That's exactly what I needed to see. Way to come through.
It's official: need to jump on either this or the sigma 20/1.8 Macro. Gonna compare.
Sorry for the thread derail.
AVCHD high bitrate tests- MUD FREE footage
- Thread starter jobless
- Start date
That's exactly what I needed to see. Way to come through.
It's official: need to jump on either this or the sigma 20/1.8 Macro. Gonna compare.
Sorry for the thread derail.
Paul Shields
Member
Yeah, the 24mm f/1.7 is just lovely on the GH-1 - so long as you don't feel inadequate being able to carry such performance in a coat pocket
.
jobless
Active member
Paul Shields
Member
ok, no problem. If anyone does want them then just shout. btw, the ones uploaded are in h.264 at highest settings and not mjpeg
.kenbostero
Member
I've got the 20mm 1. 7 pancake and 20mm 1.8 sigma. Overall I use the pancake lens more (lovely lens) but I like the sigma for its macro. Also on the Gh-1 the sigma becomes a 40mm due to sensor size......I've tried the D settings and tested both lenses, great to see a lack of mud!
____________________________
http://www.vimeo.com/user3615903
____________________________
http://www.vimeo.com/user3615903
"d" settings
native 24p
50mm f1.4
100iso
Outdoor shoot with grass, hedges and trees.
Patriot LX class 10 8Gb:
One file recorded for 2.5 minutes OK, I stopped recording
2nd file received "card speed error, etc" after 45 seconds
3rd file received "card speed error" immediately.
4th file recieved "card speed error" immediately.
AData Class 6 8gb:
Immediate "card speed error" on all tries.
native 24p
50mm f1.4
100iso
Outdoor shoot with grass, hedges and trees.
Patriot LX class 10 8Gb:
One file recorded for 2.5 minutes OK, I stopped recording
2nd file received "card speed error, etc" after 45 seconds
3rd file received "card speed error" immediately.
4th file recieved "card speed error" immediately.
AData Class 6 8gb:
Immediate "card speed error" on all tries.
Barry_Green
Moderator
I would be pretty surprised if it did. But let's do the math...
The difference between video frame rates and film frame rates (in NTSC-land) is 0.1%. Hence why 30fps is actually 29.970 (30*.999) and 24fps is actually 23.976 (24*.999). You'd think it being 99.9% the same speed would minimize any issues, right? But actually, because there are SO many frames (1,440 per minute at 24fps) it means that you can lose sync at the rate of about 1.44frames every minute. 1440*.999 = 1438.56, a difference of 1.44 frames per minute. So if you recorded video at 30.000 fps and edited on a 29.97 timeline, and tried to sync up your audio, you'd actually end up 1.8 frames out of sync after a minute.
That's why the Canon 5D's 30p mode was so problematic, prior to the firmware update that converted it to a true 29.97. And why the GH1's MJPG mode is iffy, because I think it's also 30.000 and not 29.97, isn't it?
But now let's look at 23.971 vs. 23.976. The proper rate for 24p for video is 23.976. If it's true that the GH1 footage is actually technically 23.971, that means a difference of .99979. That's really close. It means that you'd have to run a continuous take of 4,796 frames before you'd lose even one frame of sync. And at 24p, 4796 frames would mean 200 seconds, or 3:20 of running time. So, yes, potentially it could mean an audio sync issue of three frames every 10 minutes. And frankly, your audio pretty much has to be out by three frames before anyone is even likely to be able to notice; for history remember that the original DVX100 had an audio sync issue where all audio was out of sync by two frames in 24p mode, and nobody even noticed for six months! And the original Varicam had the same issue, and I don't think anybody noticed for years.
For longform work, if you're recording a continuous hour of footage, then yes perhaps the audio could be out of sync by 18 frames at the end of an hour. That would be a problem, for someone shooting a lecture or something. Of course, if there are any breaks or silent parts you could always re-sync the audio at those points and get complete sync again, in editing.
But is it really an issue? I don't know that the footage is truly running at 23.971 fps. It doesn't really make any sense as to why it should be. And remember, lots of editors round off "23.976" to become "23.98". The difference between 23.976 and 23.98 is almost exactly the same amount of difference as the difference between 23.976 and 23.971, you're talking about adding .004 instead of subtracting .005, so -- if rounding off 23.976 to 23.98 is good enough, then rounding down should be within an acceptable margin of error.
I'm not concerned about it, but I guess some longform tests should be run. Maybe someone with a Zoom or other external audio recorder should try a simple sync test; clap once at the beginning of a clip, let it run an hour, and clap again at the end, then bring in the audio and the video into an NLE and try to sync 'em up and let's see what happens.
Sounds like your cards flubbed. But try to delete everything on hte card, turn off camera, remove battery, then restart and shoot.
Paul Shields
Member
Actually, looking at the high ISO videos I uploaded they do look pretty noisy in parts, and there is quite a bit of banding going on too (this is when viewing the mts files too). I guess higher bitrate can't compensate for the failings of the sensor.
Svart, was that with Video Buffer on 40.000.00 or 84.000.00? Since both have been associated with the "D" settings here...
Also did you do a low level format on the GH1 with both those cards before shooting? That could make a difference too (or not, but just to eliminate that variable).
Vitaliy Kiselev
Carbonite Member
According to sensor readout speed exact fps is the following - 23,97602
If anyone want more decimal places, PM me
Vitaliy Kiselev
Carbonite Member
I'll clear lens discussions after some time, ok?
Which settings to use?
Which settings to use?
I'm sorry -- but which settings are the best then?
Because in the "How to use ptool" post Car3o says:
But in ptool it suggests:
VIDEO BUFFER..................................40.000.00 0
VIDEO BITRATE ADJUSTMENT)............50000000
OVERALL BITRATE ADJUSTMENT.........52000000
LIMITING BITRATE ADJUSTMENT.........60000000
Which settings to use?
I'm sorry -- but which settings are the best then?
Because in the "How to use ptool" post Car3o says:
But in ptool it suggests:
VIDEO BUFFER..................................40.000.00 0
VIDEO BITRATE ADJUSTMENT)............50000000
OVERALL BITRATE ADJUSTMENT.........52000000
LIMITING BITRATE ADJUSTMENT.........60000000
Barry_Green
Moderator
We don't really know what settings are "best". That's what these threads are for -- so people can experiment with different values and find out how they perform, and share those results with us.
Yes both cards were formatted.
this was from:
VIDEO BUFFER..................................40000000
VIDEO BITRATE ADJUSTMENT)............50000000
OVERALL BITRATE ADJUSTMENT.........56000000
LIMITING BITRATE ADJUSTMENT.........60000000
I'll redo the firmware just in case I screwed something up, reformat and then retest.
Last edited:
http://www.sendspace.com/file/f3l339 [49mb, MTS format]
More codec torture, but this time just with a super detailed scene in deep focus. "D" settings (40.000.00 buffer), "Smooth" mode with no adjustments, f/11, 1/50, 14-140 kit lens and the focal length was probably 20mm or so. And I had my Fader ND dialed in as well.
This scene froze my SD card (8GB class 6 Kingston) several times after about 8 to 10 seconds of recording. Similar results with MJPEG. It is obscenely sharp when viewed on my 42" Samsung LCD via PS3.
http://www.sendspace.com/file/f3l339 [49mb, MTS format]
More codec torture, but this time just with a super detailed scene in deep focus. "D" settings (40.000.00 buffer), "Smooth" mode with no adjustments, f/11, 1/50, 14-140 kit lens and the focal length was probably 20mm or so. And I had my Fader ND dialed in as well.
This scene froze my SD card (8GB class 6 Kingston) several times after about 8 to 10 seconds of recording. Similar results with MJPEG. It is obscenely sharp when viewed on my 42" Samsung LCD via PS3.
Good grief. So much detail. Thanks forr that, just re-enforces the need for me to grab some Class 10 Cards so I never have to run into freezing.
I think the term you're looking for is RESOLUTE. Barry taught me a long while back that resolution and sharpness are two different things. This can be sharp and have little resolved detail, but it looks "sharp" because there's SO much freakin' resolved detail.
IMPRESSIVe.
Barry_Green
Moderator
Yeah, that's pretty much the experience I've been having, watching the GH13 footage. It's so detailed. Shots that I just knew, from past experience, would be a muddy mess -- the GH13 doesn't even blink at it. Chews it up, spits it out, and asks for more.
I used to DREAD deep-focus shots. I'd avoid them however and whenever possible. Deep-focus on a DSLR used to mean D.O.A.
Not anymore. The GH13 has no qualms with deep focus, it's totally added a completely new tool to the toolbox. It's an incredible improvement.
And to anyone wanting to test it, I'd say that you should pretty much forget any Class 6 card, and any off-brand. If you want to use this high-bitrate, high-bandwidth footage, go for the gold and get a Sandisk Extreme Class 10, or a Panasonic Class 10.
Tested one of the alternate "D" settings
VIDEO BUFFER..................................84000000
VIDEO BITRATE ADJUSTMENT)............60000000
OVERALL BITRATE ADJUSTMENT.........56000000
LIMITING BITRATE ADJUSTMENT.........60000000
no card error recording however the files have encoding artifacts, large macroblocking and pulsing areas.
Tested on both Patriot LX class 10 and AData class 6. All files have same problems.
VIDEO BUFFER..................................84000000
VIDEO BITRATE ADJUSTMENT)............60000000
OVERALL BITRATE ADJUSTMENT.........56000000
LIMITING BITRATE ADJUSTMENT.........60000000
no card error recording however the files have encoding artifacts, large macroblocking and pulsing areas.
Tested on both Patriot LX class 10 and AData class 6. All files have same problems.
Psynema
Well-known member
Doubt it will happen,
but just watch Panasonic release their version of this just so it's official...hopefully the hack won't interfere with future firmware updates.
But if anything, it let's you know what the hardware can really do and maybe Panny will revamp the GH2's firmware to take advantage .
but just watch Panasonic release their version of this just so it's official...hopefully the hack won't interfere with future firmware updates.
But if anything, it let's you know what the hardware can really do and maybe Panny will revamp the GH2's firmware to take advantage .