-
HD possibilities on Final Cut Pro
-
Gary,
I think I qualified most of the arguements you plucked from my post in the post itself. The Panny codecs are fast becoming the workflow of choice because of the benifits they present in post production, an HD 4:2:2 worflow that does not destroy image quality with one render in FCP.
I can see Sean’s multigenerational arguement, it makes some sense. What I really would like to see is an HDCAM master played into whatever MPEG encoder the network uses compared to a DVCPRO HD master of the same show through the same encoder. It think that is where it gets a bit interesting.
The fact that you can not use HDCAM natively (although HDXDCAM can be) is another sticking point. How would the HDCAM look if it was edited and rendered natively (which you can’t even do). Most HDCAM will get further compressed when it is captured into ANY edit system.
Anyhow, not sure why you are picking on me. I’m still trying to figure out where this bit of animosity comes from. 🙂
David
Peace and Love 🙂
-
[walter biscardi] “Because I consider it a waste to capture DVCPro HD at anything higher than it’s native format. I can’t “add quality” to what has been recorded on tape, that’s just not possible.”
Nobody’s arguing that. As long as you capture FW, you’re right – you can’t add to the quality. I was saying that footage which you’ve applied effects, filters and CC must be rendered on an UC timeline ONLY IF you MUST have the absolute cleanest video image possible (which is overkill for many). Do this and there is no benefit to capturing UC vs. FW native.
[walter biscardi] “I just feel I have to step in here for a moment. To say “if someone can’t afford the disk requirements for uncompressed finishing workflow then I would hardly worry about it” does not apply in our case.”
Actually it could. Maybe saying “Can’t afford” was a bad way to phrase it. Rendering to uncompressed means more time rendering, more disk space, and requires space on a fast array when it isn’t if you do all DVCProHD. It also means you have to go back and resize all graphics to reflect 1280×720 vs. the DVCProHD ratio of 970×720. This means more time which means more money- either out of your pocket or the clients. That’s what I meant. So it makes sense for many people to do what you do. I do it too. But I don’t fool myself into thinking it’s the highest quality possible. I’m comfortable knowing that compared to the (still common) DV25, BetaSP, and HDV projects many people produce for similar costs, a DVCProHD production is still many, many, many times better whether you re-compress it twice or not.
[walter biscardi] “None of us could see the difference on the screen and to me that’s all that matters.”
Well I’d argue that generally things happen to the video after it leaves your shop which could be adversely affected unbeknown to you. But I don’t want to go there. As I said I do it too for some projects. We don’t own a D5 deck, so it didn’t make sense to rent one for some projects I finished which were shot on Varicam. Instead I just mastered it back to DVCProHD tape and delivered it that way. Rendering to uncompressed and mastering to D5 would have been overkill considering the budget and nature of the projects.
I am not trying to debate over how good the video looks with your workflow. I’ve done it and it looks great. Would I still do it if I were bumping up a feature to 35mm? Not in a million years. But that’s just me. Nobody’s writing a rule book so it’s fine for people to have their own opinions on the matter.
What I am saying that the workflow is not “The Absolute Best Quality”. It just isn’t. Sorry. But that’s fine. It doesn’t have to be. Many men would kill to date a woman who’s just a “9 1/2”. But it’s important that people in this field understand the concept of generational loss via multiple compressions, and how certain workflows cause the video to go through subsequent and sometimes unnecessary lossy generations. Usually it’s fine to be guilty of this. Just as long as you understand the process to be better equipped to make those decisions. I’ve heard of people who instead of using FW capture, they capture DVCProHD tapes over SDI using a DVCProHD Easy Setup – thinking they are capturing the native format of the tape (which of course they are not). These kinds of mistakes are totally unnecessary and avoidable.
Sean
-
[Sean ONeil] “I was saying that footage which you’ve applied effects, filters and CC must be rendered on an UC timeline ONLY IF you MUST have the absolute cleanest video image possible (which is overkill for many). Do this and there is no benefit to capturing UC vs. FW native.”
Have you actually done this test Sean? We’ve done this test many many times, especially when we were first establishing the DVCPro HD workflow. I’m not arguing that there is some loss in this method, Graeme and Marco can show that with all the measurements and scopes. But have you done a visual test of two sections of timeline, say 5 minutes, one rendered in a UC timeline and one rendered in a DVCPro HD timeline and simply sat back and watched them on your pro monitor? My guess is you won’t be able to tell which is which. My clients can’t, the engineers at several local Post houses can’t, other editors that have visited my shop can’t.
So when you say the “absolute cleanest video” you’re obviously referring to zooming in on the scopes to show everybody in the room, “see, here’s where the loss comes in.” But without the benefit of the scopes, I’m curious if you can simply “see” the difference just viewing the footage normally. In fact do a split screen between the two. We’ve done that here too.
The only problem I have with people saying “ONLY if you must have the cleanest video possible” it sounds like there’s an obvious loss to the viewer when you see a DVCPro HD timeline. There just isn’t in all our testing and with the myriad of people who have viewed the footage. I’m not sure what Panasonic did when they came up with the DVCPro HD codec, but it’s remarkably clean.
[Sean ONeil] “Actually it could. Maybe saying “Can’t afford” was a bad way to phrase it. Rendering to uncompressed means more time rendering, more disk space, and requires space on a fast array when it isn’t if you do all DVCProHD. It also means you have to go back and resize all graphics to reflect 1280×720 vs. the DVCProHD ratio of 970×720.”
Quite honestly, and maybe because we’ve always had some of the fastest machines available, rendering uncompressed is not all that different from rendering DVCPro HD. I believe an entire 22 minute episode renders in about 35 minutes on the Mac Pro Quad vs. about 45 minutes for uncompressed. It’s really minimal as the machines improve.
As for the graphics, we create everything natively 1280 or 1920 first and then create a copy that’s resized for DVCPro HD. This way we’re always covered in case we need to make a change to the workflow down the line. We’ve been doing this workflow over two years now and we’ve always “protected” for 1280, or 1920, whichever the case may be. Again, minimal time really, probably a total of 10 minutes over the course of working on an episode.
[Sean ONeil] “I’ve heard of people who instead of using FW capture, they capture DVCProHD tapes over SDI using a DVCProHD Easy Setup – thinking they are capturing the native format of the tape (which of course they are not). These kinds of mistakes are totally unnecessary and avoidable.”
Actually with the Kona boards, you are capturing the same as if you were coming in via Firewire however AJA says the image quality is slightly better because of how they treat the codec on the way in. Don’t know all the technical engineering of this, but we’re actually going to switch to the SDI input this week as that allows us to feed our 1200A to three different suites via SDI. So in the case of the AJA Easy Setups with the Kona boards, yes you are capturing the same native format as if you were capturing via FW.
[Sean ONeil] “Rendering to uncompressed and mastering to D5 would have been overkill considering the budget and nature of the projects.”
This is a point I just don’t understand. What does budget have to do with uncompressed? The only additional budget on my end is to rent either the HDCAM or D5 deck. I simply pass the cost along to the client if this happens. Obviously if they don’t want to rent either (which is relatively inexpensive in Atlanta) then that makes for an easy decision, but generally we’ll go with the DVCPro HD workflow because it works so well.
If your shop already has the storage available, then why does budget influence whether you go uncompressed or not? More money means you’ll allow the client to do uncompressed HD? If you have to add additional storage to facilitate uncompressed HD then you’re going to charge the client for this additional gear? Makes no sense in my mind. If you need to add some storage, you take a loss against the current job to purchase it and then you’re set for the next job. This is how we’ve built our business going on 5 years now.
If I need something for a client to complete their project, I purchase it knowing that I will need it again for another project. Especially when it comes to storage. But I never charge the client “more” for the “privilege” of working in uncompressed HD vs. DVCPro HD. They’re both high definition hence the same price.
What uncompressed HD means is that I need to make sure I manage the storage requirements better for that job to ensure that we have enough for all the current projects.
Sean, I hope I’m not being harsh here because you’re one of the great contributors to this forum.
Walter Biscardi, Jr.
https://www.biscardicreative.com
HD Editorial & Animation for Food Network’s “Good Eats”
HD Editorial for “Assignment Earth”“I reject your reality and substitute my own!” – Adam Savage, Mythbusters
-
Thank you all for your input…by initial post seems to have created some buzz. I now have another question. If I choose to go the F900 route and don’t have a problem with the traditional offline/online scenario…what sort of problems to I run into with timecode in online? For instance, if my workflow begins with shooting 24p 1080 with the F900, I downcovert (maybe not the right word) to the DVCPro HD Codec to work in FinalCut using a decklink extreme card (for example) how do I be sure of timecode accuracy in online? I’m just not sure I understand how the dropped frames work between the 2 formats. Can I online in a suite other than FinalCut Pro?
Thanks…you guys are very very helpful with your thorough responses. What a great forum for figuring these things out!!!
Laurie
-
[Laurie Case] “If I choose to go the F900 route and don’t have a problem with the traditional offline/online scenario…what sort of problems to I run into with timecode in online? For instance, if my workflow begins with shooting 24p 1080 with the F900, I downcovert (maybe not the right word) to the DVCPro HD Codec to work in FinalCut using a decklink extreme card (for example) how do I be sure of timecode accuracy in online?”
OK
capturing to DVCProHD should be frame accurate to your original tapes. 24p On HDcam is 23.98 frames- no frames should be dropped when you are capturing and there should not be any pulldown added to your video when working with files in this manner.[Laurie Case] “I’m just not sure I understand how the dropped frames work between the 2 formats. Can I online in a suite other than FinalCut Pro?”
there should be no difference in the Timecode.- Yes you can online in a different suite- EDL’s and XML’s are supported for communication with other platforms and NLE’s
gary adcock
Studio37
HD & Film Consultation
Post and Production Workflows -
[walter biscardi] “Have you actually done this test Sean? We’ve done this test many many times, especially when we were first establishing the DVCPro HD workflow. I’m not arguing that there is some loss in this method”
Remember, I have done the EXACT workflow as you on several occasions. I found it totally acceptable considering the circumstances of those particular projects. My only point is that it is not “the best” and what you see is not always what you get by the time it reaches the viewer. The “zooming in with scopes” standard isn’t always trivial. My monitor is 19″. If the show was going on a 30 ft projection screen, zooming in might not be such a bad idea. That’s just how I feel about it and there’s nothing wrong with those who disagree. Both ways provide an incredible picture that is many times better than HDV and SD no matter if you recompress it once, twice, or three times. And if you do well-lit television shows, then the effects of compression are far less damaging. The only times I’ve worked with DVCProHD is for live music concert projects. These are not brightly lit and are much, much more susceptible to compression artifacts verses a well-lit TV studio set like your show Good Eats.
[walter biscardi] “Actually with the Kona boards, you are capturing the same as if you were coming in via Firewire however AJA says the image quality is slightly better because of how they treat the codec on the way in. Don’t know all the technical engineering of this, but we’re actually going to switch to the SDI input this week as that allows us to feed our 1200A to three different suites via SDI.”
I hate to rehash old arguments, but I think they’re full of it. I’m sorry. I’m sure it still looks very, very good. And I understand that wiring each room for FW can be impractical (although it can be done with Gefen extenders). SDI-only wiring provides great benefits as it makes everything easy to patch and distribute to multiple destinations. So go for it. I can totally understand why it’s worth doing. But it is not native and it certainly isn’t “better” in the sense that detail which never existed on the tape is somehow added. The only thing added is a 2nd round of compression which is a lossy hit (albeit not visual at first). But not only that, the process itself actually upscales the image from 970×720 to 1280×720 as it is converted to an uncompressed SDI signal inside the deck. Then on the other end it downscales it back to 970×720 as the Kona converts it back to DVCProHD – causing real loss.
Take this analogy. You have 1 cup of vodka. Add two teaspoons of water to it. Mix it up. Now remove 2 teaspoons of liquid from the glass. You still have 1 cup of vodka left over. But it’s slightly diluted. It’s not like you have a magic teaspoon that only removes water from the mixture and leaves the vodka in the glass. Even the greatest vodka tasters in the world might not be able to pass the Pepsi challenge. But it doesn’t change the fact that the original glass was more pure than the one that had water mixed in.
If AJA has some special magic trick, I’d love to read about it. But all I’ve heard are ambiguous and anecdotal statements. Anyone, please feel free to prove me wrong.
I do understand why AJA would claim this. They figure it still looks great (and I’m sure it does) so why complicate things and tell “simple-minded Mac editors” that the Kona cards are not the ideal tool to capture DVCProHD footage. If we did that, some DVCProHD exclusive shops may feel they don’t need to buy one (which they don’t except for monitoring). I hope I’m wrong, but I haven’t seen any evidence to suggest otherwise. Not on the AJA website, not on the Cow, not on any FCP user groups, boards, blogs, Google search – nothing even mentions the topic let alone explains it.
Again, I’m not trying to criticize your workflow or anything like that. Believe me. I’m sure you provide better looking video to your clients than most shops would. I just think it’s important people understand these concepts so they are better equipped to make their own decisions.
Sean
Reply to this Discussion! Login or Sign Up