Final Cut Studio 2 Announced!

[Andy Mees]

It is a RT transcode with existing footage (software-based) and an instantaneous conversion with a HD-SDI/HDMI card (h/w based). It couldn't be faster. You will also be able to use it with the gamut of BMD and AJA products.

George, I appreciate that of course, I'm guessing my point is being lost. What I'm trying to argue, as it were, is that this still doesn't address the fact that we are taking a faster than realtime workflow and reducing it to realtime or slower.

I'm perfectly happy to admit that I'm looking at things from my own "niche specific" world view, and not one typical of general post production workflows. Nonetheless it is still a valid view (or so I like to think!)

To better explain, I'm considering things specifically from an ENG and EFP perspective. Logging and importing directly from the camera media. HD-SDI/HDMI etc never enters the equation, as that would only serve to restrict the workflow to good old-fashioned realtime ingest, losing the obvious benefit of the faster than realtime tapeless workflow, as does having to transcode to an intermediate codec after the fact.

Apple are actively courting major networks, promoting their hardware and software as ideal for digital news production ... and not without results. FCP is usd for field production by many of the major global news channels including the likes of CNN, SkyNews and Aljazeera.

So I guess the message "use Prores for all things post" as valid as it may be for a post house workflow, is not really one I personally want to hear from Apple ... I've got my ears peeled back for a "Quicktime natively supports all modern professional camera codec's and wrappers" message.

Here's hoping. Thanks for the debate.

[Chuck Spaulding]

Originally Posted by Andy Mees

George, I appreciate that of course, I'm guessing my point is being lost. What I'm trying to argue, as it were, is that this still doesn't address the fact that we are taking a faster than realtime workflow and reducing it to realtime or slower.

I'm perfectly happy to admit that I'm looking at things from my own "niche specific" world view, and not one typical of general post production workflows. Nonetheless it is still a valid view (or so I like to think!)

To better explain, I'm considering things specifically from an ENG and EFP perspective. Logging and importing directly from the camera media. HD-SDI/HDMI etc never enters the equation, as that would only serve to restrict the workflow to good old-fashioned realtime ingest, losing the obvious benefit of the faster than realtime tapeless workflow, as does having to transcode to an intermediate codec after the fact.

Apple are actively courting major networks, promoting their hardware and software as ideal for digital news production ... and not without results. FCP is usd for field production by many of the major global news channels including the likes of CNN, SkyNews and Aljazeera.

So I guess the message "use Prores for all things post" as valid as it may be for a post house workflow, is not really one I personally want to hear from Apple ... I've got my ears peeled back for a "Quicktime natively supports all modern professional camera codec's and wrappers" message.

Here's hoping. Thanks for the debate.

That works in an standard definition world, but as the bar is raised to HD and beyond and the quality of the display devices improves, the camera codecs are not good enough. There's no reason to constrain the quality in post production to the bandwidth and density of video tape or what you can transfer over FW.

What's probably going to happen is the migration of these post production codecs to cameras that are tapeless. The challenge here is that codecs like ProRes and Cineform require quite a bit of processing power to encode in RT and even with the Sheer Video codec which doesn't require as much CPU power, its bandwidth exceeds the pipes coming out of the cameras. But all of that is changing quickly.

[Andy Mees]

That works in an standard definition world, but as the bar is raised to HD and beyond and the quality of the display devices improves, the camera codecs are not good enough. There's no reason to constrain the quality in post production to the bandwidth and density of video tape or what you can transfer over FW.

Clearly here we're are talking about "good enough" in the post-production sense, as I think we all agree that the quality of ones footage can never be better than the camera-native original.

For what it's worth, I work in a 100% end to end HD environment, shooting, editing and delivering in 1920x1080 HD. Regardless of that working resolution, it is a tapeless ENG/EFP environment, and recompressing to ProRes, or any other intermediate codec for that matter, negates the primary benefit of the faster than realtime tapeless workflow.

I think this is the sticking point. Relatively speaking, not many are yet used to or have much experience of a truly "faster than" realtime workflow. This is still bleeding edge stuff, and to some extent it seems the bleeding is due to the nature of the camera native codecs!

What's probably going to happen is the migration of these post production codecs to cameras that are tapeless

I don't know that we'll see it in camera for a while yet, but perhaps so ... in the interim I'd expect to see these codecs being licensed to and incorporated within the many HDD field recorder companies out there.

[Chuck Spaulding]

It all depends on how you define better. A lot of footage, if not most, looks better after color correction, effects compositing, and so on. The problem is that by using the camera native codec there isn't enough headroom and after stepping on the image in post its difficult to get even the same level of quality coming out as that went in.

Where speed is important, like ENG news gathering for example, sure the camera native codecs are "good-enough." But in long form production, documentaries or commercials for example it might not be.

If everyone in my neighborhood was using camera native codecs and I was able to compete without changing my workflow then they're good enough. But if a company down the street starts using an intermediate codec and produces better quality images, then I'm at a competitive disadvantage. Thats' where the idea of an intermediate codec will take off, it at least gives us a choice we can stay with the camera natie codec or we can "re-wire it, arr, arr, arr, arrrr."

Of coarse if you have the budget and the bandwidth for uncompressed then it doesn't get much better than that.

[Luis Rolo]

Originally Posted by Boyd Ostroff

Luis, I'm like you... wondering just how well it would work on my dual G5 though. Sooner or later I'll want to upgrade to an intel machine anyway, and software like this really shows off the new hardware.

Yep, i know. But i can't upgrade to Intel yet, so, i'll be a bit disappointed if i can't run Color on my machine, and all because we don't have enough GPU upgrade options. In my country the X800 XT costs 550 €, and i won't pay that kind of money for an obsolete graphic card.
If i can't use FCS2 in my machine I'm thinking about going PC world and use Premiere CS3. But it won't be an easy decision.

[Andy Mees]

Thanks Chuck (and all)
Your perspective and experience are much appreciated

Cheers
Andy