If your shooting 28K resolution, then your might as well be shooting in IMAX!

[I Bloom]

The FF35, 645 and 617 are next gen sensors and considering how much more complicated the 617 is than anything currently available, I can see it as a product quite a ways off. The processing power needed to capture 24 261 megapixel images a second will be astounding. I am not saying it will not be possible but just take awhile to be production ready. Film 617 for motion is one thing, but digital is quite another. It will be possible but not practical. The storage issue will be a major hurdle. For advertisers, there are plenty of 617 film cameras that do not cost $70,000. Sure, you could window the frame, but then why would you need the massive sensor.

Traditional filmmaking isn't the only application for motion pictures. The future includes lots of advertising that used to be considered print that's going to need to be in motion to be competitive. Likewise the resolution of the new kinds of screens being manufactured is going to increase as the cost decreases. So the 28K Epic is for me about motion billboards and motion print advertising. As well as lots of applications for "normal" filmmaking. So unfortunately Andrew, I think you are just asking to be proven very very wrong, despite your arguments being very well thought out and practical.

IBloom

 

[Andrew Rieger]

Traditional filmmaking isn't the only application for motion pictures. The future includes lots of advertising that used to be considered print that's going to need to be in motion to be competitive. Likewise the resolution of the new kinds of screens being manufactured is going to increase as the cost decreases. So the 28K Epic is for me about motion billboards and motion print advertising. As well as lots of applications for "normal" filmmaking. So unfortunately Andrew, I think you are just asking to be proven very very wrong, despite your arguments being very well thought out and practical.

IBloom

All I am saying is, the technology needed to implement such a workflow is not practical, at least not anytime soon. People seem to get upset when you say that a product is pointless. I my opinion, the Epic 617 is pointless. There is no practical way of projecting the footage, I believe Cinerama used three projectors and a curved screen. You would need to build custom screens to show the footage in all its glory and we are talking about a screen much much larger than IMAX. For advertisers, why not just use a Linhof 617? The price is around 1/10th of the projected price of the Epic 617. Unless a company like IMAX embraces the camera, I can not see how this could be implemented. About the motion billboards. Do they need 28K? The world's largest screen (Dallas Cowboy Stadium) is capable of HD as a max, no where near 2K. Like IMAX, there will be a few instances of use but like IMAX, the uses will be few and far between. There are like 25 IMAX cameras in the world. I cannot see photographers using this, not with the price tag, and based on current workflow, graphics rendering capabilities, digital storage and projection, such a camera will not be practical for a very long time. Also looking at the camera itself, it does not appear to be compatible with the DSMC system. About lenses, the lenses used for the 617 format are stills lenses with built in trigger release shutters like large format lenses. They are slow and not suited for cinema use. Red would need to develop their own cinema lenses, adding to the release date. The Film Baraka was a Todd AO film shot on 70mm film and scanned at 8K for Bluray, the workflow was a nightmare and took many months to complete, just imagine 28K. If Red can pull it off, great for them. These massive formats are not practical, the Dark Night team could only fit around 20min of IMAX footage into the film, due to the film cost and post issues.

 

[I Bloom]

I think you may be thinking linearly. I believe it's more appropriate to predict this stuff exponentially.

In any case it's not 28K Uncompressed. It's 28K Redcode. Film scans are Uncompressed RGB. That's nearly 60x the amount of data used in Redcode. So in fact a 28K Redcode conform is about as practical as a 4K uncompressed conform. People tend to take the technology that is available to them and find uses for them. I can think of a few for Epic 617.

Ian

 

[Andrew Rieger]

I think you may be thinking linearly. I believe it's more appropriate to predict this stuff exponentially.

In any case it's not 28K Uncompressed. It's 28K Redcode. Film scans are Uncompressed RGB. That's nearly 60x the amount of data used in Redcode. So in fact a 28K Redcode conform is about as practical as a 4K uncompressed conform. People tend to take the technology that is available to them and find uses for them. I can think of a few for Epic 617.

Ian

You are correct, workflow will not be as big a deal but storage will still be difficult. Also, I feel that the hardware required in camera to capture and store the data will be a problem. As I stated previously, large format digital cameras are scanners and take a couple of seconds to capture one image. Red wants to do this in 1/24 of a second.

 

[Jeff Kilgroe]

All I am saying is, the technology needed to implement such a workflow is not practical

Once again, I disagree. RED is creating not only cameras, but support systems and workflows for these cameras. Additionally, computing power continues to increase, as does storage abilities. Desktop CPU power has effectively doubled in the lifespan of the RED One, so has storage capacity. What may be impractical for you could very well be workable for someone else. I could deal with 28K footage today a lot easier than dealing with 1080p animation frames 10 years ago, which I did. And I'm not talking about simply dealing with the finished frames, but managing render farms that created them, often two CPUs cranking away on a single frame for hours, or even a day or two.

Jim said just the other day on c.com that we would see the Monstro sensor later this year.

People seem to get upset when you say that a product is pointless. I my opinion, the Epic 617 is pointless.

I'm not getting upset. If you feel it is pointless for you, then it is. If I had one in my hands right now, I know I could ROI the thing almost instantly.

There is no practical way of projecting the footage, I believe Cinerama used three projectors and a curved screen.

Who said anything about projecting it... There's a lot more to images, still or in motion, than trying to have them projected. The immediate benefits of the Epic 617 will be from the panoramic format and optical characteristics. Resolution will of course play a role too, but 28K projection is not the immediate goal here for anyone I know of who wants one of these.

For advertisers, why not just use a Linhof 617?

No motion. 'nuff said.

As for the rest of the comments, you seem to just be focusing on the resolution and not much else. Please go shoot some 617 stills and get a feel for what the format offers. Then ask yourself just how totally cool it would be to have that at 24fps. As for the resolution, I have some uses in mind for that too, but I can't spill all the beans right now. This camera either makes sense to you or it doesn't. If you don't see a use for it, don't buy it. But for many of us on here, the promise of the 645 and 617 is still a lot more exciting than just about anything else.

You are correct, workflow will not be as big a deal but storage will still be difficult.

Funny... I'm thinking storage will be the easy part.

 

[Andrew Rieger]

Since I come from a still photography world, I may not fully understand the logistics of motion capture when dealing with images this large. Today, the only digital cameras that can capture an image in the 250 plus megapixel range, take a couple of seconds to do so. I am very excited about prospect of capturing such a large image in only 1/24 of a second but it seems so far away. I look at the cost of a 50+ megapixel Hasselblad digital back (around $45,000) and have a hard time contemplating a digital camera that can capture an image so large. Since I am not someone who is an expert in post, the numbers seem very daunting but for some with render farms and a host of servers, this may not seem so bad. My mac throws fits when I give it multiple 50 megapixel images to deal with. This camera would be in no way useful for me but it could be a cool product for others. Again, no red bashing on my end and I'm sure they can pull it off but I am skeptical about the time frame of release and the usefulness in the motion picture world. For billboards, sure. FX work, a possibility. Photographers, yes but there are easier, cheaper ways of getting the same thing. I think the lens issue is something that needs to be brought up. Does Red intend to build lenses for this camera? I could not imagine using large format stills lenses to shoot a film. If I did not believe in Red, I would not be on this forum. I'm sure others told Jim and company that the stuff they wanted to do with the One was impractical. Ya, I will probably be eating my words. What do I know, I'm only 20 and Jeff, since you think it is possible in the near future, I believe it. Is it really practical today? No, but in a couple of years, who knows.

Jeff: "Please go shoot some 617 stills and get a feel for what the format offers. Then ask yourself just how totally cool it would be to have that at 24fps."

I have shot a ton of Medium Format stills and some view camera pictures (mostly 8x10, but also some 4x5), so I do totally appreciate the power of huge negatives and yes, they would be very very cool in motion.

 

[Rick Idak]

My personal feelings to all of this is Red needs to stick with say getting the Scarlets out. With no more delays.Then the 5K Epic, while these are selling and returning a profit. Maybe then start with the 645 & 28K cameras.Otherwise none of the concepts will ever see the light of day.
Don't get me wrong who here wouldn't love to have a 28K res camera.But from a business point of view 1 camera in the hand is better than several on the drawing board.

 

[Robert McGee]

I think I've already established the fact that I would love it, but maybe it's ready for a video camera just yet. I thinking a still camera is not a bad idea for this megapixal sensor but it would have to have a very large SD card, we may have to "Super Size" it. Seeing what we can get from stills will give RED an idea what they will get from moving images.

 

[Wesley Scoggins]

That's also assuming that you'll be shooting AND saving in 28K.

I imagine quite a few people will be using 28K and pushing it a whole bunch, then converting it to 8K for the editing purposes.

That's a big reason I could see to shoot with 28K, imagine being able to push it so much, and it being so big, that even pushing it 15 stops, there still isn't visible noise once you convert it to 8 or 4K for a release print. The POTENTIAL of what you can do with 28K is astounding, and I don't think should be discounted as useless.

Just because we don't have a use for it YET, doesn't mean it's useless. A Red One would have been pretty "useless" in 1995, there simply wouldn't be anything fast enough to transcode the footage.

I am sure that Red is imagining a beefed up Red Rocket to process that footage. While it would be quiet hard to work with 28K footage natively, I imagine that Red will have some kind of option to transcode the footage to something that is easier to work with.

Also, Intel is getting ready to release 48-core 64-bit processors, imagine how much easier footage will be able to deal with in a few years, after all the software is optimized to take advantage of both of these things?

And that's just with current machines, imagine the much powerful computers that we'll have a few years from now, with OS's and editing software taking full advantage of them, and beefed up Red Rocket's working in SLI-mode.

I don't think it's that "out there" to think that 28K footage will have it's place, or be workable with the editing technology we'll have in a few years.

 

[Robert McGee]

You saying that the Red Rocket needs to be beefed up well just read this thread that I got on the off-topic section: http://reduser.net/forum/showthread.php?t=43217

It talks about a proposed item called the Red Rocket 2, and not just being the right peace of equipment for Epic and Scarlet cameras but also being used to upscale 1080i/p footage from a consumer grade camcorder to 2K and 4K in .r3d format because not all HD footage is live-action, sometimes it could be CGI-animated footage.

I would also say that the Red Rocket 2 would be the right tool for the next generation of Red Cameras but I'd keep making more Red Ones as long as people keep buying them.

 

[Peter Borosh]

I am someone who has actually worked on a few Imax productions and something the Red has over all the Imax cameras, even the one that was used for Batman Returns, is that you would be able to have a Silent Sound Sync camera. All the sound from majority of Imax productions is a product of total audio replacement. There isn't a sound recordist with you as the Imax cameras are too noisy.

 

[Robert McGee]

I guess one would say that an IMAX camera is so loud, you need ear plugs but some film cameras always give off some sort of noise of the gears inside running the film.

Now, why hasn't IMAX come out with their own digital cameras? RED is just taking it upon themselves to make cameras that shoot at 9K resolution and 28K.

 

[JohnF]

I can think of a number of subjects that are perfect for the 617 Epic and I am more than a little eager to get out and shoot them with it.

Is a 3:1 aspect ratio a problem? IMO, no. To give an extreme example I've shot 3:1 (and 3.5) a few times and squeezed it into a 16:9 format. To my surprise clients and the audience loved it and, due to downsampling, it looked fantastic - noise almost became a non issue.

I've seen even wider formats used in commercials on TV (squeezed into 14:9!).

That said the benefits of this camera quiet, small etc are extraordinary for IMAX shoots and crews but the question of how a project is delivered from such digital formats (Epic 645, 617) comes to mind? I don't see many true IMAX resolution digital projectors out there at the moment and producing a film print of such size is, as far as I know, difficult....

One thing I know is I'd be more than happy to shoot a project on these cameras for any deliverable format simply for the art and the beauty that they will be able to produce.

JohnF

 

[Jun-Dai]

I think everyone's forgotten that the real reason RED will come out with the 617 is that Jim wants one himself. We're not that important a part of that decision, though I'm sure he'd be very happy if it turned out there were more than a few others that were able to take advantage.

As to whether it's a good use of RED R&D? Well, I'm guessing this camera's something that will get worked out once the more obvious "game-changers" are past the R&D stage (and I'm guessing most of the R&D is shared across the other Monstro cameras). That said, it's Jim's money going into making this thing, and if sales never pay for the camera's R&D, that doesn't mean it wasn't worth making. If I had a billion dollars to invest in my own company, I'm sure I'd find a couple things I wanted to build where I wasn't overly concerned about (monetary) ROI.

 

[Philip Allister Anderson]

I know exactly what will happen with the 617. Producers will only let me rent one wide angle lens and then they'll just crop everything out of the wide. Composition, why do we need composition?

 

[Robert McGee]

I don't believe in wasting the image, because that's like throwing away good real estate. I can understand a filmmaker shooting HD 1.78:1 and slicing off part of the image to get 1.85:1 to make the film prints and then the full aspect ratio is preserved for the DVD and Blu-Ray release. Shooting HD 1.78:1 and going as far to slice of as much to go 2.35:1, now that right there is a waste, if Revenge of the Sith is ever released on Blu-Ray, I'd like to be able to see the full 16:9 version, we all know what Super 35 is where you shoot full frame on film and slice off the top and bottom of the image to get a 2.35:1 image, that to me is a waste but now you got a digital version like "digital super 35" where you shoot at 16:9 and slice off the top and bottom of the image to get a 2.35:1 image, well atleast there isn't any physical waste. There had been 35mm to 70mm blow ups long before the IMAX DMR, Gone with the Wind is well known for being one of them.

Now that the first Epic model, a 5K model will be ready soon. I guess now later models are to come faster then expected.

 

[Robert McGee]

The reason why I posted this in the first place is because of the fact that aside from the wikipedia article, you just got to do the math. Now IMAX 15/70 is 6 times bigger then a standard 35mm film frame.

As you can see in the image above. Now 4K is the mathematical equivalent of 35mm film. Now lets to the math 4 x 6 = 24. That means that if you where talking about an Epic or Scarlet model that shot 24K resolution, that is actually the minimum. Anyone who has seen an IMAX film in an actual IMAX venue, will tell you how it just doesn't compare to 35mm projection. Where talking about 28K and goes beyond that, that would infact be super clear on IMAX 15/70 because it is in fact much larger image then what you need.

 

[Jeff Kilgroe]

The reason why I posted this in the first place is because of the fact that aside from the wikipedia article, you just got to do the math. Now IMAX 15/70 is 6 times bigger then a standard 35mm film frame.

Your math is off as you're mixing and matching dimensions and area. Comparing to the 35mm academy frame you have referenced, a 70/15 IMAX frame is actually 8.32 times larger than that 35mm frame!

Next point, if 35mm film, or technically 35mm cinema film size resolves to about 3.2K, we can make the following deductions. 70mm is 3.19 times wider than 21.95mm or in other words, 70mm IMAX should give us approximately 3.19 times the lineal information. Talking resolution, that would be 3.19 * 3276(3.2K), we get 10,450. And 4K * 3.19 = 13,066. So that means on 70mm iMAX, the equivalent to a 35mm cinema 4K scan would be a 13K scan and it should resolve somewhere close to 10K of horizontal resolution.

The recordable width of the EPIC 617 should be 28K or more directly 28000 pixels even based on outdated, preliminary specs. That is 168mm with a 6um pixel size, or 16.8cm -- right about where it should be. If we record a 70mm wide frame on this sensor, we should have 11,666 pixels of horizontal resolution.

Mysterium-X is delivering a nominal resolution close to 90% of its full raster size. If the same holds true for the larger Monstro sensor, then we would end up with 10,500 pixels of nominal resolution at a 70mm frame size. Or about what we should see as nominal resolution from a film scan.

Of course, we won't know how all of this will come to be until the cameras are delivered. In terms of overall image size and resolution, the Epic 617 should stomp IMAX into the dirt. It is roughly 2.5 times larger than the 70/15 IMAX frame, with roughly the same amount of detail per square millimeter.

 

[Robert McGee]

Let me tell you this, Super35 is just about the same a Techniscope. The standard 35mm film frame is 4 perfs high, in Techniscope your using a 35mm film frame that is 2 perfs high, it gives you a CinemaScope aspect ratio without an anamorphic lens. Super35 is when in post, you chop off parts of the top and bottom of a full size 4 perfs high frame, now a days it may be a 3 perfs high frame, either way your doing to get a CinemaScope aspect ratio.

Now when it comes to IMAX, your recommending a 13K scan. Right now for the DMR process, only 8K scanning is available for 35mm and they usually do that for 4 perf high film. I don't know what they do for digital video, like you get from a RED.

Like you guys already said, the 617 ain't do out for a few years but how close would you say IMAX is do developing a 28K resolution digital camera of their own?

 

[Jeff Kilgroe]

In my above numbers, I did not account for Super35 for a couple reasons. First of all, to relate to the RED One 4K format, which is very close in physical size to 35mm Academy. RED One sensor is 4K at 22.12mm wide vs. 35mm Academy frame at 21.95mm wide.

The frame height makes little difference, it does not matter if it's 2 perf, 3 perf, or even 4. Scanning resolutions are represented and figured off of the frame width. Height is incidental and the same lineal resolution is maintained and the actual area scanned is extended to match the height of your frame on the film neg.

Given the same numbers, it all holds the same for Super35, 65mm and other formats. Full aperture S35 is 24.89mm wide. RED One 4.5K is 24.19mm wide. Epic 5K is 5120 pixels and 27.65mm wide.

IMAX is developing their own digital camera systems. I doubt they will put it to use if it's not at least as good as what they're getting from their current DI process for film acquisition. Most IMAX DI and post is done at 4K right now. Some is going to 8K for origination scan -- such as the remastering for Baraka, iMAX shots for The Dark Knight, etc..

My numbers above are linear abstractions, but don't represent other factors such as the influence of various film stocks, lens properties and whatnot. I think to maximize the usable detail in 70mm IMAX film frames we need a 12K to 13K film scan. The current 8K scans are not quite enough. ...Or maybe they are, I have not actually worked with them myself to know. I have seen 8K scanned images and they hold amazing amounts of detail from 70mm film, but I think there's a good bet another 3 to 4 K of resolution can be had there.