DV100 vs DNxHD vs CFHD

[David Cherniack]

You are partially correct. Full rez is available while scrubbing. Full rez in play probably by NAB

David
AllinOneFilms.com

[Graeme Nattress]

I”ve seen such shadow problems with DVCproHD footage when played back in the FCP canvas or in Digital Cinema Desktop mode, but I cannot see the same issues on the same footage played back through the Decklink + HDlink to the very same Cinema Display.

Again, if you’re saying that the software codec is inferior to the hardware codec, it’s irrelevent to talk about using a different intermediate codec. If the hardware and software codec are equal in quality, then there is no advantage to coming in over SDI to gain access to the hardware codec, and again, with P2 you don’t capture, you edit.

Graeme

– http://www.nattress.com – Film Effects for FCP

[Jason J rodriguez]

I still think that you’re being a bit short-sighted about the benfits of a good intermediate codec, especially when there’s more to the world than keeping your entire project inside Final Cut Pro.

For instance the current digital cinema project I’m working on has been onlined in Final Cut, but was then exported for 16-bit color correction and effects inside of After Effects. Since I’m working with Uncompressed 10-bit Blackmagic stuff, the file sizes are huge, and it would be a nice feature to have the same image quality in a file-size 1/6th the size, like Cineform and DNxHD can do.

DVCProHD is not the ideal solution for these types of multi-platform, multi-program, multi-compression workflows. If all you’re doing is working inside of FCP, then that’s great, but for many of us, that just isn’t the case, and that is where a nice intermediate codec that withstands the tortures of repeated compression, passes, etc. with flying colors comes in very handy.

You call the tests that Cineform did as “not fair”, yet this is exactly what you have to deal with if you use the native DVCProHD codec outside of Final Cut, in Shake, After Effects, Combustion, etc. When that becomes the playing field, then I feel the results are very fair, because that is exactly what I’m going to have to deal with when I use DVCProHD outside of Final Cut’s native DVCProHD timeline.

So, I’m not saying that DVCProHD is no good, but I’m saying, it’s NOT a true intermediate codec, meaning that it’s not made to withstand repeated YUV-RGB conversions and be visually lossless, survive multiple passes between different programs for effects, color correction, etc.

Basically, it’s a crappy codec for Digital Intermediate work. Does everybody need to do a Digital Intermediate? No. There are many programs that do just fine adding their effects, graphics, color-correction, etc. all inside of Final Cut, and then drop back to tape. But there is another side of the industry, those people who are working on films, high-end commerical work, etc. where a visually lossless digital intermediate codec can come in very handy. Uncompressed is great, but like I was saying before, it has it’s problems when it comes to speed, bandwidth, etc. If you can do all your digital intermediate work and maintain visual perfection, but have a file size that’s 1/6th the size of uncompressed, I think there are some definite advantages to that workflow. And of course AVID and Cineform have seen the market potential in that mid-range to high-range area, and I hope that Final Cut doesn’t get left behind because they think that DVCProHD is good enough for that kind of work-cause it’s not.

Who knows, maybe core video will enable all sorts of real-time stunts on uncompressed HD footage, so this whole argument is moot.

Jason Rodriguez
Virginia Beach, VA

[Luis Caffesse]

Jason,

Pretty much every project I do goes through AE at some point (in the end for color work for example).
I use Automatic duck to rebuild my timeline in AE, and then from there I can render out to whichever codec I want my master to be.

So, are you saying that CFHD codec will handle that workflow better?

It definitely gets a bit weird to follow this when you put it into the P2 context, or any direct-to-disk recording. Essentially I would have to transcode my footage to CDHD, then pull that into AE, and then render out to whatever I want once I’m done in AE?

Just trying to get a handle on this.
This is interesting to me, because I know many people who work with intermediate codecs, and it’s an issue that I never thought about before in terms of P2. The thing that really turned me off to HDV was having to transcode your footage. Granted, no one is saying you have to do this with DVCProHD, only that you may get better results.

Luis Caffesse
Studio 3 Productions, Inc.
Austin, Texas

[Graeme Nattress]

I’m sorry, but I just don’t see any kind of sensible post production workflow that repeatedly compresses the footage over and over and over. If you edit DVCproHD, send that over to AE, then AE decompresses the frame that it’s rendering into memory, and at that point, that frame is totally uncompressed. AE then does all it’s magic, and when you render, you get to pick which codec you want to render to. Say you’re going back to DVCProHD tape, then you can render back to that, or say you’re going to D5, so you render uncompressed. At what point do you need an intermediary codec, and how can an intermediate that is compressed in any way be better than going to that frame in memory totally uncompressed??? I really just don’t get what kind of workflow would warrant repeated recompressions….

Graeme

– http://www.nattress.com – Film Effects for FCP

[Graeme Nattress]

But under what circumstances would you repeatedly recompress DVCproHD?? Would you edit it in FCP, apply colour correction, export to AE, do more colour correction, and then export to shake to add some final colour correction before sending back to FCP for some extra final colour correction??

Unless you’re going to show that the software codec works differently from the hardware codec, necessitating capture of DVCproHD over SDI to 10bit uncompressed, any use of an intermediate codec to cut down files sizes is, I’m afraid to say, daft. I’m really at a loss to see how it fits in and works. Whenever you apply an effect, say in AE to some DVCproHD footage, AE must decompress it first. Do that in a 16bit project and AE will decompress to a 16it representation in memory, and there you are – uncompressed, in memory, ready to do your effect. Then you set an appropriate output codec. Final render and out. There’s no repeated mucking around between umpteen different applications, repeatedly decompressing and recompressing the video. This was all hammered out ages ago with DV, and you just don’t work like that. You can get equally good results without resorting to mucking around with proprietary intermediate codecs that you just don’t need.

I think you’ll have to show us a workflow where you’d have to repeatedly compress and decompress the DVCproHD back and forth and time and again. I’m really at a miss to see a) why you’d want to do that, and b) why an intermediate codec would help better than adopting a more suitable workflow that uses the benefits of native codec editing for you, rather than against you. I’m bemused….

Graeme

– http://www.nattress.com – Film Effects for FCP

[Luis Caffesse]

[Graeme Nattress] “I really just don’t get what kind of workflow would warrant repeated recompressions…. “

Me neither Graeme, which is why I asked.
You and I are on the same page on this one.
I’m just trying to make sure I understood the workflow that Jason was proposing,
especially in relation to P2.

Luis Caffesse
Studio 3 Productions, Inc.
Austin, Texas

[Graeme Nattress]

Thanks Luis. I’m happy that I’ve not gone completely gaga yet….

Seriously Jason – help us out here to understand what you’re getting at.

Graeme

– http://www.nattress.com – Film Effects for FCP

[Toke]

Are you really saying that camera manufacturers has left out
electronic low-pass filtering with these non-changeable lense cameras?

Anyway, if you have seen pictures from that Italian modded fx1 with eng-lens,
it shows much more resolution.

And none of real interchangeable lens camera leans to low pass filtering of the lens.

You can put Zeiss Digiprime in front of any camera and see no anti-aliasing.
All filtering is made electronically after ccd.

[Graeme Nattress]

I thought there was some kind of optical filter block after the lens, but before the CCD??

Graeme

– http://www.nattress.com – Film Effects for FCP