What seems clear to me, but not so clear to you at this point (it seems) ... is that you're very worried about a very tight level of proofing control on say PrPro yet feeding your delivery into a system that will throw it all over heck and gone. What is absolutely guaranteed ... after you upload that to YouTube, not one person on the planet will ever see every hue precisely as you have them in any NLE you might choose to use. How do I know that precisely? I have two good monitors here, both puck-calibrated, and they will NEVER show the same thing identically. Close, but not identical. Within PrPro, the same video player, whatever. I'm going to NAB, headed to the airport in a couple hours. I'll be spending a lot of time with colorists that use upwards of $30k gear at their editing suite, including all the out-board BlackMagic and AJA and Kona cards & boxes feeding out-board high-end dedicated scopes and their beautiful Flanders Scientific broad-cast monitors. If I mention this discussion, they'll just shake their heads. Yea, for the work they do, pulling secondaries to make sure that some company's logo and other 'trade-marked' colors are precise in the original is absolutely necessary ... and they do it righteously. But even among their gear, no two monitors will show precisely exactly what another one shows. Close ... it's something they routinely have to explain to clients, as so often they've got two-three computer monitors for working, the scopes thing, and a large monitor for clients to watch "over their shoulder" while working. Colorists talk routinely about wishing they had a way to make sure the client never EVER saw their working monitors and the client-monitor/screen together while working, as there are so often discussions about it ... and some clients are frightened that this screen is "right" but that one is "wrong". So they deliver those trade-marked colors as tightly according to precise number values as possible. And the commercial is broadcast ... but no two stations doing the broadcast will put out totally identical signals, there will be averaged variations ... and no two TV's that ever show that to the consumer will agree precisely on those oh-so-precise colors. Because the TV's are all over the flipping creation both for what they are capable of, and how they're set ... or not. BUT ... within that system, there are actual standards. And still it's more variable than seems reasonable. PrPro is used every day by thousands of broadcast professionals to put out highly technically prepped b-cast work, that is scanned for "standards" on submission to the studio and passes without issue. That's broadcast standards, something that can be set and technically evaluated. So ... PrPro is definitely a pro-level tool. Routinely used for pro-level b-cast deliverables. Accurately, reliably. There is NO such thing for YouTube submissions. Or for the internet in general. Period. It's not the same "field". People deliver H.264 files meeting b-cast standards from PrPro every day. And Lagarith, ProRes, and many other codecs as laid out by whoever they're delivering their content to. So claiming there's something "wrong" with PrPro's output isn't perhaps ... totally accurate. Might there be differences between PrPro's output that one can find with different codecs, especially viewed "in the wild" on different video players? Of course ... every player does slightly different things with the codecs it sees by any test I've ever made or seen. Partly because some codecs include specific things in them to "set" gamma and levels, and others don't. And partly because of the way the code for that player works. So I'm not disparaging your work in trying to puzzle this out, or understand it ... but the reason there haven't been other long-term users jumping in to either assist or suggest something else, is we've been through this. You're trying to do the most righteous output/delivery you can, and I understand that ... it's what I do also. But you're delivering into a swamp-land of no standards whatever ... seriously, which is what both YouTube and the Web and general-use video players are. And trying to relate what something is like after being dumped into the swamp with what it was like in a pristine state. You'd have better success in say creating b-cast standard stuff to a tv station, then watching that output on a b-cast standard fully calibrated expensive TV. But even then, you still would see variants at LEAST as great as what you show above. In a totally standards-controlled situation beginning to end. Which isn't at all what we do when we take something from PrPro and put it into a computer and play it on a general video player. Even on our own computer. Neil
... View more