24pA offline/online workflow

[John]

[Dave LaRonde] “Hmmm….. What on earth would an FCP offline capture preset be for DV? YouTube-quality video?”

No I am talking about a DVX100. I guess I don’t understand the confusion. I just want to edit in 4:1:1 DV codec to save HD space then recapture the final material in 4:2:2 uncompressed. Isn’t that a pretty standard ‘offline/online’ workflow for uncompressed SD? Or do I have a misperception there?

And so my question has to do with what are people doing to be able to recapture 24pA footage in this workflow? Or do people just avoid it all together and simply capture and edit in uncompressed 4:2:2 from the start?

John Christensen
cdesign@airmail.net

[Barry Green]

[John] “Or do I have a misperception there?”

The misperception would be in thinking that you’d gain anything by doing it that way. When the DVX records, it records to DV format (which means 4:1:1 color sampling, and 5:1 DV compression). Any advantage of live uncompressed 4:2:2 is GONE when you record. So yes you could recapture, but you’d be capturing compressed 4:1:1 footage into an uncompressed timeline. Which is silly, because you already have that same data on your hard disk. You can’t get better than DV footage out of a DV recording.

You can choose to edit your DV footage in an uncompressed 4:2:2 10-bit timeline if you want, but there’s no reason to go recapturing footage, you could just convert the footage you already have on your hard disk.

—————–
Get the most from your DVX camera. The DVX Book and DVX DVD are now available on ebay and at Amazon (https://www.fiftv.com/db)

[Noah Kadner]

Totally- gains you nothing. I have a section on my DVD that shows in a graphical way why this is so.

-Noah

[John]

[Barry Green] “The misperception would be in thinking that you’d gain anything by doing it that way. “

I completely understand that once you record a DV signal to tape, you have already lost information in the 4:1:1 sampling and the compression. And.. I realize that recapturing the footage to 4:2:2 uncompressed can not possibly restore the lost info. But.. am I not going to get much better color correction and graphics compositing (i.e. titles) if I am working in a 4:2:2 uncompressed timeline?

Or are you simply saying that the recapture is unnecessary because I could just throw the original DV footage into a 4:2:2 uncompressed timeline and render it out?

Also.. what about with DVCPro HD material recorded to tape? If I don’t have boundless HD space, isn’t it a normal thing to capture your tapes in a low-res format (like DV), edit as ‘offline’, and then recapture the same tapes at the full-res DVCProHD 4:2:2 (or even uncompressed) to do color correction?

John Christensen
cdesign@airmail.net

[Noah Kadner]

Or are you simply saying that the recapture is unnecessary because I could just throw the original DV footage into a 4:2:2 uncompressed timeline and render it out?

-Exactly- if you want to consider your DV timeline the offline all you need to do is export to 10-bit uncompressed when you’re totally done. Of course keep in mind that if you’re going back to DV tape it’s going to be recompressed again anyway…

Noah

[Barry Green]

[John] “Or are you simply saying that the recapture is unnecessary because I could just throw the original DV footage into a 4:2:2 uncompressed timeline and render it out?”

Exactly. Working in 10-bit is superior for color correction, etc., so your instincts are good. But keep in mind the limitation of the source footage, and that it just isn’t going to get any better if you re-capture it into uncompressed 10-bit… what you have on your hard drive already is as good as it’s ever going to get, so you can skip the whole recapture process and just re-render what you’ve got.

[John] “If I don’t have boundless HD space, isn’t it a normal thing to capture your tapes in a low-res format (like DV), edit as ‘offline’, and then recapture the same tapes at the full-res DVCProHD 4:2:2 (or even uncompressed) to do color correction?”

Not really. DVCPRO-HD doesn’t take up that much space; in the worst case it takes up 4x as much space as DV. Which means that it’s still cheaper to put DVCPRO-HD on a hard disk than it is to put it on tape! I bought an external SATA 500GB hard disk for $279 at Best Buy a while back; that’s about 10 hours of 1080 footage, or 25 hours of 720/24pN. Why go through the hassles of offline/offline/conforming/etc when you could just work with the original native footage and be done the first time?

Offline/Online was far more important when drives were slow and expensive; now drives are fast and cheap, and the whole offline/online workflow becomes less relevant. Most HVX shooters are cutting their HD originals and not bothering at all with offline/online.

—————–
Get the most from your DVX camera. The DVX Book and DVX DVD are now available on ebay and at Amazon (https://www.fiftv.com/db)

[John]

[Barry Green] “if you re-capture it into uncompressed 10-bit… what you have on your hard drive already is as good as it’s ever going to get, so you can skip the whole recapture process and just re-render what you’ve got.”

Noah and Barry… this has been very helpful. Thanks for clearing things up for me. I do have one last question though.. (maybe 2)

Does either of these two workflows have an advantage over the other:

1) capture DV25 material over Firewire; place footage into a DVCPro50 4:2:2 timeline (or 10-bit uncompressed); render.

or

2) capture DV25 material through component to a AJA/Blackmagic card directly as DVCPro50 4:2:2 (or 10bit uncompressed).

Presumably there is some reason people pay extra for decks with SDI/component out and AJA/Blackmagic cards. If it isn’t to capture at a higher res than was recorded to tape what is it? And again is there absolutely no difference in quality between the two methods?

Also.. in method 1) above, am I correct in assuming that to convert from DV25 to something higher res (most likely DVCPro50) that I would simply throw the material in a timeline and render? Or is it more practical to convert the footage using the Export command? Can Compressor do this as well in a batch mode?

Thanks much. John

[Barry Green]

When you capture DV material through SDI, you’re not going to get more information than was on the tape in the first place. But, sometimes those capture cards will perform a chroma smoothing operation, or otherwise attempt to interpolate 4:1:1 into 4:2:2. Doesn’t hurt, and can make the footage look better. You can get a similar benefit by using your NLE’s chroma smoothing filter, or something like Nattress’ chroma smoother.

Working in the 10-bit realm definitely has advantages over working in 8-bit. But the point is, your footage is 8-bit. If you convert it to 10-bit via one of those cards, or by rendering in software, either way you should end up with the same result. The main benefit to those cards is for working with footage that’s other than DV; you can use ’em with BetaSP or Digital Betacam, etc.

—————–
Get the most from your DVX camera. The DVX Book and DVX DVD are now available on ebay and at Amazon (https://www.fiftv.com/db)

[Noah Kadner]

If you have a black and white photo, does scanning at 10bit color make it become color? Of course not. You cannot by laws of physics undo DV compression. Any format you capture is going to give you equal or lesser quality but why bother? Capture via Firewire and you get exactly what you shot, nothing more or less.

-Noah

[Fgfalex]

Hey everyone.,

I am currently doing post production on a project that was shot on a DVX100 at 24P and we’ve run into a MAJOR problem.

While doing the tape to tape color correction we discovered boxy pixels — it actually look as though it were shot through a screen door — in some of the shots. When I went back to the original Uncompressed Quicktime I had delivered to the post house, I saw the boxy pixels there. When I went into the project and into the Quicktimes FCP uses there were there too! When I went back to the tapes, they were not there.

Does anyone know what is going on? More importantly, how should I proceed now? Do I need to redigitize the footage into an Avid? If I do the online in FCP what settings should I use?

Any help would be greatly appreciated.
Thanks,
Alex