A Response To Jim Jannard and 1080P

[Meryem Ersoz]

Originally posted by Lee Fingersh



OK. One word: inevitability.

History always looks inevitable in hindsight, it smooths away the many bumps in the road. But there is always a convergence of many fortuitous events that allow technologies to gain traction. The dustbins of technology are loaded with great ideas which lacked traction. I don't think there is much that is inevitable about the trajectories of new technologies. A lot of things have to line up to enable success on such a massive scale.

I can think of many reasons why 1080 delivery has lacked the traction it has deserved as a good idea. Price point has been a big one. And one thing RED has shown a will to do is reduce price points, to enable users and consumers to feel like they are getting a screaming deal, relative to what is out there.

RED has also shown a will to put their best technology forward at every possible opportunity, and, if it shows flaws, to correct them.

These are huge leaps forward in the model of doing business in technology, which often mean even more than the technology itself -

In terms of technology, there are as good and even better companies out there, in terms of pure technology. But no one seems able to match the RED business model for pure will to revolutionize industries.

Although it is easy to look into the past and claim that innovations were somehow inevitable, it is very, very hard to see five minutes into the future of technology, let alone five years, because so many stars have to align to allow a company to compete on the scale which RED has already proven it can do.

In their favor, RED will have loads of cameras in the hands of users, and those users will want to deliver the best that the camera can offer. This, too, is a big advantage for RED - moreover, there will be other users of other, lesser camera systems who have been clamoring for better delivery ever since the Z1U and the other HDV baby steps hit the streets.

There is a pent-up demand for better delivery. Much as there was a pent-up demand for better camera systems.

For a really interesting window on the past ("past," snort, five years ago...!!!!) I think when Jim Jannard posted this thread back on DVInfo on September 14th, 2005, he was speaking to a large crowd nodding their heads in agreement. He was ready to kill the trickleware model:

http://www.dvinfo.net/forum/panason...orders/51044-hvx200-vs-z1-fx1-simple-guy.html

The death of the digital video camera trickleware business model needed a champion. And he has been that champion.

The diversity and complexity of delivery systems will make this a much harder giant to slay.

But I think with the release of the EPIC and SCARLET cameras, RED will have set itself up quite nicely to push this effort. And, as I said in a different thread, I think that RED RAY, the technology itself, has been moving along, we have yet to see all that this box can do. The curtain has not yet been lifted...I'm betting that once it is, we are going to see another quantum movement, not just a little burp in history but a real revolution, much like we have seen with the release of RED. New users finding new ways to deliver media from new content providers. A release of new people and new energy systems who will rally around these machines.

One thing is for certain, win or lose, I am going to enjoy my ringside seat on history being made, both as an owner/participant and also for the sheer voyeuristic pleasure that reduser affords!

Whether or not that is ultimately a ubiquitous 4K delivery system in my lifetime or not remains to be seen, but Jim Jannard has dealt himself a very strong hand.

 

[Erwin Vanderhoydonks]

You should shoot whatever you can afford to shoot. The problem I have is the conscious decision to shoot 1080P when budget is not an issue...

The message is out there that "1080P is good enough". I'm just trying to make sure that everyone knows that concept is not accurate.

Jim

- Shoot whatever you can afford :
Well for the moment that will be the RED Scarlet Fixed 8x. And deliver on 1080p Blu-Ray (and DVD) because that's the only thing for the moment people have in there living room.
- But I want to shoot 4K (or 4.5 K) because 4K will be the future and 4K wiil not be the end. But if 4K is the near future why shoot 3K ?

I think the RED Scarlet Fixed 8x, or will it be EPIC-F in the future ;-) , will get an upgrade in the near future. Just like the REd One moved from 4K to 4.5K.
Or am I wrong... The future will tell us...

 

[M Most]

Thanks Mike.

Just thinkin' out loud.
for now the projects with 2k (P3) and 1080 (rec 709) masters have not been the same, so it is just a bit of pondering...

If you read the OpenEXR documentation, you could, in theory, actually come up with a pretty efficient way of doing this by using tiled images, sizing the tiles to some multiple of the smaller formats that fit into the largest one, and writing some code to extract only the tiles for the smaller format when required. This is where standardization comes in, and why I always bring up industry wide initiatives as keys to making these kind of things useful. My guess is that there will ultimately be a development path for OpenEXR - probably as an extension of the ACES effort - that will cover such things. But it all has to start with the data itself, and the ACES effort is currently concentrated on coming up with a theoretically unlimited color space that will accommodate any current or future display technology via specific transforms for particular displays. Things like multiple standard deliverable frame sizes within one container can come later. With any luck, larger frames would be standardized as being multiples of smaller frames, a la Red's 4K HD format. In that scenario, the EXR tiling approach would work quite well, as would scaling. Of course, in an ACES world, there's no need to separately store multiple deliverables anyway, since they would all be created from the original ACES data using standard transforms.

Progress. It's sometimes slow.

 

[Gunleik Groven]

I do think that's very forward thinking. I've actually advocated "house" mastering formats for quite some time, as long as the "house" format can be used to directly produce any deliverable that's called for. Your approach seems to have that covered (for now... :undecided:..).

Do you know of any method other than command line tools for actually creating multilayer OpenEXR files with the differing types of data - along with metadata - you're describing? Does Clipster actually do that? Not that I have anything against command line (it's actually my first choice in many, if not most, cases), but I'm trying to shortcut questions that others will likely ask...

Yup...

I am a probable asker...

 

[John Stamets]

First post after following and reading about RED since it's inception (lurking on this board for 10 years - Jim, you should really make a movie of your journey - it's been an amazing thing to follow). I am not a professional in the film industry (yet), although I spent 15 years in theatre before abandoning my passion to pursue a career as a software developer for the last 10. I hope to engage one more career change back into the entertainment industry as an independent filmmaker.

That being said, I have lived through all of this before - almost word for word - back in the early 90's when analog audio was switching to digital. The arguments were all about sample rates and bit depth and the "massive" bandwidth that was required to move the data as well as the incredible amounts of harddrive space. There was NO WAY that digital could possibly sound as good as an analog master running 30 ips. Besides, the digital audio formats were proprietary and the disk space was prohibitive.

Flash forward 15 years. No one gives a second thought to digital audio anymore - the sample rates and bit depth reached a real, live "Good Enough" level (you still have analog purists to this day - but they are the fringe). My kids have no concept of vinyl, cassettes, or even CD's (they think that they are all games or DVDs). Electronic delivery is the norm for them - the next generation of consumers. A few years ago, my son and I were sitting in a cinema watching a film (as in shot on film) and my son asked me what all the "stuff" was on the screen (he was maybe 10 or so then). I wasn't sure what he was talking about. He said that it was fuzzy, dirty - didn't look right. Turned out he was talking about film grain and he didn't like it! That is the next generation. What looks natural and desirable to today's (and yesterdays) generation of professionals is objectionable to the next generation.

When I was doing live production, I remember switching from 3-gun projectors needing hours of alignment prior to use to DLP (a godsend, I might add - I could actually get some sleep before a show). Within 4 years, our 50 thousand dollar DLPs were available for a few thousand. A couple of years later you could buy one at Staples or Best Buy.
1080 is the future of SD - we already have 1080p in our living room. I personally hate watching a real film at the cinema. Scratches, poor assembly, dust, dirt - all detract from the experience. While shooting in film may still be desirable, for now, it's at end of life. We are at a point of critical mass and, if it follows the same trajectory as audio, within 5 years everything will be digital. RED is pushing the competition and they will respond. Probably quickly and things will happen very, very quickly as the market players reposition to stay on top. Arri or Panavision may own the field now, but so did Otari - and I remember when we ordered the very last of the 1/2" analog tape that 3M produced. If it follows the same trajectory as audio, this arguing and discussion will continue for another year or two until a clear standard emerges that the audience (the ones that pay for all this technology) is happy with "good enough" and then it will switch to other improvements. The Digital audio revolution was pioneered by Bell Labs and Denon, but wasn't popularized until Sony release the R-Dat and Tascam made it affordable to the masses. Once tools like Mac/Cubase came along, it was all over for the big boys. Record Labels are having to reinvent their business models it changed the industry so much. The same will happen to cinema. I see Red as the catalyst.

 

[Mark L. Pederson]

Do you know of any method other than command line tools for actually creating multilayer OpenEXR files with the differing types of data - along with metadata - you're describing?

I'm going to beat the Foundry up until this is in Storm. They will have to do this just to shut me up. Same goes for DVS.

 

[Dj Joofa]

Regarding 24 fps or more, i.e. bringing temporal resolution into the discussion as well: AFAIK there has not yet been any research of scientific validity about the difference in perception of fiction in 24 fps vs. say 50 fps.

Several studies have been done. See for example, "Visual responses to time dependent stimuli" by D. H. Kelly in Journal of Optical Society of America, vol. 51 (1961), pages 422-429.

Joofa

 

[Uli Plank]

Thanks for the source!
Is this aimed at cinematography in particular? Our library is closed today (they should have a subscription to OSA).

 

[Deanan]

Do you know of any method other than command line tools for actually creating multilayer OpenEXR files with the differing types of data - along with metadata - you're describing? Does Clipster actually do that? Not that I have anything against command line (it's actually my first choice in many, if not most, cases), but I'm trying to shortcut questions that others will likely ask...

Going on a tangent here since it's somewhat related:
REDcineX and REDline both output full metadata used to create the image as well as all the other metadata in the R3D (excepting lens metadata will be added in the next month or so).

Both the apps also output multiview EXRs for stereo and once HDRx is out, multilayer EXRs containing both exposures.

 

[Gunleik Groven]

Does it also export the r3d frame, as Mark mentioned?

 

[Deanan]

Does it also export the r3d frame, as Mark mentioned?

We output some of the framing data but I don't think it's what mark was talking about. We can add more as desired. (ie. there is on camera framing, vs post adjusted RCX framing which you'd have to output full image but not crop to)

 

[paulherrin]

I'm going to beat the Foundry up until this is in Storm. They will have to do this just to shut me up. Same goes for DVS.

haha, good job...

and deanan, that's good news indeed! thank you

 

[Tom Lowe]

Regarding TVs/monitors. The same pattern happens with every new technology. You could use the same argument for there for 1080p before 1080p was a common format.

This is a good point. I remember watching "1080p" videos on my computer and LCD monitor long, long before I had a large-screen 1080p TV at my house. Till this day, my cable company only broadcasts 720p "HD".

This further boosts my hunch that 4K could begin at the computer, with 40-inch 4K LCDs, for example. A 4K monitor would be beautiful for amateur photographers to grade their photos in Photoshop. It would also be ideal for video editing, and video games.

What happens when Vimeo starts to offer 4K videos? I could master my latest video, Rapture, as 4K right now and upload it. When "1080p" first hit computers, the videos were limited mainly to movie trailers and short "HD" videos. The same could happen again with 4K on computers.

To me, 4K is incredibly exciting. It's like the difference between SD and HD. 4K looks like a moving photograph, not video. All of us would like to see our work presented in the best possible way.

 

[cali]

I've decided. I need to talk louder...

Jim

OVER THE ROOFTOPS JIM....OVER THE ROOFTOPS...YOU'VE GOT THE FUTURE BEHIND YOU ALREADY!

 

[M Most]

We output some of the framing data but I don't think it's what mark was talking about. We can add more as desired. (ie. there is on camera framing, vs post adjusted RCX framing which you'd have to output full image but not crop to)

I think what Mark is talking about is using the file as a pure data container, not unlike what S.two does in rewrapping Arri RAW files in DPX "containers." I'm not so sure I agree with that concept, as I think theres a lot to be said for keeping a single image file as an image file and not confusing either the user or the software as to what is and what isn't RGB image data. It is technically feasible, but I would resurrect my oft stated mantra on this one: just because you can do something doesn't mean you should. Trying to do too much in one file is a sure ticket to unacceptance, especially when what you're doing is non-standard.

 

[paulherrin]

don't be a grinch mike

 

[M Most]

don't be a grinch mike

???????

Do you have any experience with using S.two DPX-wrapped Arri Raw files, and do you understand how confusing it can be to send out something that calls itself a DPX file, but doesn't actually contain an image that any software can display? Do you understand the difference between what should be done with an open file format like Open EXR and what a single vendor can do with their own proprietary formats (like R3d, for instance)? And lastly, do you understand the need for files and file extensions to predictably contain what they're supposed to contain so that software can be written to properly interpret it?

 

[Dane Brehm]

I've done some work for some 4K TV Manufacturer's and here is what we used to capture a true 4K image

This was in addition to shooting with the RED MX but this was what the client wanted to do for a portion of the shoot.

Mike I completely agree with you as I have indeed used S.two.

 

[paulherrin]

easy mike, it was a joke. and yes, i understand... you made some good points.

 

[Ivan Kovax]

To me, 4K is incredibly exciting. It's like the difference between SD and HD. 4K looks like a moving photograph, not video. All of us would like to see our work presented in the best possible way.

Tom - I remember the first time I saw a 4K image on a flat-panel - an Astro demo nearby my office here in Tokyo. We put up some footage I shot of a Ducatti motorcycle tearing it up across the Rainbow Bridge from our camera car.

It was truly hard to believe that these still photos were moving. The jump from SD to HD finally took the veil off our TVs. HD to 4K will bring clarity... now just TRY to imagine 4K S3D! Try it, I dare you!