Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

65mm from RED

Here's what I'll say in somewhat deliberately abstract ways.

RED took a major freaking gamble being the first to market with a larger format and higher resolution VistaVision style sensor.

This was a major gamble in two directions.

1. Resolution, it was the first 8K digital cinema camera

2. Lens selection towards the format size, not many professional options at the time VV was first shown


However, those were educated gambles. 8K holds a great deal of importance for those who are trying to deliver high quality 8K through 4K or even lower content. I don't like speaking of other filmmakers if I can avoid it, but conversely much of what Nolan has been up to over the last few years lines up with exploring larger format sizes and better fidelity towards the image quality.

The larger format is incredibly attractive to creative professionals who have been looking to explore some new horizons.

Towards the industry, I think much of this is truly needed as various film formats held so much allure and creative potential and really not until recently did we truly have format freedom.

Remember, some of the reason Monstro and Dragon VV make so much sense is having that 8K resolution in the larger format, but also covering S35 in the 5K-6K range. I love the idea of a multi-format and multi-resolution camera. This is something film couldn't exactly do as you needed different cameras to tackle those formats and film neg sizes.


To RED's credit on that gamble I think it's clear to say that other manufacturers got the memo on the larger formats. RED's two main competitors found their way into this arena in 2018.

And in 2019 you will see they also were ahead of the 8K trend by a fair amount.


Personal take though. In many ways I think VV is the sweet spot format-wise for general larger format photography. Glass options for days and really creative ones at that. 65mm-ish would also work great. 70mm full 15-perf, I'm not totally sure. I don't think many of us want to be shooting and/or lighting at T8 or beyond if we can avoid it. I've worked with most motion picture formats and for me it all sings in the 16mm through 65mm-ish range approximately.

Beyond that, I'll take my thoughts off the air as I'm loving 8K VV.
 
The Gemini sensor is the closest to get 65mm and 8K. It has the best individual pixels of the sensors and, I'm not sure about the math here, but either it resolves at around 8K for a 65mm sensor, or it becomes a bit higher, able to downsample to a better 8K output.

All in all, I've been proposing a Red 65mm for a long time. At the moment, with Arri and others momentum towards higher resolution and larger sensor sizes, it's a logical step for Red to create such a sensor as the highest resolution and size sensor in the world, with the best downsample available. Using the Gemini architecture for this would be the most logical if no other photosite tech is in development.

Personally, I don't like squeezed pixels that much. I rather see a totally new sensor type with perfect 8K acquisition in 65mm than anything else. Larger photosites at 8K with 65mm would be a perfect balance in my opinion, between all the aspects of a CMOS sensor tech.
 
here's some specs for the Larger Sensor Cameras out there

(54.12 x 25.58) max FPS 60 - Alexa 65 6K
(36.70 x 25.54) max FPS 60 - Alexa LF 4.5K
(40.96 x 21.60) max FPS 60 - Red VV 8K
(35.90 x 24.00) max FPS 25 - Sony Venice 6K
(28.90 x 15.20) max FPS 120 (external recorder) - Canon c700 4.5K
(38.10 x 20.10) max FPS 60 - Canon c700FF 5.9K (not released)
 
Last edited:
Is there a market for digital cinema cameras that have even bigger sensors than the current crop? Of course - but how many?

I LOVE larger format photography, still or motion, but at some point the physics start to limit practical use cases. I can see 60 x 30 being pretty sweet, but beyond that...

Cheers - #19
 
Christoffer Glans said:
The Gemini sensor is the closest to get 65mm and 8K. It has the best individual pixels of the sensors and, I'm not sure about the math here, but either it resolves at around 8K for a 65mm sensor, or it becomes a bit higher, able to downsample to a better 8K output.

All in all, I've been proposing a Red 65mm for a long time. At the moment, with Arri and others momentum towards higher resolution and larger sensor sizes, it's a logical step for Red to create such a sensor as the highest resolution and size sensor in the world, with the best downsample available. Using the Gemini architecture for this would be the most logical if no other photosite tech is in development.

Personally, I don't like squeezed pixels that much. I rather see a totally new sensor type with perfect 8K acquisition in 65mm than anything else. Larger photosites at 8K with 65mm would be a perfect balance in my opinion, between all the aspects of a CMOS sensor tech.
Click to expand...
Yes, for RED, it would be a very logical step to create a 65 mm camera with a resolution of 8 k. I think that would give a huge step in the business of RED. I hope that RED will do it.
 
I don't see a big market at the time for DCI-8k on the otherhand the 8k SUHD (7680x4320) is happening at the moment and the only camera at the moment that I know of that can produce true 8k SUHD output is the Sony UHC-8300 3 sensor camera https://cvp.com/product/sony_uhc-8300.

With a bayer sensor you would need 7680x4320 ~ 33 green Mpixels to produce a true 8k RGB output. Since there are not many lenses(Otus and MasterPrime) that can handle the lp/mm of the 3.65 um helium pixel, the excellent Monstro sensor seems to be a good candidate to be scaled up to 65mm and get a perfect de-bayered 8k output.

7680xsqrt2~10.9k and 4320xsqprt2~6.12k, with a pixel pitch of 5um the sensor would be around 54.5mm x 30.6 ~ 62.5mm image circle.

Haven't seen an 8k DCI theater yet, but Samsung is already selling 8k tv's for around $5k.
 
I think that we’ll see all kinds of video options being made over the next decade. It started with adding video to DSLRs and now we have 360-degree specialty cameras, action cameras of all kinds, drones, and, yes, eventually we’re going to see medium format, large format, and panoramic video cameras. They might be specialty items or just higher end versions of what we have but no matter what kind of camera it is, video is almost always expected now.
 
Theatrical exhibitors are motivated to offer an experience you can't get at home. If that includes 8K, which seems likely at some point, it will require upgrading several elements of the DCI ecosystem. There are some DCP variants (B & B1) codified by SMPTE that go beyond the 250mb/s cap on IOP (interoperable) compliant DCPs, but even at 500mb/s, 8K HDR @24fps isn't supported.

I fully expect a new school version of DCI that can support 8K at up to 60fps with HDR and 24 audio channels to manifest. Will it still rely on the JPEG2000 entropy codec? If not, then what? What would be the economics of going to a 1,500mb/s data stream - both for mastering and exhibition? Is there any codec that could provide enough image data at less than a gigabit/s to properly support 8K HDR 12bit @60fps content?

UHD/4K to the home has been a slow road so far, but by 2020 I'd expect it to become the entry point for premium content. If so, how much investment would exhibitors be willing to make to go 8K? As of today, the cost of that move would be extremely difficult to ROI. When might that equation start making sense...

Cheers - #19
 
Blair S. Paulsen said:
Theatrical exhibitors are motivated to offer an experience you can't get at home. If that includes 8K, which seems likely at some point, it will require upgrading several elements of the DCI ecosystem. There are some DCP variants (B & B1) codified by SMPTE that go beyond the 250mb/s cap on IOP (interoperable) compliant DCPs, but even at 500mb/s, 8K HDR @24fps isn't supported.

I fully expect a new school version of DCI that can support 8K at up to 60fps with HDR and 24 audio channels to manifest. Will it still rely on the JPEG2000 entropy codec? If not, then what? What would be the economics of going to a 1,500mb/s data stream - both for mastering and exhibition? Is there any codec that could provide enough image data at less than a gigabit/s to properly support 8K HDR 12bit @60fps content?

UHD/4K to the home has been a slow road so far, but by 2020 I'd expect it to become the entry point for premium content. If so, how much investment would exhibitors be willing to make to go 8K? As of today, the cost of that move would be extremely difficult to ROI. When might that equation start making sense...
Click to expand...

Without breaking any NDAs what I can say is every needle points to a:
- 4K @ 120fps max
- 8K @ 60fps max
- 12-16bit (10-12 today)

Future.

I'll leave sound out of the mix, pun intended, for now :)

I think for cinema they really need to look at data rate in the face of what is coming over 5G and Gigabit. Color and resolution are where cinema will still be able to really differentiate itself from home streaming offerings.

As for the ATSC 3.0 standards, I'm behind the effort, but I fear they are lagging a bit far behind the line and likely can't fully compete with the streaming marketplace within two years. Though in our country it will be a moderate requirement and likely subsidized eventually. At that point a bigger issue will be at hand though.

12bit display technology is happening in the R&D world aggressively at the moment. Getting a good 10bit display for a decent price truly happens next year and you will see offering from some unexpected corners.

I'd chop off a toe for a decent DCI 8K monitor in the 30-40" range. Not my toe mind you, but somebody elses to pay for that thing. 8K goes single cable HDMI next year proper from what the crystal ball says. That's the biggy for consumers really. Then what drives them. The 8K stream box today is around $2500 USD. That will come down in price rapidly.
 
Blair S. Paulsen said:
Theatrical exhibitors are motivated to offer an experience you can't get at home. If that includes 8K, which seems likely at some point, it will require upgrading several elements of the DCI ecosystem. There are some DCP variants (B & B1) codified by SMPTE that go beyond the 250mb/s cap on IOP (interoperable) compliant DCPs, but even at 500mb/s, 8K HDR @24fps isn't supported.

I fully expect a new school version of DCI that can support 8K at up to 60fps with HDR and 24 audio channels to manifest. Will it still rely on the JPEG2000 entropy codec? If not, then what? What would be the economics of going to a 1,500mb/s data stream - both for mastering and exhibition? Is there any codec that could provide enough image data at less than a gigabit/s to properly support 8K HDR 12bit @60fps content?

UHD/4K to the home has been a slow road so far, but by 2020 I'd expect it to become the entry point for premium content. If so, how much investment would exhibitors be willing to make to go 8K? As of today, the cost of that move would be extremely difficult to ROI. When might that equation start making sense...
Click to expand...

You do realize that essentially all tentpole releases today are finished at 2K, don't you? Unless someone can magically extend the post schedules for these kind of pictures - and for the most part, those schedules are always getting shorter, not longer - there are significant technical and artistic barriers to getting 2500 VFX shots done at the quality level that's expected at anything higher than that in the near future. That's the reality for large scale productions, not what people here seem to think it is...
 
By the way, I'm thinking about buying an 8K monitor, and since the choice is not big, I look at a monitor from DELL or QLED 85-inch TV from Samsung 8k. Accordingly, I will work with 8K material from Monstro and will do Color Grading. What do you think is more appropriate? and is it worth it?
 
M Most said:
You do realize that essentially all tentpole releases today are finished at 2K, don't you? Unless someone can magically extend the post schedules for these kind of pictures - and for the most part, those schedules are always getting shorter, not longer - there are significant technical and artistic barriers to getting 2500 VFX shots done at the quality level that's expected at anything higher than that in the near future. That's the reality for large scale productions, not what people here seem to think it is...
Click to expand...

I'm from your side of things Mike generally speaking. And while I echo and acknowledge what tent pole style productions have done, especially before now, I don't think anybody can ignore productions that are having VFX work done in 4K at this point. And there has been 8K VFX work done for premiere productions even longer ago, though that's a rare and high budget bird.

I agree from a logistics and deadline point of view the whole 2K and upscale thing has worked out, but that's clearly not what a good portion of the industry is doing these days. And that's growing very rapidly.

It is a peculiar world where one hour dramas with tighter turnaround than features are handling 4K VFX, yet the typically 3-18 month post VFX schedule can't handle it on features. That literally won't be able to go on much longer.

I'm in no way disagreeing with you, but it's pretty interesting to see where, what, and why a lot of those barriers exist and where they have been broken through. Especially in 2018.

Even at my time at R&H as well as doing work for ILM, DD, Weta, etc throughout the 2000s the vast majority was 2K, but there were jobs that came through and worked on in 4K every now and again as well as a smidge higher resolution. That was a decade ago.

Yes, 2500 shots is a lot to handle through large pipelines with high resolution assets, but major houses can and have been doing it occasionally. Storage and bandwidth are not where they were a long while ago, same can be said for full raster rendering. We're not exactly waiting a day per frame these days.

I would question hard for larger productions at a minimum not working in 4K within the next two years. Very hard.

Smaller houses are gaining ground in this way as it's been easier for them to integrate higher resolution workflows into a studio of sub-200 people. Beyond that, the big houses are going to need to adapt and studios shortly are going to demand it for future work. It's already happening in a rather staggered way.
 
Big surface area makes more sense for film. I think we're trying to impose old paradigms onto new ones. Who among you would prefer the Alexa 65 (which delivers a fine image) over the Helium or Monstro? You either want progress in digital or you don't.

Of course, a sensor that's too small makes no sense - look at how awful GoPros are. OTOH Apple has some magic going on with the 2018 iPhones. We'll see.
 
You must log in or register to reply here.
Back
Top