I think our workflow is messed up at DVinfo.net

Ryan Szulczewski · November 28th, 2007, 08:26 PM

I posted a thread similar to this in the JVC forum but am not sure I got the answer I needed. First let me detail our workflow as it is.

We shoot on a JVC HD100 with the 30p HDV setting. We than capture from the JVC decks component out to a Kona cards component in. All the info comes from the miniDV tape, we aren't capturing live from the component out of the JVC while we are filming.

I can't explain why I think this but the only benefit to capturing from the component in on the Kona Card would be if we are capturing live from the camera and bypassing the HDV compression that get puts on the tape, right?

I am thinking we should just capture using the HDV preset in Final Cut via firewire. Edit in HD and when we are done downconvert via the Kona Card to the Beta Deck.

Can anyone tell me if I am right or not. I would of course prefer to here that I am right but if we are actually getting quality/filesize benefits from the Kona card than I am all for it.

Oh yeah: this is all done on a Mac in Final Cut Pro

Carl Middleton · November 28th, 2007, 08:49 PM

Unless you are using the hardware of the camera to perform a good job of upscaling, and saving directly into an intermediate format that you would like to use, there's really no advantage.

The data on the tape is 1440x1080 (I am assuming? Not familiar with the HD100) 8 bit color.

The firewire captures that exact digital information that is on the HDV tape, an m2t file. That is all the camera would have to work with, though if it is automatically converting it in realtime to an intermediate codec (i.e. Cineform, ProRes, AIC) that will help your editing process, such as 10bit+ color, I frame, etc it will hold that advantage. As well, it's very possible that the results of using the camera to upsample with the component out might be quite nice.

Carl

Steve Oakley · November 28th, 2007, 09:59 PM

well if you are capturing to DVCpro100, you're losing 25% of your horizontal resolution, plus you are adding another compression hit. not a good gain. if you are capturing using prores, thats good. uncompressed is good too, but uses disc space at a horrendous rate.

if you can capture native via FW, thats still the best rate. very small files, no loss coming in as you are just copying the bits over. problem is FCP is just plain flaky with capturing it. DVHS on mac works much better, actually just about anything works better then FCP for HDV capture.

Ryan Szulczewski · November 29th, 2007, 06:45 AM

I guess I should just do a comparison. It just seems like the 8bit uncompressed files are ginormous with no quality benefit from the HDV footage.

When we capture we aren't upscaling we are downscaling. So we go from HDV to SD through the Kona Card.

Carl Middleton · November 29th, 2007, 08:23 AM

If that is the case, I would take and use the camera to convert HD->SD via firewire. I would then try the SD downconvert using the Kona card as you've been doing.

Compare! :) It's two methods that many people use for the same end result, whichever gives you the best quality and simplifies your workflow (probably firewire, imo) is the right choice for you.

Carl

Ryan Szulczewski · November 30th, 2007, 09:52 AM

So I talked to the other editor about our workflow and why we do it this way.
We do a lot of After Effects work and basic VFX stuff. So if you capture and edit on an HDV timeline in Final Cut Pro if FCP has to render anything it will be lossy (at least that is what he understands).

Plus if we do any After Effects work on the footage and rerender to HDV that is lossy as well.

I haven't tested it yet but I am going to do some tests to see how much the picture degrades working through HDV because those 8-bit uncompressed files are huge.

I captured the same clip (about 10 seconds in length) and the Uncompressed was 222MB while the HDV clip was 44MB. That is no joke.

Ryan Szulczewski · November 30th, 2007, 10:53 AM

OK so I took some HDV footage and messed around with it in AE and rerendered it out using various codecs (HDV, Animation, Konas 8-bit) and no matter what I do it comes out flat. Its like After Effects is doing something weird to the Gamma curve of the footage. I can use Synthetic Aperture to offset it but it usually introduces a lot of noise.

Any thoughts on why the HDV footage comes in looking very flat and muddy in After Effects?

November 28th, 2007, 08:26 PM	#1
Ryan Szulczewski Regular Crew Join Date: Jul 2007 Location: Pensacola, FL Posts: 60	I think our workflow is messed up I posted a thread similar to this in the JVC forum but am not sure I got the answer I needed. First let me detail our workflow as it is. We shoot on a JVC HD100 with the 30p HDV setting. We than capture from the JVC decks component out to a Kona cards component in. All the info comes from the miniDV tape, we aren't capturing live from the component out of the JVC while we are filming. I can't explain why I think this but the only benefit to capturing from the component in on the Kona Card would be if we are capturing live from the camera and bypassing the HDV compression that get puts on the tape, right? I am thinking we should just capture using the HDV preset in Final Cut via firewire. Edit in HD and when we are done downconvert via the Kona Card to the Beta Deck. Can anyone tell me if I am right or not. I would of course prefer to here that I am right but if we are actually getting quality/filesize benefits from the Kona card than I am all for it. Oh yeah: this is all done on a Mac in Final Cut Pro __________________ http://www.OneProVideo.com http://www.OneProDesign.com Last edited by Ryan Szulczewski; November 28th, 2007 at 08:27 PM. Reason: more info added

November 29th, 2007, 06:45 AM	#4
Ryan Szulczewski Regular Crew Join Date: Jul 2007 Location: Pensacola, FL Posts: 60	I guess I should just do a comparison. It just seems like the 8bit uncompressed files are ginormous with no quality benefit from the HDV footage. When we capture we aren't upscaling we are downscaling. So we go from HDV to SD through the Kona Card. __________________ http://www.OneProVideo.com http://www.OneProDesign.com

November 30th, 2007, 09:52 AM	#6
Ryan Szulczewski Regular Crew Join Date: Jul 2007 Location: Pensacola, FL Posts: 60	So I talked to the other editor about our workflow and why we do it this way. We do a lot of After Effects work and basic VFX stuff. So if you capture and edit on an HDV timeline in Final Cut Pro if FCP has to render anything it will be lossy (at least that is what he understands). Plus if we do any After Effects work on the footage and rerender to HDV that is lossy as well. I haven't tested it yet but I am going to do some tests to see how much the picture degrades working through HDV because those 8-bit uncompressed files are huge. I captured the same clip (about 10 seconds in length) and the Uncompressed was 222MB while the HDV clip was 44MB. That is no joke. __________________ http://www.OneProVideo.com http://www.OneProDesign.com

November 30th, 2007, 10:53 AM	#7
Ryan Szulczewski Regular Crew Join Date: Jul 2007 Location: Pensacola, FL Posts: 60	OK so I took some HDV footage and messed around with it in AE and rerendered it out using various codecs (HDV, Animation, Konas 8-bit) and no matter what I do it comes out flat. Its like After Effects is doing something weird to the Gamma curve of the footage. I can use Synthetic Aperture to offset it but it usually introduces a lot of noise. Any thoughts on why the HDV footage comes in looking very flat and muddy in After Effects? __________________ http://www.OneProVideo.com http://www.OneProDesign.com

November 28th, 2007, 08:49 PM	#2
Carl Middleton Major Player Join Date: Sep 2007 Location: Fernandina Beach, FL Posts: 562	Unless you are using the hardware of the camera to perform a good job of upscaling, and saving directly into an intermediate format that you would like to use, there's really no advantage. The data on the tape is 1440x1080 (I am assuming? Not familiar with the HD100) 8 bit color. The firewire captures that exact digital information that is on the HDV tape, an m2t file. That is all the camera would have to work with, though if it is automatically converting it in realtime to an intermediate codec (i.e. Cineform, ProRes, AIC) that will help your editing process, such as 10bit+ color, I frame, etc it will hold that advantage. As well, it's very possible that the results of using the camera to upsample with the component out might be quite nice. Carl

November 28th, 2007, 09:59 PM	#3
Steve Oakley Trustee Join Date: Apr 2006 Location: United States Posts: 1,158	well if you are capturing to DVCpro100, you're losing 25% of your horizontal resolution, plus you are adding another compression hit. not a good gain. if you are capturing using prores, thats good. uncompressed is good too, but uses disc space at a horrendous rate. if you can capture native via FW, thats still the best rate. very small files, no loss coming in as you are just copying the bits over. problem is FCP is just plain flaky with capturing it. DVHS on mac works much better, actually just about anything works better then FCP for HDV capture.

November 29th, 2007, 08:23 AM	#5
Carl Middleton Major Player Join Date: Sep 2007 Location: Fernandina Beach, FL Posts: 562	If that is the case, I would take and use the camera to convert HD->SD via firewire. I would then try the SD downconvert using the Kona card as you've been doing. Compare! :) It's two methods that many people use for the same end result, whichever gives you the best quality and simplifies your workflow (probably firewire, imo) is the right choice for you. Carl