Mark Spano
Forum Replies Created
-
No need to upgrade to make Blu-ray encodes. The reason you’re getting low bit rate as an option is probably due to how Compressor thinks you are going to use this file.
Does yours look like this?
If so, take a look again. It is saying AVCHD as the usage. That means it thinks you want to burn Blu-ray content onto a standard DVD. Not what you want? Try changing it to Blu-ray:
Now you get proper Blu-ray encoding parameters.
-
You’re welcome – glad to hear it.
-
[Sarah Davis] “One issue is that there is a flash to colour bars near the beginning which I guess is compressor’s way of saying – “on no lost the plot for a moment”. No other visual issues at all.”
Yeah I have seen slight visual issues like the one you describe. They are rare and usually only crop up near the beginning or end. I usually bring the resulting file into FCP to do a QC check anyway, and excise any errant frames like that. It’s definitely a good idea to watch it down to check.
[Sarah Davis] “Wondering if I should do the NTSC conversion to the same master format as the PAL master I am working from and then do the mpeg version?”
If you’re saying that you think you should make the DVD from the new 29.97 master file, then I agree. Compressor is happier to do one thing at a time, so doing a straight MPEG encode from the new file will be relatively quick. Your settings look OK, but think about setting the frame controls to best resize and fast everything else, since you’ve already handled the retiming and there is no deinterlacing necessary. Everything else is good.
-
[Glenn Camhi] “it sounds a bit like taking a shot in the dark”
Yeah the whole thing does.
I can’t understand how this can get screwed up. DVD encoding has been around and perfected over the last, I don’t know, two decades? It’s not like this is uncharted territory.
Point one: if you have DVD encoding equipment that can’t handle 23.976 fps, you have very outdated and poor encoding equipment. It might be awesome at what it does, but 23.976 is the norm for every major feature release on DVD. And that was established a long time ago.
The simple fact that you can do professional 23.976 encoding and simple authoring yourself with a $50 piece of software (Compressor) should be enough to elicit the proper amount of derision toward this company who’s handling your encoding. There’d have to be a pretty amazing reason why it doesn’t look right. And them thinking that another pre-encode process (they’re going to try lowering the brightness themselves a bit) will solve it is just a joke.
-
You could have learned that by reading this thread, specifically above where I said:
[24p encoding] actually looks better since you can take the fixed amount of bits and spread them across fewer frames.
About your other comment regarding washed out video: whoever is encoding is clearly doing something wrong. What you put in should come out looking relatively the same, video levels wise. People think they’re doing you a favor by adding filters in encoding – they’re not. If you’re paying for this, keep having them do it over and over until they get it right. There’s no reason it should look washed out.
-
[Sarah Davis] “speed and pitch are important”
If that is the case, then abandon what you’ve already started. Cinema Tools does not do frame rate conversion. It does conform. Conform basically just tells the file to play back at a different speed. So now your 25fps file is playing back at 29.97fps, so it is playing back faster. Check the total running time. Much less I bet. So speed/pitch are not going to be accurate with this method. If you listen to the audio, it’s in squeaky voice land.
What you need is the hardware conversion you alluded to. This will be the best. However, if you’re looking for “good enough for now”, then you should use Compressor. Compressor will allow you to transcode, and with optical flow, it will create the frames in between to get you to 29.97 with the same running time. Audio is not touched at all, naturally, since you’re keeping the same running time.
You want to create a preset in Compressor that changes frame rate, using Frame Controls to engage optical flow to retime your video from 25fps to 29.97fps. If you need help with a preset, I can probably whip one up and pass it along. Be prepared for the processing to take a while, as retiming takes a while in software.
-
Well it’s clear that the second one looks better than the first in these images. But what are we looking at? I mean, I always need to view output material on a broadcast monitor to judge true quality. You say these are widescreen, so clearly now in these images there’s been a conversion to full height anamorphic, presumably for your tape output. There’s another avenue where distortion could take place that you wouldn’t necessarily be able to judge unless you’re looking at the tape output. The screenshots – are they of a 100% raster viewer window? I know that I can’t judge anything accurately on my computer screen unless I’m looking at a 100% size viewer. And even then, the viewer is showing either a single field, or two fields blended, so that’s no way to judge for interlacing effects. Both of these could look identical when played out to a CRT. So we’re back to the question of proper workflow. You are rendering out of AE as what codec? AVI is a container, so what’s inside? I would use either Animation codec (this is called Lossless) or straight to Apple ProRes. Thinking that it’s a progressive animation, I’d render without fields, scaled to the output resolution (720×576, 25fps). Then you’re making sure AE is handling any scaling involved, which is probably better than Premiere/FCP/MC. I’d import the video straight to a sequence that matches the codec/fps/field order of the source. That’s really it. I don’t see how it can get messed up in that workflow. The card will handle the interlacing to get it out to display/tape. But this will look identical to your progressive animation, so it won’t matter that the output is interlaced as you won’t see any interlacing lines. Note that I am not talking about the computer screen viewer, as I mentioned before, that is not reliable to judge for quality. Only the actual video output from the card is important. If it is still not to your satisfaction, then there may actually be something wrong with the hardware, and you may need to contact the manufacturer.
-
I don’t think it makes a difference there, since you’ve got just two tracks those two settings should come out identical…
-
As long as the sequence setting is set for 2 Outputs, Stereo Monitoring, and the channels set to Channel Grouped (with A1 panned hard left and A2 panned hard right), then the sequence playback should be exactly what is in the Quicktime. I can’t think of anything you’d have wrong there.
-
This doesn’t surprise me, since I believe 25p PAL is not in the standard definition spec for SDI. 25p PAL exists only inside your computer. As soon as it goes out over SDI, it has to be interlaced. This is usually not a problem, as I said before:
DigiBeta is always interlaced. To get video onto DigiBeta tape, it has to go down as 720×576 / 25i. So the Decklink is doing you a favor by taking your progressive signal and splitting the frames into fields.
So your 25p PAL signal is 25 frames per second, progressive. This gets translated by the Decklink into 50 fields per second, interlaced. There is no degradation or signal loss. The 50i signal goes to tape and is 50i on DigiBeta, since DigiBeta is interlaced. I’ll say it again: DigiBeta is interlaced. You can’t change that. So the card is doing it correctly. This 50i interpolation is the best (and only) possible way for your 25p PAL stuff to get to tape.
If you want true progressive, you need to go HD, at 1280×720 50p.
If you’re concerned about quality, look at your signal and monitoring path. This is the only advice I can give based on what you have provided.