The xha1 shoots 1440x1080 with a non-square pixel which, expanded to hdv output in Quicktime with square pixels, becomes 1920x1080. Just the way it's made..../ Battle Vaughan/miamiherald.com video team

what about the length of time its taking me to encode HDV 2.5 times the original length of the clip

It looks great

I used to capture as HDV, set my sequence as HDV with ProRes renders. Quality-wise it works well. However, exports take forever. Why? Because Final Cut is not smart enough to know that it already rendered clips to a non-GOP format when you apply effects. It has to rerender every single clip into a non-GOP format and rerender the effects. Exports usually take 5X the length of the sequence.

The method I've now settled on is capturing HDV to retain smaller file sizes for whole tapes. Then I set my sequence as ProRes422 1440x1080. This way when I render effects Final Cut knows that they are non-GOP and I can export a Quicktime Reference file in a matter of seconds. I can still insert HDV clips on the sequence and get pretty good playback quality without rendering. It's not full-res but it is good enough until I can do a render.

In this scenario, to what are you exporting? What are the settings?

Here's a brief article on HDV workflow: HDV Workflow Tips | ECG Editorial

Quick summary:

#1) If It’s All HDV, Always Edit Native
#2) If It’s A Mixture Of HD and HDV, Convert (to ProRes) ASAP
#3) If You Can, Use Two-System Audio
#4) Know The Limits Of The Format In Advance

I don't get that? The more complex the edit (color, resize, motion), the more issues you have working in HDV and the more reasons to get out of HDV, and get out soon.

Ah, but did you read the article? He specifically mentions to be sure to set renders to ProRes.

With regards to the whole "ProRes transcode quality drop": I have been experimenting with HDV for the first time myself (with thanks to fellow dvInfo poster Vito DeFilippo) and I have to say -- thus far -- the quality loss from going to ProRes seems to be very insignificant -- and I say that from comparing 1080i footage (HDV source and the ProRes conversion) on a large calibrated HD broadcast monitor.

Yet another article
 

This one is Editors Guild Magazine - Tech Tips "Taming the HDV Beast: Final Cut Pro 6 and ProRes 422 to the Rescue" by Larry Jordan. It covers much of what has been said here, but also has very useful technical data.

At some point I should think about organizing these into one post for a "stickie" ;-p

Yes, I'm sure that the Prores drop in quality is very minimal, I agree, I agree. And the only real drawback is file size which I am fine with HD space provided I do not work on several long form projects at the same time. But I may spring for more HDs like I buy socks and undies, just something that one does every six months or so.

No, I did not read the whole article. Rendering Prores rather than Prores capture I am not that well versed at. Native HDV blows imo, I've seen my movie fall apart. But I will read article, can't hurt thanks!

I have captured concert footage (lots of reflections and details) both native HDV and ProRes. There is (with my equipment at least) a small bit of very fine artifact gain in ProRes but it's not MPEG artifacts and that's the really important point. The artifacts don't increase perceptibly with renders and that's important as well.

Very interesting. Do you have any stills of the same frame in HDV and ProRes that you could post that would demonstrate? The footage I have been looking at is a talking head, which would be very different in terms of lighting, spectral highlights, etc.

HDV Audio when using tapeless DR-HD100
 

Mike (or anyone who knows),

Am I right in assuming that HDV files recorded from my JVC HD100U camera directly onto my DR-HD100 hard drive via firewire are subject to the same low-quality audio characteristics that you are referencing in point #3) when you recommend “If You Can, Use Two-System Audio” in the ECG Editorial article?

I understand that using something such as the Edirol R-4 Pro digital recorder would be highly preferable but I’m curious if there is anyway to take advantage of the DR-HD100’s tapeless recording options to capture better quality audio when shooting in HDV than if I was using mini-DV tapes. My guess is that the HDV format is inherently poorer at capturing audio in order to achieve such good compression and small file sizes, and that it doesn’t matter whether one uses a HD to directly capture HDV footage when filming.

Thanks for any illumination you can provide and the stickie sounds like a great idea.

Go Habs!!

I would have to contact the producer first as the project is set for commercial release in the next few months. I would say that talking head shots would work fine, any change in quality would be imperceptible.

The digital recording on the DR-HD100 is the encoded output of the firewire port of the camera, exactly what is going to tape. The audio is compressed more than what many professionals find acceptable for their work. Unfortunately there is no way to separate the audio once it is going thru the encoding process, the cameras are not built that way. You have to record the audio before it gets into the camera or use analog output of the camera. The best way is to record audio to a separate audio recorder (DAT, ZOOM recorder, EDIROL, laptop, whatever) but this means more wires and devices to worry about so it's a decision on your part. I find that HDV is usually good for interview voice recording. Music is not great. When possible, I have the audio recorded separately with an output of the digital recorder going to the camera for an editing reference track.

Large Project Workflow clarification, please
 

Hi friends,

I know a lot of this has been discussed, but please bear with me, any help or suggestions would be really appreciated. I have only recently thrown in the towel with PC based editing so Macs and FCP are all new to me and really want to set the project up right at the beginning!

I’m about to start a reasonably long edit, 90-120 mins. Most of the footage is HDV with bits and pieces of SD DV etc. I expect to capture at the least - 20 hours of footage (always hard to predict). This is after a paper edit as there is in excess of 160 of tapes...

From what I've read, I can:

1) capture/convert to ProRes (my only hesitation here is disc space, I currently have 3TB internal). I would consider upgrading the space, but in order to backup everything is then doubled, and the expense is probably beyond my finances...

2) Capture as HDV and set the render to ProRes.
This is appealing as is saves space and even in the event that every clip is altered, I still will have saved a vast amount of space, than capturing/converting everything to ProRes. In terms of this will I have problems getting the project into Color once the edit is locked off and rendered to ProRes?

3) Capture as HDV and edit in a ProRes sequence.

Thanks
Errol

If you have a new system, all will work fine. 2 is probably the easiest as it will play the sequence back, on your computer monitors, with having to engage real-time conversion (no low res playback for unrendered clips.) except for transitions and real-time filters. A unrendered ProRes sequence will play HDV back in real-time with various levels of degradation depending on the system you have.

By the way, if your drives have more than 1TB free, you should have enough room for 20 hours of ProRes capture. ProRes capture has less issues compared to direct HDV capture (unexpected clip breaks, missing footage around camera starts). Rendering is faster as well. Remember to take into account how much you intend to filter the footage. A 1 hour ProRes project will need around 50GB if the whole project requires rendering.

The DV footage can be captured as DV and rendered in the sequence.

Where to I change the render setting to ProRes?
 

This may sound really basic, but whereabouts do I change the 'render settings' to ProRes.
As I said, most of this is new to me, but I can't seem to locate anything that resembles what I’m looking for in the Audio/Video settings... Have looked around and tried to find out with no luck... Thanks again

Under Sequence Settings you'll find the render tab. Click on it and change the render codec to ProRes. You will have to re-render any effect you have done already but that's fine.

William/Jason,

Thanks so much for the fine and detailed explanation. I can truly say I have a better understanding of Prores and HDV.

My questions is what is your workflow for getting it to SD DVD (not getting enough requests for Blue Ray yet) in the best quality?
Do you send your sequence straight to Compressor out of FCP?
Do you export as QT using Prores then send it to Compressor? As explained in the Ken Stone article.
And what are your settings in both the export process of FCP and in Compressor?

Thanks,

Carlos

OK this is a little tricky and depends on how you set the sequence.

Going directly from the sequence into Compressor works great for frame-based codecs such as DV or ProRes as Compressor will compress each clip from it's native codec directly into MPEG2. The problem with HDV is that it's already a MPEG2 codec and for complicated reasons rendering MPEG to MPEG doesn't always make good results. Transcoding HDV to ProRes makes the re-encoding to MPEG less problematic. That's why it's suggested to export HDV sequence edits to ProRes before making a DVD. In the past I have exported the HDV sequence to DV first and then made the DVD and that worked well enough for me at the time. A ProRes sequence can go directly to Compressor.

William,

OK, let me make sure I got you straight.

For reference, I shoot with a JVC HD200 mostly in 720p30, but at times 720p24 and 720p60

So if I capture in HDV - edit in the "HDV/ProRes" using ProRes only for render - I should export the sequence using QT conversion and select ProRes for the compression type. Then use that file for compressor.

If I use the "Pure ProRes" capture, edit, and render - I could send that sequence straight to compressor.

If I capture in HDV - Edit in a ProRes sequence and render in ProRes - Could that go straight to compressor or should one use QT conversion?

And if I may (bit off the subject) I am getting more and more requests for the web. Usually 3 to 5 minute clips (i.e. highlights of a ground breaking or fundraiser) that the client wants to put on their website. What would you recommend as the best workflow to get it into compressor and what setting in compressor (from the web folder I'm guessing?) would you use? Or is there a better way to deliver the file to the client?

Thanks again,

Carlos

You've got it right. And HDV files in a ProRes sequence should be rendered first and then you should make a QuickTime reference file and send that to Compressor.

Web files are a different story. Find out what your client needs, QuickTime, Windows Media, Flash and then work from there. Compressor gives you a set of h.264 presets in the Web/QuickTime 7 folder that cover a lot of ground and you can adjust them to your specs. Windows Media and Flash require additional software, DV Kitchen is good for this at a reasonable price. Some clients have a web person who can format the files to their standard so all you have to give them is a quality file that they can use.

Where does AIC fit into this picture?
 

I've never paid much attention to Pro res up until now as I've only recently jumped from FCP 5 to 7. I've got an all but completed project that I captured in FCP 5 using AIC, just doing some final tweaks. Made the jump to FCP7/SL without any catastrophes. I'm still using the AIC sequence/render setting I started with and have made a couple of SD DVDs using the QT conversion to Prores to do that. For the most part I'm happy but there are a couple of shots that could be better, shots where there was some heavy colour correction, (shot under an orange tent and no time to manual white balance.)

I'm going to change my render settings today and see if that helps but I've also been thinking about recapturing the whole thing as either HDV or Prores. Obviously my concern is that my current workflow means one extra recompression from HDV to AIC before the one to Prores for the SD DVD. How beneficial would avoiding that first recompression be?

I'm assuming that if I keep my current workflow and only change render setting to Prores all I have to do is delete the original render files first or is there more to it than that?

Thanks

AIC is a frame-based codec one generation before ProRes. You don't need to convert it to ProRes before making a DVD. ProRes is a better codec all around and has gained widespread acceptance where AIC never did. However AIC is not bad just not as good. You might want to stay in AIC unless recapturing the footage as ProRes is easy enough. I'm not sure if ProRes is available as a render option in an AIC sequence.

Thanks for that William. For reasons that are beyond me, compressor 3.5.1 on 10.6.1 hated AIC. After 2 attempts where the time remaining indicator went up to 15 hrs plus I gave up and went back to the prores conversion which worked just fine. I've since run Disk Utility which gave me the following.

Warning: SUID file "System/Library/CoreServices/RemoteManagement/ARDAgent.app/Contents/MacOS/ARDAgent" has been modified and will not be repaired.

Warning: SUID file "System/Library/PrivateFrameworks/DesktopServicesPriv.framework/Versions/A/Resources/Locum" has been modified and will not be repaired.

Fixing permissions using Onyx gave me the same result. Googling these two warnings didn't help. The best I got was that some people got the same when installing 10.5 I've been on 10.6 for months now. According to the posts I found, some people had contacted Apple support who said it wasn't anything to be worried about. Thanks Apple.

At the same time as Compressor went on holiday with the AIC project, FCP has been dropping frames like crazy on the same, but not on an SD project on a different drive. Smaller files of course might explain that. But then it didn't have an issue with the Prores file. I haven't taken the issue any further just yet but I do have a SL compatible version of Disk Warrior on the way and plan on defragging my boot drive as well.

Any thoughts? Thanks.

Are your playing HD video files off the system drive? This may be the problem and you are suffering from extreme file fragmentation. OS 10 is a UNIX based operating system which makes and deletes a lot of tiny files during it's normal hour to hour operations. If you are using a laptop the combination of a fragmented drive and the slower drive speed of laptop drives will cause this dropped frame problem. Other uses of the computer, graphics/music/ document work will never show a problem as they don't require the data bandwidth requirements of HD video.

Transfer the video files to an external drive and see if the problem remains. You might also need to reinstall your system OS. Do a search here as Robert Lane has posted a good method to "self-heal" 10.6.

I'm on a MBP and all media is on external drives. The AIC project on a G Raid which never had trouble until now.

Anyway, I found Robert's self healing post so will give that a shot.

G-Raids are good but I've heard about breakdowns with them and a client of mine had one that completely bit the dust while capturing a live shoot thru FCP. I was there for that. The company replaced it.

Good to know that but still don't think that explains why Compressor handled Prores easily but not AIC, given they were both on the G Raid. Hope to get to this tomorrow and will report back once I do.