAL camera if it matters(no native 25p patch).

EDIT: I also added the battery, 30 min removal and audio encoding bitrate patch. Maybe it's the bitrate patch that causes the problems?

Thanks!

Anders

it's because you changed the audio bit rate

jobless
06-10-2010, 03:08 PM
to headintheclouds46:

In 1080 your bitrate with this settings is only 11 Mbps...

Did you try to shoot 720? And what bitrate you getting?

AndersM
06-10-2010, 03:14 PM
it's because you changed the audio bit rate

Thank you, my good friend ;)

I did a rebuild without the audio patch, and everything works fine now!

Has anyone done a comparison with and without the patch on 720 avchd footage? Any improvements in this mode?

headintheclouds46
06-10-2010, 03:15 PM
test 2 and 3:
Video buffer 84000000
Video bitrate adjustment 2)55000000 3)60000000
Overall bitrate adjustment 2)57000000 3)62000000
Limiting bitrate adjustment 2)63000000 3)70000000

2) 720 36,5 Mbps
1080 36,8 Mbps
No problems with recording

3) 720 44,1 Mbps (camara stops recording on fast pans after 3-4 sec)
1080 5478 Kbps

Setting #3 works great for me. I'm using a SanDisk Extreme class 10 card.
Just donated. This is great work by tester!

headintheclouds46
06-10-2010, 03:41 PM
AVCHD 720P 50fps bitrate is 44.7Mbps . M-JEPG 720P 30fps is 59.8Mbps. 1 minute recording with no problems.

bumkicho
06-10-2010, 09:37 PM
Testing out higher bitrate AVCHD although the setting I used is not as high as what some people try. Shot with 28-70mm fd with 2x extender, 80-200mm lens with 2x extender, 20mm pancake lens, all on a monopod. Color graded in AE CS4. All I can say is this higher bitrate codec shines when color grading.

12477730

Kholi
06-10-2010, 09:49 PM
Testing out higher bitrate AVCHD although the setting I used is not as high as what some people try. Shot with 28-70mm fd with 2x extender, 80-200mm lens with 2x extender, 20mm pancake lens, all on a monopod. Color graded in AE CS4. All I can say is this higher bitrate codec shines when color grading.

12477730


Very awesome, man. Looks very good!

Barry_Green
06-10-2010, 10:11 PM
Posted this in the other thread, but ... wait until you see 1080/24pN footage patched with the "D" settings, playing back on a 67" screen. It's fantastic. If you have a blu-ray player with an SD slot on it, or a PS3 or something, you should be able to take the card straight out of the camera and play it on the big screen.

x_WOrPIG_x
06-10-2010, 10:34 PM
I just spent about an hour testing different footage and testing the " #3 " setting and a few others had this problem. Everything I recorded did not stop or freeze ( AVCHD 1080 stuff ) but when I went to go upload it ( via usb cable to computer ) It froze on the PLEASE WAIT . . . screen and i had to remove the battery. ( I tried 2 different cards and they both did the same ) Anyone know?

sammysammy
06-10-2010, 10:38 PM
x worpig take a still pic ,just make sure the last thing on the card is not a patched 24p file, and the usb connection will wrk..

Barry do you mind sharing the D settings .i also tried to find it on here with out any luck,thanks!

Barry_Green
06-10-2010, 10:41 PM
Barry do you mind sharing the D settings .i also tried to find it on here with out any luck,thanks!
They're in this same thread, post 10.
http://www.dvxuser.com/V6/showpost.php?p=2012699&postcount=10

x_WOrPIG_x
06-10-2010, 11:08 PM
x worpig take a still pic ,just make sure the last thing on the card is not a patched 24p file, and the usb connection will wrk..

Barry do you mind sharing the D settings .i also tried to find it on here with out any luck,thanks!

Thanks.

When I uploaded my clips something weird happend. I only got 0.5mb/s for each clip. Just a bunch of choppin blocks. I was using:

Video buffer 84000000
Video bitrate adjustment 3)60000000
Overall bitrate adjustment 3)62000000
Limiting bitrate adjustment 3)70000000

sammysammy
06-10-2010, 11:15 PM
Thanks Barry!

sammysammy
06-10-2010, 11:17 PM
x worpig , these settings are higher than the D settings, im not sure if it has to do with your card speed with those settings..

Mike@AF
06-10-2010, 11:34 PM
I'm tempted to give this all a shot, but I do have some questions about all this and not sure if this is the right place to ask, but I'll do it anyways.

I always prefer to shoot 1080p. Which is going to be better overall quality for 1080p @ 23.98fps with these hacked firmwares, AVCHD or MJPEG?

With AVCHD, aside from not having to do pulldown in post is there any quality advantage to shooting native 24p in camera? This is a concern because I'd want to be able to play back shots in camera.

Also, do these improved bitrates help with the banding in GH1 footage?

Thanks!

butler360
06-10-2010, 11:38 PM
I always prefer to shoot 1080p. Which is going to be better overall quality for 1080p @ 23.98fps with these hacked firmwares, AVCHD or MJPEG?

With AVCHD, aside from not having to do pulldown in post is there any quality advantage to shooting native 24p in camera? This is a concern because I'd want to be able to play back shots in camera.

The 1080P MJPEG is just upscaled, and Barry's test shows AVCHD 1080P has more detail.

I asked the same in the main thread, tester13 said 60i is better for quality. Plus it can be played back in camera.

Barry_Green
06-10-2010, 11:44 PM
I always prefer to shoot 1080p. Which is going to be better overall quality for 1080p @ 23.98fps with these hacked firmwares, AVCHD or MJPEG?
That one's easy -- MJPG is still 30.000 fps, AVCHD is 23.976. So if you want 24p, you want AVCHD.


With AVCHD, aside from not having to do pulldown in post is there any quality advantage to shooting native 24p in camera?
Perhaps. I have to test this further, but it's possible that the native 24p is using progressive 4:2:0 chroma, vs. the 60i-pulldown version using interlaced 4:2:0. Progressive 4:2:0 is just quite a bit better. But I haven't tested properly for that yet.


Also, do these improved bitrates help with the banding in GH1 footage?
Depends on what you mean by banding. Are you talking about fixed-pattern noise, or are you talking about gradient banding in the skies or on flat walls or such? Because it should have a huge effect on improving the gradient banding (that happens in an 8-bit codec when the compression is taxed too much) but I can't see how it would have any effect on fixed-pattern noise.

Barry_Green
06-10-2010, 11:48 PM
The 1080P MJPEG is just upscaled, and Barry's test shows AVCHD 1080P has more detail.
My testing is preliminary. What I found was that the 1080p MJPG was wrong somehow, and it is my guess that it's an upscaling artifact. I don't want that proclaimed as absolute fact until we know for sure. All that I'm prepared to say for sure is -- look at the picture of the concentric circles, there's obviously something going on with the MJPG that shouldn't be there, and I'm guessing that it's from uprezzing.

It is my guess that the 1080P MJPG is really just 720p MJPG being blown up into a 1080p frame. And if that's the case, you'd probably be better off just shooting 720p MJPG, as Pappas has been doing. You can certainly do a better upscale in post than the in-camera upscaler, I'd bet.


I asked the same in the main thread, tester13 said 60i is better for quality. Plus it can be played back in camera.
Not sure that the statement could be directly interpreted to say that 60i is "better for quality." I believe what tester13 said (although I could be wrong) is that the 60i version was showing higher bitrates. But that doesn't necessarily mean it's better quality! It might be requiring higher bitrates because 60i is more complex to compress than 24p is. My guess is that on a strictly quality basis, the 24pN would look better than the embedded-in-60i. But even if that's true, you'd still have to ask "how much better, and is that improvement worth the inconvenience of not being able to play back in-camera".

Mike@AF
06-10-2010, 11:49 PM
That one's easy -- MJPG is still 30.000 fps, AVCHD is 23.976. So if you want 24p, you want AVCHD.

Okay I thought I saw an option in Ptool to make MJPEG 24p. I must be mistaken.


Perhaps. I have to test this further, but it's possible that the native 24p is using progressive 4:2:0 chroma, vs. the 60i-pulldown version using interlaced 4:2:0. Progressive 4:2:0 is just quite a bit better. But I haven't tested properly for that yet.

That would be good to know and I'll wait for results on that. Color is very important to me, although I've been happy with color from the camera thusfar without the hack.


Depends on what you mean by banding. Are you talking about fixed-pattern noise, or are you talking about gradient banding in the skies or on flat walls or such? Because it should have a huge effect on improving the gradient banding (that happens in an 8-bit codec when the compression is taxed too much) but I can't see how it would have any effect on fixed-pattern noise.

I meant the gradient banding on sky shots and flat wall shots. FPN hasn't bothered me that much. I usually put those things out of focus (slightly or very much) whenever possible to reduce the noise.

Thanks for your answers!

Barry_Green
06-10-2010, 11:53 PM
Okay I thought I saw an option in Ptool to make MJPEG 24p. I must be mistaken.
You're not mistaken, the option is there. It just doesn't work right (at least not yet). So far it appears that the camera is still running at 30fps, but the encoder only records 24, so it effectively is just dropping one frame out of every six. So not "real" 24p at all.

butler360
06-10-2010, 11:57 PM
My testing is preliminary. What I found was that the 1080p MJPG was wrong somehow, and it is my guess that it's an upscaling artifact. I don't want that proclaimed as absolute fact until we know for sure.

Ah, well I thought it had basically been determined that it was upscaled 720P. Guess that's not for sure yet.


Not sure that the statement could be directly interpreted to say that 60i is "better for quality."

Well I assumed since tester had it added to the FAQ that he was saying it was the best mode to use, bitrate-wise. His answer was a bit brief, though, so I guess we can't say that for sure. Although the FAQ does say:

"And it has also been suggested to record in 60i as the files will have a better bitrate"

Of course, that says bitrate and not quality. So I should be more precise with my words.

Barry_Green
06-10-2010, 11:59 PM
Ah, well I thought it had basically been determined that it was upscaled 720P. Guess that's not for sure yet.
I just don't want anyone taking a first impression and running with it as if it's a known fact. We need more testing.




Well I assumed since tester had it added to the FAQ that he was saying it was the best mode to use, quality-wise. His answer was a bit brief, though, so I guess we can't say that for sure. Although the FAQ does say:

"And it has also been suggested to record in 60i as the files will have a better bitrate"
Yes, but "better bitrate" doesn't necessarily mean "better quality". That's the distinction I was trying to draw. The question over which is "better quality" is, in my mind at least, unanswered as of yet. Some direct side-by-sides of repeatable circumstances would need to be done. Or, ideally, we'd need two cameras that could be set up side-by-side; that'd be the right way to do it.

butler360
06-11-2010, 12:07 AM
I get what you're saying. Thanks for clarifying.

Ozpeter
06-11-2010, 02:04 AM
Okay, one last tidbit for you folks -- the 24pN footage plays on a blu-ray player! It doesn't play back in camera, but it sure did when I put it in my blu-ray and played it on my 67" DLP TV. It will probably work from a PS3 as well.

The PS3 certainly plays unmodified firmware AVCHD footage (1080 & 720) - even sent from my PC across a wireless network!

I'll try the PS3 with the "D" settings in the morning (if it works on my existing card) as well as trying the 'turntable' test.

Update (fwiw) - I've played the original 'Jobless' example and the ' garden-whipping' example on my PS3 with the files stored on my PC, across the network - the garden-whipping played a bit jerkily but that could be a network issue rather than a PS3 issue. Anyway, both looked great on the 42" Bravia connected to the PS3.

rambooc1
06-11-2010, 02:41 AM
Update (fwiw) - I've played the original 'Jobless' example and the ' garden-whipping' example on my PS3 with the files stored on my PC, across the network - the garden-whipping played a bit jerkily but that could be a network issue rather than a PS3 issue. Anyway, both looked great on the 42" Bravia connected to the PS3.

Ditto when played on a WD media Player via HDMI.

bumkicho
06-11-2010, 03:20 AM
Very awesome, man. Looks very good!
Thanks, Kholi. I hope you will get your hands on a GH1 soon!!

Adventsam
06-11-2010, 03:23 AM
Thanks, Kholi. I hope you will get your hands on a GH1 soon!!

Wonder if they'll withdraw the price incentives:beer:

jobless
06-11-2010, 03:39 AM
Testing Ptool 3,36:

I can confirm playback of native 24p in camera.

I didn't check buffer for this test and I use C suggested setting from ptool...

1080 -- 41 Mbps ( camera on heavy pans recording without of problems)

http://www.nikolicnemanja.com/pics/1080p.jpg


720 --- 36 Mbps ( camera records normally on heavy pans but on really heavy pans ( like samurai with sword) stops recoding...




http://www.nikolicnemanja.com/pics/720p.jpg


I suppose that Video buffer option i Ptool don't work...

ProjX v2.0
06-11-2010, 03:43 AM
Jobless, how are you reading Streameye to find the bitrate? When you mouseover the blue bars, is it the "size" that is the bitrate?

Vitaliy Kiselev
06-11-2010, 03:47 AM
to jobless

1) Adjust Limiting bitrate to lower value to remove stops in recording.
2) Try and do systematic comparison between Video bitrate patch and Video bitrate simplified patch (check one or the other, both are useless).
3) I also think that buffer can be removed in later versions. But more testing is required.

jobless
06-11-2010, 03:47 AM
No, I read bitrate with mediainfo.. In streameye size stands for size of that specific frame...

Vitaliy Kiselev
06-11-2010, 03:47 AM
Jobless, how are you reading Streameye to find the bitrate? When you mouseover the blue bars, is it the "size" that is the bitrate?

Streameye have separate window for bitrate information (Alt-S).
In popup it only show info about frame.
MediaInfo can be not very accurate as it uses default fields, and not exact calculations.

jobless
06-11-2010, 03:57 AM
Streameye have separate window for bitrate information (Alt-S).
In popup it only show info about frame.
MediaInfo can be not very accurate as it uses default fields, and not exact calculations.

ok, no more MediAInfo ...

ProjX v2.0
06-11-2010, 04:00 AM
Ah , thanks guys. ALT-S was the trick.
Is it the "bitrate declared" number that is the actual bitrate or is it the "bit allocation min/avg/max" numbers?

jobless
06-11-2010, 04:09 AM
New test:

Same settings, just changed Limiting bitrate adj. to 5500000

1080 -.- 40 361 800 kbps ( no problem with Cam)

720 ... 38 941 620 kbps ( with samurai style paning camera stops rec.)

Oedipax
06-11-2010, 04:19 AM
Native 24p in-cam works for me with the "D" settings!

However, when I patched MJPEG to 1920 and used the following settings:





MJPEG Size 1280M = Ticked
MJPEG Enco 1280M = Ticked
MJPEG E1 Quality = 400
MJPEG E1 Table = 103
MJPEG E2 Quality = 350
MJPEG E2 Table = 116
MJPEG E3 Quality = 250
MJPEG E3 Table = 120
MJPEG E4 Quality = 200
MJPEG E4 Table = 125

I got write speed errors to the card after just a second or two of recording. Didn't seem scene dependent, either (I never successfully recorded a clip with MJPEG with those settings). Transcend 16gb card, class 6. Maybe I need a class 10, or maybe it's just too much for the camera's hardware.

Ozpeter
06-11-2010, 04:22 AM
Downloaded the "garden whipping" video right into the PS3 and it plays perfectly (c.43Mbps says the PS3 display). The big screen confirms that there's clearly still 'mud' in the fine foliage (eg base of big tree and the bush that appears above the top of the birdbath) but you have to step through frames to see it - in normal play nobody would be troubled by it. Actually perhaps the really fast pan isn't such a good test - it helps obscure possible problems as they flash past too quickly. A pan at the speed of a typical real pan might be better, one where you can still judge the quality of what you are seeing in normal playback.

jobless
06-11-2010, 04:22 AM
Limiting bitrate adj. to 5000000

1080 --- Streameye want open the file, Mediainfo shows 37.4 Mbps.. I didn't have any problems during recording and player plays hte file normally...

720 --- 40 265 416 kbps ( camera stops recording after 5sec)





Limiting bitrate adj. to 4500000

Camera freezes when I hit record button ( However, LCd works normal but all other functions are not available... I needed to remove the battery....
It happens in 1080 and 720...

Vitaliy Kiselev
06-11-2010, 04:23 AM
Fun thing is that as we now know weakest point of this camera is SD controller :-)
In case of CompactFlash, controller is inside card, and it uses UDMA. So, you can have really good speed, like 240Mbits/s. May be it's worth experimenting with 7D :-)

Vitaliy Kiselev
06-11-2010, 04:25 AM
Limiting bitrate adj. to 4500000

Camera freezes when I hit record button

Never ever make limiting bitrate lower than overall bitrate and, of course, video bitrate. It'll lead to freezing due to firmware bug (function do not process error code correctly). So, next time I hope they'll read something about unit testing.

ProjX v2.0
06-11-2010, 04:32 AM
Native 24p in-cam works for me with the "D" settings!

However, when I patched MJPEG to 1920 and used the following settings:
[/list]
I got write speed errors to the card after just a second or two of recording. Didn't seem scene dependent, either (I never successfully recorded a clip with MJPEG with those settings). Transcend 16gb card, class 6. Maybe I need a class 10, or maybe it's just too much for the camera's hardware.

I'm using those MJPEG settings and haven't had any issue with recording up to 3 min on an Acumem 16GB class 6 card.

Oedipax
06-11-2010, 04:33 AM
I'm using those MJPEG settings and haven't had any issue with recording up to 3 min on an Acumem 16GB class 6 card.

Hmm, okay, thanks. It seems I am in the market for a faster SD card then!

Adventsam
06-11-2010, 04:34 AM
People should go with the Sandisk extreme 200x cards for now, these are highly rated by toms hardware.

All others have a habit of dropping the write speed=errors. I think the issue is the card, the sd controller is probably getting a write error back? If everybody moves to the sandisk 200x, sustained writing of 22.6MB is probable, ie 180mb/sec.

http://www.tomshardware.co.uk/compactflash-sdhc-class-10,review-31878-8.html

ProjX v2.0
06-11-2010, 04:35 AM
I think if you switch from Transcend to anything else, you should be fine. They are the worst I've ever used.

Adventsam
06-11-2010, 04:43 AM
I think if you switch from Transcend to anything else, you should be fine. They are the worst I've ever used.

This is the best sdhc out there at the moment.

jobless
06-11-2010, 04:47 AM
Video bitrate adjustment 43000000
Overall bitrate adjustment 45000000
Limiting bitrate adjustment 46000000

1080 --- 39 440 032 kbps ( samurai pans) No problems with CAm

720 ----- 38 392 884 ( samurai pans ) camera stops after 7 sec..

Adventsam
06-11-2010, 04:50 AM
Video bitrate adjustment 43000000
Overall bitrate adjustment 45000000
Limiting bitrate adjustment 46000000

1080 --- 39 440 032 kbps ( samurai pans) No problems with CAm

720 ----- 38 392 884 ( samurai pans ) camera stops after 7 sec..

I cant do it until later tonight! anyone? jobless, what card you using?

ProjX v2.0
06-11-2010, 04:51 AM
Video bitrate adjustment 43000000
Overall bitrate adjustment 45000000
Limiting bitrate adjustment 46000000

1080 --- 39 440 032 kbps ( samurai pans) No problems with CAm

720 ----- 38 392 884 ( samurai pans ) camera stops after 7 sec..

Is this with Video Buffer patch on? If so, set to 40 or 84?

jobless
06-11-2010, 04:53 AM
I cant do it until later tonight! anyone? jobless, what card you using?

It is not Card problem...
It's SD controller...

jobless
06-11-2010, 04:54 AM
Is this with Video Buffer patch on? If so, set to 40 or 84?

Buffer seems to not have any influence...

jobless
06-11-2010, 04:56 AM
Video bitrate adjustment 40000000
Overall bitrate adjustment 42000000
Limiting bitrate adjustment 44000000

1080 --- 37 431 240 kbps ( without problems)

720 .... 36 270 044 kbps ( stops after 13 sec)

Adventsam
06-11-2010, 04:58 AM
It is not Card problem...
It's SD controller...


Class10, its the only one able for sustained high bit rates. I'm ordering 1 today, I want some of that mjpeg 80mps+ too.:happy:

Vitaliy Kiselev
06-11-2010, 05:00 AM
With time we'll understand coder behaviour better.
As for pans - you can have not card problem but hardware problem as time distance between frames is much smaller and if encoder can't complete frame before next one it can freeze. Seems reasonable.
Try with 720p fps patch (bitrate drops significantly of course, but you can try to adjust accordingly).

Vitaliy Kiselev
06-11-2010, 05:01 AM
Class10, its the only one able for sustained high bit rates. I'm ordering 1 today, I want some of that mjpeg 80mps+ too.:happy:

Please try to find Barry's and mine explanations about Class ratings.
To be short - you are wrong about Class 10.

jobless
06-11-2010, 05:12 AM
Without buffer option checked
Video bitrate adjustment 38000000
Overall bitrate adjustment 40000000
Limiting bitrate adjustment 42000000

1080 ---- 34 364 832 kbps ( no problems ) On heavy pans I don't see any MUd...

720 ---- 35 512 400 kbps I recorded 1 min 15 sec clip and then I stopped the camera. So no problems with cam.... Also I can't see any MUd...

I'll upload clips letter....

ProjX v2.0
06-11-2010, 05:13 AM
I didn't check buffer for this test and I use C suggested setting from ptool...

1080 -- 41 Mbps ( camera on heavy pans recording without of problems)

This stops recording for me if I swish the camera around too much.

Adventsam
06-11-2010, 05:50 AM
Please try to find Barry's and mine explanations about Class ratings.
To be short - you are wrong about Class 10.

Sandisk c10 200x is as good as it gets by the looks of it, but I'll check.

jobless
06-11-2010, 07:04 AM
Download links for clips:

http://www.nikolicnemanja.com/test6/1080samurai.MTS
http://www.nikolicnemanja.com/test6/720samurai.MTS

settings:

Without buffer option checked
Video bitrate adjustment 38000000
Overall bitrate adjustment 40000000
Limiting bitrate adjustment 42000000

1080 ---- 34 364 832 kbps ( no problems )

720 ---- 35 512 400 kbps I recorded 1 min 15 sec clip and then I stopped the camera.

ProjX v2.0
06-11-2010, 07:41 AM
Jobless, do you record all your clips for 75 seconds? The reason I ask is because in StreamEye, I'm noticing bitrate differences depending on clip length.

jobless
06-11-2010, 08:03 AM
Jobless, do you record all your clips for 75 seconds? The reason I ask is because in StreamEye, I'm noticing bitrate differences depending on clip length.

Avchd Is variable codec. So, if you have shoot with different levels of detail bitrate is changing accordingly... You should looking for average Avg bitrate...

jobless
06-11-2010, 08:10 AM
Also in shoots with lot of movements, when you stop panning bitrate should drop down...

Adventsam
06-11-2010, 08:54 AM
Download links for clips:

http://www.nikolicnemanja.com/test6/1080samurai.MTS
http://www.nikolicnemanja.com/test6/720samurai.MTS

settings:

Without buffer option checked
Video bitrate adjustment 38000000
Overall bitrate adjustment 40000000
Limiting bitrate adjustment 42000000

1080 ---- 34 364 832 kbps ( no problems )

720 ---- 35 512 400 kbps I recorded 1 min 15 sec clip and then I stopped the camera.

:dankk2:

jobless
06-11-2010, 09:26 AM
Video bitrate adjustment simplifed TEST:

Without buffer option checked
Video bitrate adjustment simplifed 38000000
Overall bitrate adjustment 40000000
Limiting bitrate adjustment 42000000

1080 --- 36 653 036 kbps ( no problems)

720 --- 33 927 580 kbps (no problems)

Vitaliy Kiselev
06-11-2010, 09:30 AM
jobless

Please, shoot identical footage with identical settings using still scenes and motion (use tripod) using original video bitrate and simplified video bitrate patches. After this look at clips at low level.
Start with simplified, as this is current firmware.

jobless
06-11-2010, 09:33 AM
jobless

Please, shoot identical footage with identical settings using still scenes and motion (use tripod) using original video bitrate and simplified video bitrate patches. After this look at clips at low level.
Start with simplified, as this is current firmware.

ok, no problem... I'll shoot tomorrow,'couse I'm going out now...

jcintron
06-11-2010, 11:31 AM
Love the Cat digging thru the flower pots on the 720 samurai shot.... :-)
Quick question to Tester or Jobless.

Are these all the tweaks with these setting? or are these being left out (default)?:

MJPEG Size 1280M = Ticked
MJPEG Enco 1280M = Ticked
MJPEG E1 Quality = 400
MJPEG E1 Table = 103
MJPEG E2 Quality = 350
MJPEG E2 Table = 116
MJPEG E3 Quality = 250
MJPEG E3 Table = 120
MJPEG E4 Quality = 200
MJPEG E4 Table = 125

alignment1
06-11-2010, 11:33 AM
just updated w/ jobless 'd' settings........oh my....

have a shoot tonight---will use AVCHD1080p24 jobless 'd'!!!!

If needed how hard is it to revert to pre-hacked firmware??

seanmcleod
06-11-2010, 11:39 AM
just updated w/ jobless 'd' settings........oh my....

have a shoot tonight---will use AVCHD1080p24 jobless 'd'!!!!

If needed how hard is it to revert to pre-hacked firmware??

Answered in the FAQ sticky

Barry_Green
06-11-2010, 11:48 AM
Tested AVCHD 60i vs. 24pN, and yes, no question, the chroma is recorded much better in the 24pN mode.
http://dvxuser.com/barry/GH1-Chroma-Compare.jpg

Also tested MJPG vs. AVCHD/pN, using Pappas settings. Resolution looks actually pretty comparable, but there's a disturbing chroma alias that looks very very much like what the Canons do. Question here though is: why did it do that on the Pappas settings, but not on the other settings? Weird.
http://dvxuser.com/barry/GH1-MJPG-vs-AVCHD-chroma-alias.jpg

Barry_Green
06-11-2010, 11:59 AM
Also, the chroma is just crazy higher on the newest firmware build. The 1080pN shots were from yesterday, the 1080i and MJPG shots were from today. Nothing changed on the camera at all, the settings are exactly the same, but the colors are far more saturated on today's shots.

Svart
06-11-2010, 12:17 PM
Not sure I'm following, do you mean that Ptool's latest update causes chroma differences between yesterday's update and today's?

bumkicho
06-11-2010, 12:29 PM
I just tested out D setting with native 24fps. Oh man.. IT IS BEAUTIFUL..

Yes, samuri pan in SH mode will cause the camera to stop recording, but when will I ever need to samuri pan? NEVER. It handles just fine when panning faster than normal.

Barry,
Thanks for 60i vs 24fps test shot and avchd vs mjpeg shot. To me this higher bitrate AVCHD at native 24fps with no recording time limit is the winner over 1080p MJPEG with 3 minute limit.

Paul Shields
06-11-2010, 12:42 PM
Personally I'm struggling to think of a reason to shoot high bitrate MJPEG as opposed to high bitrate AVCHD (including native 24/25p + playback in camera). Would be good if someone can do a test of both at their optimum settings. As bumkicho mentions though, the file benefits of AVCHD make it a no brainer for me.

Kholi
06-11-2010, 12:44 PM
Personally I'm struggling to think of a reason to shoot high bitrate MJPEG as opposed to high bitrate AVCHD (including native 24/25p + playback in camera). Would be good if someone can do a test of both at their optimum settings. As bumkicho mentions though, the file benefits of AVCHD make it a no brainer for me.

I think just aesthetically, it offers something different. It almost looks like HPX3000 footage or Adapter footage versus Straight DSLR video footage if that makes sense.

Paul Shields
06-11-2010, 12:49 PM
I think just aesthetically, it offers something different. It almost looks like HPX3000 footage or Adapter footage versus Straight DSLR video footage if that makes sense.

Yes, it does make sense. I guess you're saying it looks a bit more "film-like"? Could that be approximated later when editing?

Kholi
06-11-2010, 12:51 PM
Yes, it does make sense. I guess you're saying it looks a bit more "film-like"? Could that be approximated later when editing?

Nah, not saying more film-like. Haha. I've long since realized that film is film, digital is digital. Within both realms there are different looks to choose from.

I'm just saying that the AVCHD looks different, to my eye, than the Photo JPEG. I like the look of both, having both wouldn't be harmless.

Dunno what you would do to get the Photo JPEG look out of the AVCHD.

Trucci
06-11-2010, 12:51 PM
Download links for clips:

http://www.nikolicnemanja.com/test6/1080samurai.MTS
http://www.nikolicnemanja.com/test6/720samurai.MTS

settings:

Without buffer option checked
Video bitrate adjustment 38000000
Overall bitrate adjustment 40000000
Limiting bitrate adjustment 42000000

1080 ---- 34 364 832 kbps ( no problems )

720 ---- 35 512 400 kbps I recorded 1 min 15 sec clip and then I stopped the camera.

I tried this setting and I'm getting weird splotchy artifacts at 720p as if the bitrate is too low. I just did a test shot of my room right after installing the patched firmware, so my iso was set to 1600. 1080p mode doesn't seem to have these artifacts. Maybe I set one of the parameters wrong?

Here's what I'm talking about: http://dl.dropbox.com/u/1017296/splotch.MTS

Paul Shields
06-11-2010, 12:53 PM
Nah, not saying more film-like. Haha. I've long since realized that film is film, digital is digital. Within both realms there are different looks to choose from.

I'm just saying that the AVCHD looks different, to my eye, than the Photo JPEG. I like the look of both, having both wouldn't be harmless.


They certainly seem to have obvious differences - though from what I'm seeing the higher bit-rate AVCHD holds it together a bit better? Lack of moire and other artifacts.



Dunno what you would do to get the Photo JPEG look out of the AVCHD.

Add artifacts :).

Bobomatic
06-11-2010, 01:06 PM
Apologies in advance if I've missed this in the thread, but does the latest patching impact the 720/60P mode? I love my slo mo and I'm curious if applying this would impact this mode?

Barry_Green
06-11-2010, 01:07 PM
Not sure I'm following, do you mean that Ptool's latest update causes chroma differences between yesterday's update and today's?
Unless I'm crazy... that's what appears to be happening.

I set the camera on the tripod and shot with yesterday's settings. Then I updated the firmware and shot with today's settings. Those shots were taken about 30 seconds apart from each other, but the chroma is hyper-saturated on today's, versus yesterday's. Same camera settings, same white balance, same everything. I left the camera in the mode I'd established for yesterday's shots, and shot. Then I updated the firmware to today's and shot again. No other differences in operation, but the resulting images are quite different.

tommygdawg
06-11-2010, 01:39 PM
Hey all,


I've been putsing around DvxUser for some time now. However, I just decided to sign up and really get active since I started getting involved with this firmware hack. I just wanted to add my two cents to what I've discovered.



Everything I've shot has been beautiful since the hack
AVCHD seems to be holding up well even upon up-ressing ine AE
Only issue I've seen is that 24pN (native 24) footage, according to After Effects, is 23.971 frames per second.

Isn't 23.971fps going to be an issue when it comes to distributing films? Also if you try to conform it to 23.976 won't there possibly be audio sync problems after a while? Maybe this isn't a huge issue, but it's something that I've been quite curios about and is my only trepidation when it comes to using the native 24p patch.

By the way for anyone who's interested I've been using the C and D settings primarily and I'm really loving my GH1! It's like a new camera.

x_WOrPIG_x
06-11-2010, 01:47 PM
By the way for anyone who's interested I've been using the C and D settings primarily and I'm really loving my GH1! It's like a new camera.

I've been testing with the D setting and what's weird is in 720p I can get around 35mb/s with my shutter speed at about 1/200 if I shot the same scene ( trees and bush ) with a 1/50 i would only get about 10mb/s and the there isn't enough detail.

tommygdawg
06-11-2010, 02:00 PM
I've been testing with the D setting and what's weird is in 720p I can get around 35mb/s with my shutter speed at about 1/200 if I shot the same scene ( trees and bush ) with a 1/50 i would only get about 10mb/s and the there isn't enough detail.

It's because a lower shutter speed gives more motion blur and thus less detail in movements and thus a lower bit rate (I think :P).

Bungess
06-11-2010, 02:51 PM
just donated $10.
Has anyone tried a green screen shoot with these new hacks? Regular GH1 codec did not work for my purposes (much too blocky). If there's there's noticeable improvement in regards to keying this new footage, then I'm pumped (already pumped that there's no more mud).
Scott Jenson

Kholi
06-11-2010, 02:53 PM
just donated $10.
Has anyone tried a green screen shoot with these new hacks? Regular GH1 codec did not work for my purposes (much too blocky). If there's there's noticeable improvement in regards to keying this new footage, then I'm pumped (already pumped that there's no more mud).
Scott Jenson

Count on it. That's the first test we're doing at Rainfall.

If you check the reel ( www.rainfallfilms.com/showreel ) about the third or forth shot in is with a 5D, the guy with the cowboy hat walking through the canyon.

5D wasn't all that great, but it worked for now. I think the GH1 will melt it in comparison once converted to a 4:4:4:4 or uncompressed space.

jobless
06-11-2010, 03:02 PM
Love the Cat digging thru the flower pots on the 720 samurai shot.... :-)
Quick question to Tester or Jobless.

Are these all the tweaks with these setting? or are these being left out (default)?:

MJPEG Size 1280M = Ticked
MJPEG Enco 1280M = Ticked
MJPEG E1 Quality = 400
MJPEG E1 Table = 103
MJPEG E2 Quality = 350
MJPEG E2 Table = 116
MJPEG E3 Quality = 250
MJPEG E3 Table = 120
MJPEG E4 Quality = 200
MJPEG E4 Table = 125

Doesn't matter... These are for MJPEG only. You can chack them or not, they don't influence AVCHD
The cat is callled Nutella... You got really good eyes..:)

Barry_Green
06-11-2010, 03:16 PM
If there's there's noticeable improvement in regards to keying this new footage, then I'm pumped (already pumped that there's no more mud).
Unquestionably there will be (or at least SHOULD be) a substantial improvement in keying quality. The codec is far more robust and accurate now, and if you use the 24pN patch you get the "good stuff" - 24p progressive chroma, instead of the interlaced chroma that the 24p-within-60i version gives.

dmpsk8
06-11-2010, 03:25 PM
big improvements for the GH1 world! Just sold my GH1 for a 7D or I'd be all over this.

jobless
06-11-2010, 03:36 PM
I just tested out D setting with native 24fps. Oh man.. IT IS BEAUTIFUL..

Yes, samuri pan in SH mode will cause the camera to stop recording, but when will I ever need to samuri pan? NEVER. It handles just fine when panning faster than normal.


Yes, It's true. But occasionally you'll have some scenarios that D settings in 720 mode want hold.
For example:

From 30 sec to 1 min


http://www.nikolicnemanja.com/locationtest.mp4

sammysammy
06-11-2010, 03:43 PM
Barry my question to you , once yo compared a cheap lens to an expensive lens,and you said that regarding shooting video it make not much difference ..now with all these extra robust codec,does this change things regarding lenses or do you still think we are still far from seeing the difference from a good lens to cheap lens ..thanks!

jobless
06-11-2010, 03:50 PM
I tried this setting and I'm getting weird splotchy artifacts at 720p as if the bitrate is too low. I just did a test shot of my room right after installing the patched firmware, so my iso was set to 1600. 1080p mode doesn't seem to have these artifacts. Maybe I set one of the parameters wrong?

Here's what I'm talking about: http://dl.dropbox.com/u/1017296/splotch.MTS

Nice finding!
I checked and I have problems on 720 with 1600 and 800 iso with this settings...

Kholi
06-11-2010, 03:56 PM
Barry my question to you , once yo compared a cheap lens to an expensive lens,and you said that regarding shooting video it make not much difference ..now with all these extra robust codec,does this change things regarding lenses or do you still think we are still far from seeing the difference from a good lens to cheap lens ..thanks!


Good question. Barry might be able to test this as well.

nobbystylus
06-11-2010, 04:00 PM
anybody got any reliable settings for AVCHD? I'm not getting reliable results from Jobless's settings, getting glitches and stuttering with those..

i'm also not noticing any improvement in lowlight in terms of mud and codec nastyness in dark areas.

I used joblesses settings with a Sandisk 200x class 10 16gb card.

Psynema
06-11-2010, 04:03 PM
Do you guys notice an improvement WITHOUT pixel peeping footage you would have used anyway? (usually during quick pans, you cut that stuff out of a final edit anyway).

But does anyone who installed the hack find their footage more 5D ish ? LOL.

PappasArts
06-11-2010, 04:13 PM
I think just aesthetically, it offers something different. It almost looks like HPX3000 footage or Adapter footage versus Straight DSLR video footage if that makes sense.

Totally understand. It's the codec. AVCHD codec looks more Video, edgy to me. MJPEG's tonality is smother and feels dense. Like high bit rate codecs are.

Here is a shot done today with a macro on the GH1 in 1080P mode

http://www.dvxuser.com/V6/showpost.php?p=2015709&postcount=4

This is the thread to the new 1080P Mpeg settings.

http://www.dvxuser.com/V6/showthread.php?t=213364


Pappas

tired
06-11-2010, 04:13 PM
Can everyone stop posting for about a week - I've got to get some work done!

I'ts extremely frustrating - I've got 2 GH!'s & so much editing to do I haven't got time to have a play! - it's just EDIT, check the forum - RENDER, check the forum - BURN, check the forum
:badputer:


I think we need to design a TShirt that prople can buy (profits to the fund)

TESTER13 rocks!

jobless
06-11-2010, 04:18 PM
Totally understand. It's the codec. AVCHD codec looks more Video, edgy to me. MJPEG's tonality is smother and feels dense. Like high bit rate codecs are.

Here is a shot done today with a macro on the GH1 in 1080P mode

http://www.dvxuser.com/V6/showpost.php?p=2015709&postcount=4

This is the thread to the new 1080P Mpeg settings.

http://www.dvxuser.com/V6/showthread.php?t=213364


Pappas

Mjpeg looks nice...
But you need to shoot some wide shots with lot of details!!!
It's just upscaled 720.

PappasArts
06-11-2010, 04:24 PM
Mjpeg looks nice...
But you need to shoot some wide shots with lot of details!!!
It's just upcaled 720.

It's not upscaled in camera I believe. I have never had any of the best of 720P look anywhere near that. If that is up-scaling, then Panasonic needs to sell that chip to Hollywood.

More shooting of-course needs to be done.


Pappas

Kholi
06-11-2010, 04:29 PM
Yes, as we've seen from Pappas' earlier testing, his upscaled 720 looked much softer. This 1080, if it's upscaling--I'd have to agree--I'll have some of that.

It reminds me of HPX3000 footage, and the AVCHD reminds me of EX-1 footage.

Pappas used a great term, I remember using it when I first got to touch an HPX3000: Dense. Thick. Heavy. Full.

Good thing it won't hurt to have both if we can get 24P out of it.

jcintron
06-11-2010, 04:34 PM
These are good for 1080 only... is there a good compromise for 720 and 1080? (Other than factory default?)



Video bitrate adjustment simplifed TEST:

Without buffer option checked
Video bitrate adjustment simplifed 38000000
Overall bitrate adjustment 40000000
Limiting bitrate adjustment 42000000

1080 --- 36 653 036 kbps ( no problems)

720 --- 33 927 580 kbps (no problems)

PappasArts
06-11-2010, 04:38 PM
Yes, as we've seen from Pappas' earlier testing, his upscaled 720 looked much softer. This 1080, if it's upscaling--I'd have to agree--I'll have some of that.

It reminds me of HPX3000 footage, and the AVCHD reminds me of EX-1 footage.

Pappas used a great term, I remember using it when I first got to touch an HPX3000: Dense. Thick. Heavy. Full.

Good thing it won't hurt to have both if we can get 24P out of it.

Indeed "Dense Thick. Heavy. Full." That is how it feels. When massaging it in post. It just feels right! :-)


Great analogy EX1 vs HPX3000. Both are good, both produce nice images. Just the HPX3000 is just different... In a good way.


This 1080, if it's upscaling--I'd have to agree--I'll have some of that..

Man, I would attach one of those to my bank account, credit line, chick magnetism, carrer; why hell I would attach that chip to everything. :-D

Pappas

.

Kholi
06-11-2010, 04:54 PM
Hahah! I know what I'm about to attach to it. >:] Master Primes.

Pappas, you got PM. Need to know what SDHC card you have, getting my GH-1 tonight.

ATL Media Group
06-11-2010, 04:54 PM
I think we need to design a TShirt that prople can buy (profits to the fund)

TESTER13 rocks!

heh... something like this:
18233

PappasArts
06-11-2010, 04:56 PM
hahah! I know what i'm about to attach to it. >:] master primes.

lol.........

Barry_Green
06-11-2010, 04:58 PM
Barry my question to you , once yo compared a cheap lens to an expensive lens,and you said that regarding shooting video it make not much difference ..now with all these extra robust codec,does this change things regarding lenses or do you still think we are still far from seeing the difference from a good lens to cheap lens ..thanks!
No no no -- the codec hasn't changed how the sensor works! It still aliases, and still benefits from some softening. I shot most of my stuff at f/8 and f/11 and the diffraction helped to minimize some of the aliasing, but the detail is incredible.

It's vastly improved, but at heart it's still a DSLR.

I think what it's done is set itself definitely above the Canons now. It's sharper, without the chroma moire, and it's got no mud, which was the real knock against it.

And it's still $1200 including lens. Huge improvement, but at heart it's still a DSLR/still camera and so it will still be affected by those things that affect DSLRs.

Kholi
06-11-2010, 05:02 PM
lol.........


Haaaa PM me man! Must get SDHC card that Pappas has. PLEASE PLEASE

PappasArts
06-11-2010, 05:08 PM
Haaaa PM me man! Must get SDHC card that Pappas has. PLEASE PLEASE

Just did Kholi.

As to cards, as I told Kholi in a PM. Go Extreme 8gig. I bought mine at Best buy overpriced. 74buck if I recall. I know there were II III, Those seem to have been changed out. The best cards are important at these data rates.

Also, if we could get a SDdummy card with ribbon cable to Compact flash. If such thing existed, that could interesting.

Out of here to shoot tests----------->


Pappas

Kholi
06-11-2010, 05:09 PM
Yes, that would be ideal. UDMA Cards would probably moot any card issues for sure.

Gonna roll to Samy's Camera and Best Buy in a second here to figure out which ones to go with.

Thanks a lot Pappas.

Photo JPEG 1080/24 WHAT!?

jobless
06-11-2010, 05:10 PM
It's not upscaled in camera I believe. I have never had any of the best of 720P look anywhere near that. If that is up-scaling, then Panasonic needs to sell that chip to Hollywood.

More shooting of-course needs to be done.


Pappas

As I said, You need to shoot some wide shot detailed footage( use tripod) and compare AVCHD 1080 and MJPEG 1080...

As for upscaling; look at NURBS test

Very interestin thing too:

MJPEG mode let do change resolution ;-)

Duration : 10s 80ms
Bit rate mode : Variable
Bit rate : 28.9 Mbps
Width : 2 048 pixels
Height : 854 pixels
Display aspect ratio : 2.35:1
Frame rate mode : Constant
Frame rate : 25.000 fps
Bits/(Pixel*Frame) : 0.661

He got squeezed image. I suppose that is not true 1080, it's 720 upscaled output from sensor.
We need much more testing to achieve true 1080 __Mjpeg, if it's possible?
Resolution test that I made with Andrew settings (MJPEG) are not even close to 1080 AVCHD....
+ 2GB Size Limit
+ FPN in Shadows in MJPEG
+ Card limit with high detailed scenes
+ 60p
+ 24p

PappasArts
06-11-2010, 05:18 PM
As I said, You need to shoot some wide shot detailed footage( use tripod) and compare AVCHD 1080 and MJPEG 1080...

As for upscaling; look at NURBS test

Very interestin thing too:

MJPEG mode let do change resolution ;-)

Duration : 10s 80ms
Bit rate mode : Variable
Bit rate : 28.9 Mbps
Width : 2 048 pixels
Height : 854 pixels
Display aspect ratio : 2.35:1
Frame rate mode : Constant
Frame rate : 25.000 fps
Bits/(Pixel*Frame) : 0.661

He got squeezed image. I suppose that is not true 1080, it's 720 upscaled output from sensor.
We need much more testing to achieve true 1080 __Mjpeg, if it's possible?
Resolution test that I made with Andrew settings (MJPEG) are not even close to 1080 AVCHD....
+ 2GB Size Limit
+ FPN in Shadows in MJPEG
+ Card limit with high detailed scenes
+ 60p
+ 24p



<<<Width : 2 048 pixels
Height : 854 pixels>>>>
Those settings are not good. They failed in my testing.


How's this, it's all gravy.......We should have all options 1080/720 30/24fps and AVCHD/Mjpeg. All these tools have there rightful place. Would you agree?


Pappas

jobless
06-11-2010, 05:21 PM
<<<Width : 2 048 pixels
Height : 854 pixels>>>>
Those settings are not good. They failed in my testing.


How's this, it's all gravy.......We should have all options 1080/720 30/24fps and AVCHD/Mjpeg. All these tools have there rightful place. Would you agree?


Pappas

absolutely....

Kholi
06-11-2010, 05:25 PM
absolutely....

Then let's all vote and donate to get them enabled. I would greatly appreciate a Photo JPEG @ 1080/24 along with my AVC-HD @ 1080/24pN. I feel as though I could use the AVCHD for commercial, videography, etc and the Photo JPEG for all of my narrative work.

This is turning out to be just as exciting as the GH1's launch.

Good times.