HDV to ProRes 422

[Abd Allateef]

Capturing HDV footage through the HDV-ProRes422 step store large files but easy to edit HDV footage...fact
but what about pre-captured HDV footage through normal 1080 50i HDV capture....will it look and behave the same as the HDV-ProRes footage after importing and rendering it into a ProRes422 sequence?

"i have 12hrs of last week trip HDV transferred footage on my hard drive... i hate to recapture them again"


thank you all,

[Michael Galvan]

I have done this before...

Unfortunately for you, recapturing into ProRes will yield a cleaner image. Taking captured footage that was captured into the HDV codec and then transcoding into ProRes yields noisier images. It must be from going through multiple compressions.

[Alex Humphrey]

how much faster is it to render pro-res vs HDV? I live capture with a FE-DTE drive so I skip the capturing process, edit and i generally take 2:1 ratio for time to downconvert 720p 24fps HDV to 24p DVD at best settings. so 1.5 hour final product takes 3 hours. What would capturing in pro-res (lose 1.5 hours) save me on the render time?

thanks

[Abd Allateef]

thanks allot....


re-coding be it then........there go 12 more hrs and few 100s gigs of space.


thanks again

[Aric Mannion]

Originally Posted by Michael Galvan

I have done this before...

Unfortunately for you, recapturing into ProRes will yield a cleaner image. Taking captured footage that was captured into the HDV codec and then transcoding into ProRes yields noisier images. It must be from going through multiple compressions.

Are you saying that changing your HDV sequence settings to prores in Final Cut does not create an identical image as capturing prores in the first place? Cause I think it does. Or are you only referring to transcoding (compressing?) the footage lowering the quality?

[Shaun Roemich]

I've noticed the same noisy artifacts Michael mentions with HDV captured material run through fairly aggressive shadow/gamma/highlight correction using the built in 3 way cc in FCP, even when rendered to ProRes. The ProRes captured media (from 720P60 HDV) is MUCH cleaner in terms of mosquito noise.

[Aric Mannion]

Well that sucks cause I've asked this question 100s of times, and have done hours of research and everyone has always said that changing HDV sequence settings to prores is the same as capturing the HDV as prores in terms of quality.

[Shaun Roemich]

I can only speak to my very own specific set of circumstances but in my case it was a two camera shoot of a live band in a VERY dark venue and when I captured the ENTIRE performance in HDV versus the initial run of two songs to cut, I noticed SIGNIFICANT mosquito noise in the HDV material (after dynamic range processing) when rendered ProRes whereas the ProRes media in a ProRes timeline was pretty darned decent. Again, YMMV.

[Chuck Spaulding]

Nomenclature is important in these types of conversations. Are you talking about converting HDV to ProRes OR editing HDV in a ProRes timeline for example.

I edit approximately 4TB's a month of XDCAM EX (Really cool HDV). Part of the confusion regarding workflows is that there are several variables to consider. I'm not going to go into all of them, I'll just explain what and why we have the work flow we do.

First and foremost, the priority for us is time, we have almost 40TB's of shared storage so although there's never enough storage we manage it well without consuming a lot of time copying clips from editor to editor.

We bring the XDCAM clips from solid state media into the file system using the Sony XDCAM Transfer, this is analogous to Log and Transferring HDV from tape. We leave it in native XDCAM and do our first editorial cut is in an XDCAM timeline. We set the codec in the timeline render control to ProRes so if we add any effects or filters that require rendering the render files are ProRes. After editorial the project is sent to Color, corrected and exported to ProRes. Color works natively in ProRes so renders happen at about quarter realtine (six frames a second or so).

You can place XDCAM and HDV clips into a ProRes sequence and you should not have to render anything to edit. If occasionally you will have to render effects or transitions depending on what they are. The quality of this workflow is outstanding.

The first couple of projects we converted XDCAM and HDV to ProRes, this took a while and required considerably more disk space. If your limited on disk space then you can also edit in an XDCAM EX timeline and get the same quality with much smaller output file. The tradeoff is time, our ProRes sequences render in near (or sometimes faster than) real time but the file is about six times larger. Outputting from an XDCAM sequence yields great quality and a smaller file but it takes about 7 hours to render/conform a 2 hour XDCAM project.

There's always a trade off, but since we use Color for everything, that's the point in our workflow where we made the transition to ProRes. We get the great quality with manageable render times.

I hope this helps.

[Abd Allateef]

My Q was : is it the same quality(for editing) to capture pure HDV then decode and edit on a ProRes422 time line as capturing ProRes422 and edit on a ProRes422 timeline.......

My main edit steps are : levels, speed, smoothcam then output 20-30mbs 720p H.264.....

Dont we all wish cameras record internaly ProRes 422 on SSDs swapables (without the Aja Kipro shoe box)


thanks,

[Chuck Spaulding]

This is the nomenclature part...

If you capture HDV natively, you don't have to do anything to it to edit in a ProRes timeline, just place the HDV clip directly into the the ProRes timeline and edit.

For display purposes FCP just edits the HDV natively, the encoding only takes place when you export it. If you place titles, transitions or video filters on an HDV clip in the ProRes timeline most often you might get a green render bar. If you want to view that full quality then you'll have to render it. If you set your render files to ProRes in user preferences, your render files will be ProRes and you won't have to render those again.

As far as quality goes, Apple claims the ProRes is "virtually" lossless, since most editors do not work with uncompressed HD they are "virtually" correct. There will be no perceptible difference from your native HDV source files and the ProRes exported file.

[Sherif Choudhry]

Chuck, in yr opinion would you capture in prores standard or prores HQ if your final output is DVD (and included color correction).

Thanks

[Chuck Spaulding]

Short answer:-| ProRes Standard is all you need for SD-DVD, Blu-ray and good color correction.

Prior to FCP7 ProRes standard is 145Mbs/sec and ProResHQ is 220Mb's/sec. With the release of FCP7 there are some other variations.

For SD-DVD and Blu-Ray, Standard ProRes is more than adequate, just for reference Sony HDCAM is 144Mb's/sec.

ProResHQ and the newer (FCP7) ProRes 4444 are for outputting to film, again for refence the latter is equivalent to Sony HCCAM SR @ 440Mb's/sec (although I believe ProRes 4444 is 330Mb's/sec with alpha channel). Your more likely to use this codec if you shot with a Red or HDCAM SR.

I have not tested the output of the newer ProRes(LT) 100Mb codec but I'm pretty sure that it is sufficient for Blu-ray as well.

Although this sounds confusing, its actually quite exciting. Depending on your source (camera) with devices like the CD NanoFlash and Aja Ki Pro you can capture full raster HD in 4:2:2 with the potential for stunning results. I have not tried either of these devices yet but really look forward to it.

I hope this helps.

[Brian David Melnyk]

so just to be clear, if best quality is the only factor, and one does a lot of titles, CC, transitions, etc etc, would the best HDV workflow be:

1. capture HDV, edit in HDV timeline with prores render
2. capture HDV, edit in prores timeline
3. capture HDV as prores and edit in prores timeline

i am unsure what the differences are if everything is being rendered as pro res anyway... how does HDV captured as prores differ from HDV rendered as prores?

[Shaun Roemich]

Brian: my completely unscientific testing has shown ME (YMMV) that extensive dynamic range recovery (dark images in a dark club) hold up better by capturing HDV as ProRes and editing that way. My blacks and dark greys exhibit significantly less mosquito noise.

This is versus capturing HDV, placing in an HDV timeline and rendering as ProRes.

Please do not take this as the gospel, this is just based on a recent live band EPK I did and the differences were tangible and noticeable even to the client.