Raven Plenty
Forum Replies Created
-
Thank you very much Zane. I did not notice that there was a forum for “Compression techniques” – I looked for Apple Compressor, and since I didn’t see that went with what I thought was the next best. I’ll repost there since I’m still interested in more information about this (i.e. if there is anything unusual happening or if this is normal).
-
I have a feeling that this thread isn’t being read from the beginning.
I realize that it takes longer to transcode from one compressed format to another. However, this H.264 file is taking way longer than a different (and much higher res) H.264 file to transcode to MPEG-2. Please review my earlier postings for more details.
Also, thanks very much for your help.
-
I’m not using Final Cut, nor am I doing any editting. Only transcoding using Compressor.
I’m not making the case for Handbrake as a professional worktool, it was a last ditch effort to create a backup copy of a single copy-protected DVD we were going to use for a convention. Nor am I intending to use H.264 files in an editing workflow.
All that aside, are you suggesting that the H.264 file was taking extra long to transcode to mpeg-2 because it was created by Handbrake?
-
The other difference is that the 20-minute video was created by Handbrake (as an H.264 mp4 file, which I resaved as MOV) whereas the 2-minute video was created from a Combustion animation. But H.264 is H.264, no?
-
FCP –> Compressor –> DVD Studio Pro
The blockiness looks the same in both the m2v file and the DVD itself, which makes sense since DSP shouldn’t be changing anything.
-
Thanks Rafael. Actually, I have successfully applied the 4:1:1 filter to my nested sequence. The result is clearly visible in the canvas as well as in my exported m2v file. No problem there. Also, yes I agree the blockiness is introduced in the MPG2 encoding. And it looks like the compression markers are retained even with the nested sequence. I have tried a higher bit rate with no visible difference. (It was already the default “best quality” setting of 6.2Mb/s – max 7Mb/s.) Is this what people are talking about when they say Compressor sucks?
I don’t know much about GOPs yet…I’ll look into that, and your 10-bit suggestion. Thanks.
-
Thanks Rafael. Actually, I have successfully applied the 4:1:1 filter to my nested sequence. The result is clearly visible in the canvas as well as in my exported m2v file. No problem there. Also, yes I agree the blockiness is introduced in the MPG2 encoding. And it looks like the compression markers are retained even with the nested sequence. I have tried a higher bit rate with no visible difference. (It was already the default “best quality” setting of 6.2Mb/s – max 7Mb/s.) Is this what people are talking about when they say Compressor sucks?
I don’t know much about GOPs yet…I’ll look into that, and you’re 10-bit suggestion. Thanks.
-
I should add that these are going to be displayed at one location on 16×9 LCD tvs, hence my concern that the blockiness will be more visible than on a CRT.
-
Well, the source footage is DV NTSC and I’ve only nested it in an 8-bit sequence so that I can apply color smoothing. (Applying color smoothing to a DV sequence doesn’t seem to work – effect reverts after rendering.) In short, I’m going from DV –> FCP –> DVD.
I don’t see a change in the canvas view when I apply the 4:2:2 filter, compared to no filter at all. 4:1:1 looks great to me (makes sense, no? since DV is 4:1:1), and looks great on the DVD output as well. Isn’t the color smoothing not relevant to my blocky transition issue though?
Thanks for your help.
-
It’s not DV video, it’s a Flash animation exported as a PNG image sequence, then saved as a Quicktime Movie using Animation compression. I did try your suggestion though…no difference in this case, the Color Smoothing filter looks the same as in my posted sample from earlier. I’m pretty sure it’s simply that the detail is too fine for crappy m2v to handle, particularly this colour combination.
Why do pink/red/orange appear so much worse than other colours?