Keith Paisley
January 27th, 2009, 09:54 AM
I saw an earlier thread (http://www.dvinfo.net/conf/cineform-software-showcase/133853-importing-cineform-into-virtualdub-601-709-rgb-issues.html) that touched on some of what I'm about to ask, but there were some statements by David Newman that left me with a few questions.
First off, I am a Canon 5d MKII owner and I edit with Vegas on a fairly stout PC (quad core, 6GB, 2.5+TB, XP/Vista 64-bit (dual boot). Not that it matters, but I also own a Mac. I also own a Canon XL2 and a pair of HV20s but the biggest reason I am looking at Cineform is the 5D Mk II.
The two key things I am looking for from my intermediate codec are:
1) very easy integration into the 5d MK II editing workflow. I don't want to add more than one step. Ideally I dump the files from the card straight into a transcoding app and get EXACTLY what I need in my intermediate files
2) Superb quality and fast timeline performance.
BONUS) points are awarded if the software somehow enhances my editing when mixing the 5D Mk II footage with HV20 or other HD footage.
Okay, so now for my questions.
I downloaded and installed the Neo Scene trial app and i'm racing against the clock (3days?!?) to see if I can make it do what I need it to do.
Obviously the included Neo Scene application doesn't help me with transcoding the 5D MK II footage (all 1080p), so I've been trying to make other things work. So far the only reliable way to get to where I need to be seems to be using Virtualdub. That kills #1 on my list because to use Virtualdub I have to create avisynth scripts for each of my files and then use virtualdub's batching tool, which is a little bit on the weak side.
In any case, it seems that if I want to enjoy the benefits of Cineform (it meets #2 on my list above), I'm stuck with the Virtualdub method for now. So this is where the questions come in.
Everybody is probably familiar with the 5D MK II's infamous YUV-RGB conversion woes (there is still a minor glitch with Quicktime 7.6) and I'm just trying to make sure I'm doing everything correctly and that the transcoded files contain every bit of detail that they possibly can.
So, here go the questions:
1) using CoreAVC as the decoder, what is the best format to present to cineform (CoreAVC settings)? I would assume it would be the native colorspace that cineform uses. Is that YUY2? YV12? RGB? Note that counter to what Mr. Newman stated in the referenced thread above, according to the Virtualdub documentation: "For an uncompressed source (or Avisynth script), the desired format is produced through image conversion if it is not the same as the source. YCbCr-to-YCbCr conversions do not round-trip through RGB.". To me this means that we can transcode YUV-native files to other YUV codecs without any RGB conversions, and I would assume that even if Cineform's native color space is RGB, that it should handle the YUV->RGB conversion possibly better than Virtualdub or CoreAVC.
2) I assume that to maintain maximum detail I should set the levels to 0-255 for both input and output on the coreavc configuration, correct? Otherwise, it seems I would be companding the data and I can always use color correction to set back to Studio RGB levels.
3) the default 709 matrix should be used on the cineform HD codec configuration, correct?
thanks.
First off, I am a Canon 5d MKII owner and I edit with Vegas on a fairly stout PC (quad core, 6GB, 2.5+TB, XP/Vista 64-bit (dual boot). Not that it matters, but I also own a Mac. I also own a Canon XL2 and a pair of HV20s but the biggest reason I am looking at Cineform is the 5D Mk II.
The two key things I am looking for from my intermediate codec are:
1) very easy integration into the 5d MK II editing workflow. I don't want to add more than one step. Ideally I dump the files from the card straight into a transcoding app and get EXACTLY what I need in my intermediate files
2) Superb quality and fast timeline performance.
BONUS) points are awarded if the software somehow enhances my editing when mixing the 5D Mk II footage with HV20 or other HD footage.
Okay, so now for my questions.
I downloaded and installed the Neo Scene trial app and i'm racing against the clock (3days?!?) to see if I can make it do what I need it to do.
Obviously the included Neo Scene application doesn't help me with transcoding the 5D MK II footage (all 1080p), so I've been trying to make other things work. So far the only reliable way to get to where I need to be seems to be using Virtualdub. That kills #1 on my list because to use Virtualdub I have to create avisynth scripts for each of my files and then use virtualdub's batching tool, which is a little bit on the weak side.
In any case, it seems that if I want to enjoy the benefits of Cineform (it meets #2 on my list above), I'm stuck with the Virtualdub method for now. So this is where the questions come in.
Everybody is probably familiar with the 5D MK II's infamous YUV-RGB conversion woes (there is still a minor glitch with Quicktime 7.6) and I'm just trying to make sure I'm doing everything correctly and that the transcoded files contain every bit of detail that they possibly can.
So, here go the questions:
1) using CoreAVC as the decoder, what is the best format to present to cineform (CoreAVC settings)? I would assume it would be the native colorspace that cineform uses. Is that YUY2? YV12? RGB? Note that counter to what Mr. Newman stated in the referenced thread above, according to the Virtualdub documentation: "For an uncompressed source (or Avisynth script), the desired format is produced through image conversion if it is not the same as the source. YCbCr-to-YCbCr conversions do not round-trip through RGB.". To me this means that we can transcode YUV-native files to other YUV codecs without any RGB conversions, and I would assume that even if Cineform's native color space is RGB, that it should handle the YUV->RGB conversion possibly better than Virtualdub or CoreAVC.
2) I assume that to maintain maximum detail I should set the levels to 0-255 for both input and output on the coreavc configuration, correct? Otherwise, it seems I would be companding the data and I can always use color correction to set back to Studio RGB levels.
3) the default 709 matrix should be used on the cineform HD codec configuration, correct?
thanks.