Robert M Wright
Well-known member
"News" gathering is protected from all sorts of legal issues, otherwise Fox "News" would be getting sued all the time!
Bob's Tests & Comments on the HMC40
- Thread starter BobDiaz
- Start date
Well, then... clearly you were gathering 'news' about smart cars.
And Fox isn't the only new agency to take artistic license with the truth (they did go about making themselves easy targets, though -- and from the group that trademarked 'fair and balanced,' too..').
And how does 'ambush journalism' work? Or all those fools chasing down celebrities with CIA quality spy gear?
Really, I think there is a lot of gray when anyone with an eye toward a blog can be an aspiring journalist/news reporter. Yes, it is better to get a release, but I don't think there is much need to panic.
And Fox isn't the only new agency to take artistic license with the truth (they did go about making themselves easy targets, though -- and from the group that trademarked 'fair and balanced,' too..').
And how does 'ambush journalism' work? Or all those fools chasing down celebrities with CIA quality spy gear?
Really, I think there is a lot of gray when anyone with an eye toward a blog can be an aspiring journalist/news reporter. Yes, it is better to get a release, but I don't think there is much need to panic.
ShugPro Digital Studio
Active member
Oh I didn't mean it like that at all. I was just joking around. :happy:
--------------------------------------------------------------------------
If you're filming in a public place, you're covered, no release needed. If they are waving at you in a friendly way or talking to you and looking at the camera, you're covered. For better or for worse, the explosion of internet video (YouTube) has made it very hard for anyone to take legal action against anyone unless the video is making that person look bad, and even then....
Last edited:
BobDiaz
The Professor
An update on the 720/30p glitch.
Toast 9 does offset the video by 11 frames and for every clip, the video is shifted by 11 frames after the audio. The good news is that by being exactly 11 frames every time, I can fix it one way or another.
VoltaicHD has a different problem: In normal conversion mode, No 11 frame glitch, BUT the video runs at 2x the speed and the audio is normal; very odd. It also takes FOREVER to convert the video...
In the "QUICK MODE" VoltaicHD produces 29 frames of a glitch, BUT the audio stays in sync. I'll have to test this on several other clips to make sure this approach is OK.
This problem is only with 720/30p footage.
For my next test, I'll have to install the Edius software on the PC side of my iMAC and see how it runs.
Bob Diaz
Toast 9 does offset the video by 11 frames and for every clip, the video is shifted by 11 frames after the audio. The good news is that by being exactly 11 frames every time, I can fix it one way or another.
VoltaicHD has a different problem: In normal conversion mode, No 11 frame glitch, BUT the video runs at 2x the speed and the audio is normal; very odd. It also takes FOREVER to convert the video...
In the "QUICK MODE" VoltaicHD produces 29 frames of a glitch, BUT the audio stays in sync. I'll have to test this on several other clips to make sure this approach is OK.
This problem is only with 720/30p footage.
For my next test, I'll have to install the Edius software on the PC side of my iMAC and see how it runs.
Bob Diaz
I don't remember the specifics but both 1080/30p and 720/30p are not native. I am pretty sure 1080/30p is recorded as "over 60i". I suspect that is what is causing VoltaicHD its problems - perhaps is has some setting for that. I use Cineform NeoScene and it does have a setting to resolve the problem. Of course, I never use either of the 30p modes because it is not native. I prefer native progressive so I always shoot it either of the 24p modes or 720/60p. Except for a couple brief tests I have never used 1080/60i, 1080/30p or 720/30p for anything serious. It used to be the case (and I think it still is) that Vimeo converts all videos to 24p; so I dthink 24p works best for Vimeo uploads.
Robert M Wright
Well-known member
It would be nice if the video hosting sites, like Vimeo and YouTube, actually provided detail on what they do when re-encoding uploaded videos. For goodness sake, then we could actually optimize for it nicely without a ton of experimenting!
I suppose simple encoding suggestions for most casual users is a very good idea, but even that is really weak (the suggestions Vimeo and YouTube offer). It floors me that both Vimeo and YouTube suggest encoding videos, for uploading to them, with small keyframe intervals. That makes absolutely no sense whatsoever. To crank down file size and maintain quality, essentially encoding an intermediate optimized for low bandwidth transmission, large keyframe intervals (in the hundreds of frames) are highly desirable. I seriously doubt they actually re-encode videos using the same keyframe intervals as the uploaded files are encoded with.
I suppose simple encoding suggestions for most casual users is a very good idea, but even that is really weak (the suggestions Vimeo and YouTube offer). It floors me that both Vimeo and YouTube suggest encoding videos, for uploading to them, with small keyframe intervals. That makes absolutely no sense whatsoever. To crank down file size and maintain quality, essentially encoding an intermediate optimized for low bandwidth transmission, large keyframe intervals (in the hundreds of frames) are highly desirable. I seriously doubt they actually re-encode videos using the same keyframe intervals as the uploaded files are encoded with.
Robert M Wright
Well-known member
While I'm on a mini-rant, I wish they would allow advanced users to actually do their own encoding. I can't think of any good reason they couldn't offer up specs to conform with, and let us upload our own H.264/AAC encoded MP4 files. That would allow for faster uploads, no need for the service(s) to spend CPU resources on re-compression (simply an automated check for compliance with specs instead), and better final quality. If Vimeo did that, I'd actually consider paying for their service.
BobDiaz
The Professor
The 1080/30p is laid onto a 1080/60i format. I don't know if there's anything to say which 2 fields are part of the same frame or not, maybe every even field is part of the odd field before it. (Pages 39 & 118 of the user manual.)
I do know that the 720/30p is laid onto a 720/60p file. This is the part that's odd, you would think that if the conversion software could deal with 720/60p, that it would be the same for 720/30p laid onto 720/60p. (Page 118 of the user manual.)
The 1080/30p does not pose any problems for Toast or VoltaicHD conversion software. There must be something in the 720/30p file that really screws things up.
I don't have the newest version, Toast 10, so I don't know if it was fixed there or not.
Bob Diaz
imaginator
New member
I was wondering if anybody had any insight on how the low light performance compares to the HMC-150? I primarily do indoor recording for local bands, so the low light performance is a touching point for me, and I'm pretty satisfied how the 150 handles it.
For reference, I am weighing the options between a Canon HF-S100, Pan TM300, and the HMC-40. Pretty big range there, I know. It would be replacing an HV-20 being used as a stationary cam, tired of tapes and the exposure lock trick!
Just curious, but I'm using FCP and can ingest the video from my HMC-150 in directly using the log and transfer functionality. Does this not exist in FCE? I believe support was added in a patch, not sure what version...
For reference, I am weighing the options between a Canon HF-S100, Pan TM300, and the HMC-40. Pretty big range there, I know. It would be replacing an HV-20 being used as a stationary cam, tired of tapes and the exposure lock trick!
Just curious, but I'm using FCP and can ingest the video from my HMC-150 in directly using the log and transfer functionality. Does this not exist in FCE? I believe support was added in a patch, not sure what version...
BobDiaz
The Professor
In comparing the ISOs of the HMC40 and the HMC150, the HMC150 is 3 F-Stops (8x) more sensitive to light than the HMC40. When the lighting is minimal, the HMC40 does a poor job. Yes, I can crank the gain up all the way, but the noise level is very bad. Here's my example of the different gain settings & shutter speeds:
http://vimeo.com/8732993
This video was shot on a well lit street corner. I have not posted the video of my neighbor's house across the street; the lighting was VERY dark. The only way to see the house was to use way too much gain, which resulted in a massive amount of image noise.
With good lighting, even at night, the HMC40 video camera performs well, but with limited lighting, the camera does poorly.
Bob Diaz
http://vimeo.com/8732993
This video was shot on a well lit street corner. I have not posted the video of my neighbor's house across the street; the lighting was VERY dark. The only way to see the house was to use way too much gain, which resulted in a massive amount of image noise.
With good lighting, even at night, the HMC40 video camera performs well, but with limited lighting, the camera does poorly.
Bob Diaz
Robert M Wright
Well-known member
A TM300 and HMC40 use the same imaging chips. There shouldn't be any difference for light sensitivity. Judging from the reviews at camcorderinfo.com, the HF-S100 should actually be pretty close to the TM300 and HMC40 for light sensitivity. All three of the cams you mentioned that you are considering are far less light sensitive than an HMC150, but in the same general ballpark as your HV20.
Robert M Wright
Well-known member
If this cam isn't going to be used for audio, you might want to consider the HS250 instead of a TM300. Same guts for imaging, but considerably less money, and with a 120GB HDD in it, you can set it on an unmanned tripod shooting all day.
in another thread about scene file settings, someone posted an image from shooting a guitarist playing in low light, and the settings they used to do so. Have a look and decide for yourself.
As a stationary cam, and with the 150 to shoot the good stuff, you might want to save what money you can. Canon has new cameras coming out in April. If you want to go the Canon route, then you could probably pick up and HF100 series a lot cheaper once the replacement hits the shelves--assuming you can wait that long.
As a stationary cam, and with the 150 to shoot the good stuff, you might want to save what money you can. Canon has new cameras coming out in April. If you want to go the Canon route, then you could probably pick up and HF100 series a lot cheaper once the replacement hits the shelves--assuming you can wait that long.
Robert M Wright
Well-known member
I kind of wonder how low the retailers will go on the HS250 when Panny releases newer models. At $600 right now, I can image retailers going perhaps below $400 when they clear them out. I may have to twist my son's arm to get one to replace his HV20. He doesn't use an external mic anyway.
BobDiaz
The Professor
I decided to try out the settings used for reduced image noise at high gain levels:
Detail Level -4
V Detail Level 0
Detail Coring +7
Chroma Level -1
Chroma Phase 0
Color Temp 0
Master Ped -3
A.Iris Level 0
DRS Off
Gamma B.Press
Knee High
Matrix Norm1
Skin Tone DTL ON
Now I wish I could remember which thread I found these in, so I could give the proper person credit. The settings are NOT perfect and tend to darken the image too much. The easiest way to play with the settings was to connect the camera to my HDTV and try different settings.
This becomes a matter of personal taste, but these are the settings that seemed the best for low light.
Detail Level -4
V Detail Level 0
Detail Coring +7
Chroma Level -5
Chroma Phase 0
Color Temp 0
Master Ped 0
A.Iris Level 0 (You will need to play with this one.)
DRS Off
Gamma CINE-LIKE V
Knee (Unable to set due to gamma setting)
Matrix CINE-LIKE
Skin Tone DTL ON
These settings will make the image look darker and there's still noise in the image at high gain settings, but it seems more like film grain, rather than video like noise. At about +24dB, the fixed patter noise seemed to appear and I could not find a way around that.
To me these settings seem to make the noise a bit more pleasing to look at.
Bob Diaz
Detail Level -4
V Detail Level 0
Detail Coring +7
Chroma Level -1
Chroma Phase 0
Color Temp 0
Master Ped -3
A.Iris Level 0
DRS Off
Gamma B.Press
Knee High
Matrix Norm1
Skin Tone DTL ON
Now I wish I could remember which thread I found these in, so I could give the proper person credit. The settings are NOT perfect and tend to darken the image too much. The easiest way to play with the settings was to connect the camera to my HDTV and try different settings.
This becomes a matter of personal taste, but these are the settings that seemed the best for low light.
Detail Level -4
V Detail Level 0
Detail Coring +7
Chroma Level -5
Chroma Phase 0
Color Temp 0
Master Ped 0
A.Iris Level 0 (You will need to play with this one.)
DRS Off
Gamma CINE-LIKE V
Knee (Unable to set due to gamma setting)
Matrix CINE-LIKE
Skin Tone DTL ON
These settings will make the image look darker and there's still noise in the image at high gain settings, but it seems more like film grain, rather than video like noise. At about +24dB, the fixed patter noise seemed to appear and I could not find a way around that.
To me these settings seem to make the noise a bit more pleasing to look at.
Bob Diaz
BobDiaz
The Professor
Still photos can never reproduce the dynamic effect of a video, but at least you can get some idea on how the settings look. All photos are a cropped image from a frame in the source video.
The first photo is at +18dB gain with the new settings.
Here's the image at the factory default, Scene 1:
This is +24dB Gain with the new settings:
For those who crank the gain beyond reasonable, +30dB Gain, New Settings:
Here's it is set to maximum with the new setting, +34dB Gain:
I'll have the videos posted later on today, but for now I have to get ready for a class and don't have the time to deal with uploading it. The bottom line is that increasing the gain is never good, but with some changes, we can make the image noise a bit less harsh, but not by much.
Bob Diaz
The first photo is at +18dB gain with the new settings.
Here's the image at the factory default, Scene 1:
This is +24dB Gain with the new settings:
For those who crank the gain beyond reasonable, +30dB Gain, New Settings:
Here's it is set to maximum with the new setting, +34dB Gain:
I'll have the videos posted later on today, but for now I have to get ready for a class and don't have the time to deal with uploading it. The bottom line is that increasing the gain is never good, but with some changes, we can make the image noise a bit less harsh, but not by much.
Bob Diaz
> For reference, I am weighing the options between a Canon HF-S100, Pan TM300, and the HMC-40.
Consider also the HF-S200 simply because of the large 3.5-inch hi-res LCD screen, should be great for focusing. Native 24p. Two card slots. Focus aids like color peaking and magnification. LANC compatibility. Because there is no stupid built-in memory, it is cheaper than the HF-S21.
Consider also the HF-S200 simply because of the large 3.5-inch hi-res LCD screen, should be great for focusing. Native 24p. Two card slots. Focus aids like color peaking and magnification. LANC compatibility. Because there is no stupid built-in memory, it is cheaper than the HF-S21.
The HS300 also has a built-in HDD. The 250 has no focus ring.