24p Capture comes to FCP 6.02

[Nathan Pawluck]

Here are the new capture specs for FCP 6.02

Sony HVR-V1 HDV Tape-Based Camcorder Support
Final Cut Pro 6.0.2 is compatible with the Sony HVR-V1 HDV camcorder, which is capable of recording 1080p24, 1080p25, and 1080p30 footage. You can capture natively or capture to either the Apple Intermediate Codec or the Apple ProRes 422 codec. You can also output back to the Sony HVR-V1 camcorder using the Print to Video command.

To natively capture 1080p25 or 1080p30 footage, you should use the HDV 1080i50 and HDV 1080i60 Easy Setups, respectively. Your footage will retain its progressive scanning even though it will be stored in an interlaced format. You can capture 1080p24 footage using the 1080i60 Easy Setup, but your captured footage will retain 3:2 pull-down in this case.

For transcoded capture of 1080p24 footage, 3:2 pull-down is removed during transcoding, resulting in footage stored in the 1080p24 Apple Intermediate Codec format or the 1080p24 Apple ProRes 422 codec format. You can also capture 1080p25 and 1080p30 footage to either format, although Easy Setups are not included for these formats. In these cases, your captured footage is stored in the 1080p25 or 1080p30 Apple Intermediate Codec or Apple ProRes 422 codec format.

Here are the recommended workflows for capturing from and outputting to the Sony HVR-V1 camcorder with Final Cut Pro 6.0.2:

24p/60i on tape: Capture to the 24p Apple Intermediate Codec or Apple ProRes 422 codec, then output to the HVR-V1 camcorder in 24p/60i mode.
25p/50i on tape: Capture to the 25p Apple Intermediate Codec or Apple ProRes 422 codec, then output to the HVR-V1 camcorder in 25p/50i mode.
30p/60i on tape: Capture to the 30p Apple Intermediate Codec or Apple ProRes 422 codec, then output to the HVR-V1 camcorder in 30p/60i mode.

[Nathan Pawluck]

Here is the link to all the spec's on FCP 6.02 http://www.apple.com/support/release...ut_Pro_6.0_rn/

[Matt Devino]

Has anyone tried to capture to ProRes while removing pulldown? This would save me a ton of time, as I'm removing pulldown and transcoding to ProRes through compressor. Is there an easy setup now that is specifically for the V1? Thanks,
Matt

[Steve Mullen]

Originally Posted by Nathan Pawluck

Here is the link to all the spec's on FCP 6.02 http://www.apple.com/support/release...ut_Pro_6.0_rn/

Remember you must update to 10.4.11 BEFORE you try to download this update. I'm waiting a few days because the update to QT 7.2.3 broke JES.

"24p/60i on tape: Capture to the 24p Apple Intermediate Codec or Apple ProRes 422 codec, then output to the HVR-V1 camcorder in 24p/60i mode."

Is Apple saying there is now "24p/60i" Print-to-Tape that adds 2-3 pulldown?

[Benjamin Eckstein]

Will you ever be able to capture from the V1U with the 1080p24HDV preset like you can with Canon 24F footage?

[Steve Mullen]

Originally Posted by Benjamin Eckstein

Will you ever be able to capture from the V1U with the 1080p24HDV preset like you can with Canon 24F footage?

That's the point of the upgrade.

[Nathan Pawluck]

I love people who comment without reading.

[Matt Devino]

Hey Everyone,
So I got home an updated my home system to 6.0.2 on Leopard, ready to test out the new pulldown removal on capture.

The short answer is it works.

The long answer is it works but with some annoying things apple seems to have left out of it's release info.


Here it goes:

I hooked up the V1U via firewire and loaded a tape that was shot at 24pA. I first tried the Easy Setup HDV-1080P24, then HDV-1080P24 firewire basic. Both worked, and the pulldown was removed correctly. BUT when I jogged through my clips frame by frame there was a duplicate frame every 5 frames. I right clicked on my clips and they were still at 29.97! WTF!

Now, I checked the easy setup and my sequence setting was for HDV 23.98, so I took one of these clips and dropped it into a new sequence, FCP prompted me to change my sequence settings to match the clip, I said NO, and to my delight when I jogged through the clip in the timeline the extra frames were gone. It is realtime, needing no renders. So although it captures at 29.97 it's really 23.98 with duplicate frames added and by putting it into an HDV 23.98 sequence the extra frames are properly removed.

Weird way to do things, but it works.

Now, I'm not a fan of editing in HDV, so I was very excited to try out a conversion to ProResHQ while removing pulldown on capture. So I open up the Easy Setup HDV-Apple ProRes 1080P24. I opened up my log and capture window and... NO LOG AND CAPTURE WINDOW, WTF!!!! A window pops-up that says "name HDV clip" or something like that. So I go along with FCP and name a test clip, and hit OK. It goes right to the capture screen, and starts my camera rolling. It begins capturing footage off of my tape, with a little warning saying "preview is 10 seconds behind, view camera for playback". I let it roll for a few seconds, it goes through a clip break, and I hit ESC.

Now 2 clips pop up in my Bin, both with the pulldown removed and at 23.98, and it successfully created a new clip at the clip break. So at least it works, now I don't have to go to compressor to do ProRes/Pulldown conversions of my HDV footage. But seriously, why can't I add logging info? I can't even add a reel name, or have playback controls. I have to switch between easy setups or (gasp) use my on camera playback controls to cue up footage? That really really really sucks. It doesn't make any sense. It's like Apple brought some of the Windows code writers to add these new functions to FCP.

Anyway, I guess I'll have to deal, it's better than recompressing everything I capture.


But one more thing, there is no Easy Setup to do the transcode to ProResHQ, only to standard ProRes. Weird right? SO I made a custom easy setup where the compressor was ProResHQ...and it worked exactly the same as the regular ProRes Easy Setup. So Apple seems to have left it out, but it works. The lag concerns me though, I'll need to try a longer capture (the test was about 30 seconds), I'm worried that maybe the lag happens because of the transcoding eating up processor power and memory, I could see a long capture (like a full tape) possibly causing the system to run out of memory or lag to the point where it couldn't keep up anymore.

Although I did do this test on a Dual 1.8 GHz G5 with 1.25GB of RAM over to a firewire800 drive. So guys with shiny new 8-cores and RAIDs should be fine. I'll have to take my V1 to work tomorrow and try it on my 8-Core there and see what happens :-).

[Leslie Wand]

somewhat different setup - v1p to winxp - however, i was under the impression that there was NO 'batch capture' under hdv since it's 15 gop would give inaccurate in/out points?

i hope someone will tell me there is, but from my experience with vegas 8, and other people i've talked to, tc capture is a none starter. bugger.

leslie

[Steve Mullen]

Originally Posted by Matt Devino

Now, I'm not a fan of editing in HDV, so I was very excited to try out a conversion to ProResHQ while removing pulldown on capture. So I open up the Easy Setup HDV-Apple ProRes 1080P24. I opened up my log and capture window and... NO LOG AND CAPTURE WINDOW, WTF!!!! A window pops-up that says "name HDV clip" or something like that. So I go along with FCP and name a test clip, and hit OK. It goes right to the capture screen, and starts my camera rolling. It begins capturing footage off of my tape, with a little warning saying "preview is 10 seconds behind, view camera for playback". I let it roll for a few seconds, it goes through a clip break, and I hit ESC.

Now 2 clips pop up in my Bin, both with the pulldown removed and at 23.98, and it successfully created a new clip at the clip break. So at least it works ...

There is no advantage to editing with ProRes 422. It's a waste of drive space and time.

What's happening is HDV is converted to ProRes 422 -- which is why you don't get logging. Another reason to stay with HDV.

[Matt Devino]

I think people will be interested to see this:

http://www.flickr.com/photo_zoom.gne...7000334&size=o

This is a frame grab comparing the footage I captured using on the fly conversion to ProResHQ to HDV footage I had on my system that was captured as HDV in 6.0.1

The top frame is HDV that was captured in FCP 6.0.1, and pulldown was removed and converted to ProResHQ in compressor (at the time the only good way to remove pulldown in final cut studio). Even before the conversion I would see the same artifacting (mostly aliasing) you can see in the top half of the image (the raw HDV footage in my FCP looked the same, I did a bunch of side by side comparisons, as I wasn't happy with the amount of compression at all, but I figured that's what happens when you put HD onto a miniDV tape).
Now when I capture straight to ProRes in 6.0.2 I get what you see in the bottom half of the image, an image that is almost completely free of aliasing (it's very apparent on the black trim around the windows of the car).

Needless to say I'm a lot happier with my V1U (or should I say Final Cut's ability to capture HDV).


As far as what Steve said above:

- "There is no advantage to editing with ProRes 422. It's a waste of drive space and time."

This isn't true, it's a 10-bit codec (much better looking renders) and is intra frame (faster renders) and it's 4:2:2 instead of 4:2:0 so any kind of color effects/correction you add will look MUCH better. Sure it's bigger file-wise, but you can get 6 hours of footage on a 500GB drive, so who's worried about disc space?

-"What's happening is HDV is converted to ProRes 422 -- which is why you don't get logging. Another reason to stay with HDV."

Ok, but why is there no deck control?

[Steve Mullen]

Originally Posted by Matt Devino

As far as what Steve said above:

- "There is no advantage to editing with ProRes 422. It's a waste of drive space and time."

This isn't true, it's a 10-bit codec (much better looking renders) and is intra frame (faster renders) and it's 4:2:2 instead of 4:2:0 so any kind of color effects/correction you add will look MUCH better.

-"What's happening is HDV is converted to ProRes 422 -- which is why you don't get logging. Another reason to stay with HDV."

Ok, but why is there no deck control?

You don't understand how FCP works. HDV is automatically converted, whenever a render is performed, to 4:4:4. If you are working with a 10-bit Sequence -- HDV is converted to 10-bit.

IT MAKES NO DIFFERENCE "WHEN" HDV IS CONVERTED TO 10-BIT 4:4:4 .

If you do it upon capture, you only waste disk space. There will be no difference in quality. None.

You have bought into one of the oldest anti-HDV myths! Shame on you. :)

The reason for no deck control is that there has never been deck control with AIC. All Apple has done is replace the AIC codec with the ProRes codec.

But, once you stop believing that you can use HDV, you'll have no need to use ProRes captures. Problem solved.

PS: the only time ProRes is needed is when Color renders. And, you can ask Color to do so.

[Matt Devino]

Originally Posted by Steve Mullen

You don't understand how FCP works. HDV is automatically converted, whenever a render is performed, to 4:4:4. If you are working with a 10-bit Sequence -- HDV is converted to 10-bit.

But Final Cut isn't an RGB system, it only work in YUV 4:2:2 unless you have Kona or Blackmagic codecs installed (and this is still completely un-native to how the guts of FCP work). Even then for it to ever end up as 4:4:4, or anything 4:2:2 for that matter, you would need to take your final edited sequence and put it in a sequence that uses a 4:2:2 or 4:4:4 codec and render your entire sequence. Seems like a lot of extra rendering when you can just convert everything to 4:2:2 in the first place right?

And Final Cut won't render all HDV footage to 4:4:4 automatically, it may render it into 10-bit 4:2:2 IF you set up you sequence's render option to use 10-bit YUV super white etc.

Anyway you're right there's nothing "wrong" with editing in HDV native, it just may take a few extra steps at the end of your edit and some extra render time to get your final piece looking as good as it can.

[Steve Mullen]

Originally Posted by Matt Devino

Seems like a lot of extra rendering when you can just convert everything to 4:2:2 in the first place right?

There is NO rendering as HDV is edited in real-time. Only 3 things can happen to any frame of HDV or ProRes:

1) Nothing

2) FCP uncompresses the frame to 4:2:2 either 8-bit and displays it.

3) FCP uncompresses the frame to 4:4:4 YUV either 8-bit or 10-bit (your choice) and uses the frame in some way. For example CC or FX. The 4:4:4 YUV 8-bit or 10-bit result is sent to a display. The result is NEVER saved or re-used! When you use the source frames again -- FCP always uncompresses ALL frames from scratch.

Note: even ProRes is "chroma up-sampled" from 4:2:2 YUV to 4:4:4 YUV before being used.

Obviously, it makes no difference if FCP decompresses HDV to 4:4:4 YUV (either 8-bit or 10-bit) during capture OR at the moment the HDV frame is needed. The only difference is how many bits you waste holding the HDV information on disk.

The actual source information is never other than 4:2:0 at 8-bits.

Your export is ONLY dependent on the codec you use when you export. All frames including the "nothing" frames are are converted at this time.

PS1: if you MANUALLY render -- it is only for you to see the frames play in RT. When you export -- FCP uncompresses ALL frames from scratch. The rendered HDV frames are ONLY used to drive your display.

PS2: Color renders are THE exception. They go back to the timeline as ProRes, but this is done automatically.

PS3" Even Avid has no need for anything but HDV source files. Avid does REUSE renders -- which is where the myth started -- but Avid is smart. All renders are automatically DNxHD. Just like Color.

[Matt Devino]

Ok, yes there is upsampling to 4:4:4 happening in order to display video on your computer's monitor, because computer monitors are RGB.

But this doesn't matter if you're trying to view your timeline on a broadcast monitor through SDI. Say I want to look at my timeline on a video scope through SDI. An HDV timeline would look slightly different than a ProRes timeline of the same video because of the different color space and 10-bit vs 8-bit.

Another thing to consider is graphics. Working in an HDV timeline, if you place a title into your timeline you will eventually need to render it, if not right away. When you render it the HDV codec is used, and let's face it HDV is pretty compressed, and you can get aliasing and quantizing errors notorious to 8-bit (even "uncompressed" 8-bit) and adding the amount of compression that HDV calls for along with 8-bit is just yucky. Now, if you were working in ProRes, you will not get this problem as ProRes is 10-bit (say goodbye to those quantizing errors) and it's much less compressed which will reduce aliasing. These errors will never be fixed simply by outputing to something that is 4:2:2 10-bit like a D5 tape, your sequence must be rendered at a 10-bit 4:2:2 codec before the output. Hence the reason to use ProRes in a situation like that. Fo an extreme example drop a title into a DV timeline, turn on 10-bit renders in your timeline settings, and render your title. Still see all that aliasing? Yup, it's because it was rendered using the DV codec, not uncompressed 4:4:4 or 4:2:2.


Anyway, I think anyone reading our convorsation is probably confused or annoyed by now. Everyone should just read this article by Philip Hodgetts: http://www.kenstone.net/fcp_homepage...ay_native.html

And I think after reading this article neither of us will be proven right or wrong, just maybe we're thinking of different workflows.

Most of the stuff I do will go to broadcast and never go back to HDV tape, it will go to D5 or HDCAM, and sometimes DigiBeta. This is why I like to convert to ProRes.

Thanks for the good geek debate, it was fun.