Thoughts regarding S35 vs. FF35 sensor size

[Esteban Sosnitsky]

Did I miss something?
Can the camera "brains" be upgraded?

Sure... the last poster wanted to hold us to the free upgrade but that was with the old program... which we are happy to honor. Upgrading brains will be cheaper.

Jim

Also, may I quote Jim from his Recon post:


"We also said we would offer a RED ONE sensor upgrade to keep it current in a world of ever-improving electronics. We have kept that promise with a Mysterium-X upgrade path when it is finished, then to the Monstro sensor which will follow."

So I guess one could keep the R1 as is and be able to upgrade the sensor for a price, my question then would be, can the board and electronics of the current R1 handle such those sensors?

 

[Stephen Pruitt]

Geez, talk about Johnny on the spot! Estebanred found Jim's quote instantly!

Now, just because he said it doesn't mean that it will come to pass, but it is at least a hopeful sign that a S35 Monstro could be available for Epic in the near future.

As for me, I cannot imagine trying to deal with the depth-of-field with a FF35 chip. It's hard enough with the RED One's Academy S35!

Stephen

 

[Jason Ing]

Or somehow make the Mysterium-X system be 16-bit with more dynamic range, like the Monstro, since the size of the sensor is about perfect for S35 photography.

This would be ideal for people who are happy with their Red Ones, don't necessarily need the new modular system, but still wants the best image quality possible.

I personally think a lot of R1 owners would be happy about this upgrade path.

 

[Stephen Pruitt]

I'd be deliriously happy with the ability to keep upgrading at least one of our RED Ones if it would yield a picture quality essentially equal to an Epic.

I honestly cannot imagine Jim wanting to leave the RED One in the dust. It was his baby, too many people have them, and he would be perhaps the first techno-geek in history to offer a true "non-obsolescence" promise and actually keep it.

A S35 Monstro would utterly dominate the market.

Stephen

 

[Esteban Sosnitsky]

Select?

Select?

As I was re-reading the announcement I came to find a "mysterium x select" in the spec sheet... what would the Select stand for???

 

[Alexander Nikishin]

Thanks for answering my questions Michael & David.......

The select as I understand it is sort of a way to signify that the Epic X sensors are as Jarred put it, "The best of the batch of Mysterium X sensors" allowing for Redcode 250 vs. 225.

 

[AIO Films]

i'm worried that red might be spending to much time working on unproven formats instead of just being ahead of every body else with current ones.
we are still waiting for the pl mount red primes, red are now going to make 8 more cameras and loads of different lenses with different lens mounts. it will all get very confusing.

And put the 18mm and 14mm T1.9 primes back in the PL lens series would be really helpful for those who can't afford Ultra Primes or Cooke S4 like me.

 

[Jason Ing]

As I was re-reading the announcement I came to find a "mysterium x select" in the spec sheet... what would the Select stand for???

I believe it indicates that is the mysterium x for the special limited edition. Note that it is the only sensor that has Redcode 250.

 

[Michael Hastings]

I'd be deliriously happy with the ability to keep upgrading at least one of our RED Ones if it would yield a picture quality essentially equal to an Epic.

I honestly cannot imagine Jim wanting to leave the RED One in the dust. It was his baby, too many people have them, and he would be perhaps the first techno-geek in history to offer a true "non-obsolescence" promise and actually keep it.

A S35 Monstro would utterly dominate the market.

Stephen

Stephen: I suspect the REDONE would be easy to upgrade all of the way to the same as an FF35 Monstro Epic. Since the REDONE is so much larger, you would just replace all of the processing circuit boards and the sensor - probably with room left to spare - but it wouldn't make much sense - like buying the epic brain and i/o module and removing the circuit boards and sensor and transferring to the REDONE. All you are saving by doing a full upgrade of a REDONE would be the external case - which doesn't save much and might be offset by the cost of making internal adapters to hold the smaller boards inside the REDONE. And it would still be an oversize clunky package compared to the EPIC package.

But what would be the point?

It's like upgrading your old Pentium III computer. You can buy a modern computer with latest processors, high speed DVD reader/writer, 3 gigs RAM, 250 gig harddrive, USB2, built in wifi, flash card slots, high speed video, etc. etc. for $299 from major vendors. Could you replace the motherboard and all that other stuff in your old case - yep but again all you save is the case.

In reality it is like replacing the old Pentium with a current much more powerful laptop and 15" screen for $399. You get all of the goodies, in a much tighter, nicer packaage.

 

[Stephen Pruitt]

Hi there, Michael. . .

Your point is very well taken, BUT. . . Jim has already announced an upgrade to Mysterium X for $4500. If Monstro is the same $4500 from there, things could really get interesting.

I'm not convinced that the modularity of Epic is going to be as good in practice as it seems just now. One thing is certain: That big block of machined aluminum is as solid as a you-know-what. I suspect that, for hand-held use, rigidity will prove to be more useful than some people think right now. Indeed, while the RED One is a very heavy camera for hand-held use, I'm quite happy with its stability. If I lifted more weights, I'd have no trouble using it at all.

But, you're right. . . we'll have to see what Epic looks like when it comes out.

Stephen

 

[Michael Hastings]

A $4500 Monstro upgrade would be nice, but (Total speculation here) I think the Mysterium is as far as RED will practically want to go. At some point you get to the limitations of the brain rather than the sensor, and beyond that I think RED would like to gather in the REDONE hardware, not so much for the size but for the fact that it is based on off-the-shelf FPGA hardware which allows for greater flexibility in developing the software upgrades and capabilities. But in the long run you are kind of stuck with the high power draw, long boot up times, etc rather than using custom designed ASICs - like Canon does with their Digic !V chips - so you can get near-instant bootup, much better power consumption, and probably better physical reliability because you have a single chip doing the job of an entire complicated processor board that is more likely to develop problems in the rough and tumble of field use.

I have to believe that this is the direction RED has gone - they have the money and the volume to make it worthwhile, and the REDONE has been the proof of concept that should allow them to bake a lot of the functions into silicon rather than RAM.

 

[I Bloom]

A $4500 Monstro upgrade would be nice, but (Total speculation here) I think the Mysterium is as far as RED will practically want to go. At some point you get to the limitations of the brain rather than the sensor, and beyond that I think RED would like to gather in the REDONE hardware, not so much for the size but for the fact that it is based on off-the-shelf FPGA hardware which allows for greater flexibility in developing the software upgrades and capabilities. But in the long run you are kind of stuck with the high power draw, long boot up times, etc rather than using custom designed ASICs - like Canon does with their Digic !V chips - so you can get near-instant bootup, much better power consumption, and probably better physical reliability because you have a single chip doing the job of an entire complicated processor board that is more likely to develop problems in the rough and tumble of field use.

I have to believe that this is the direction RED has gone - they have the money and the volume to make it worthwhile, and the REDONE has been the proof of concept that should allow them to bake a lot of the functions into silicon rather than RAM.

That's an interesting suggestion Mike, because essentially what you mean is that the ongoing firmware upgrades of RedOne will not be a part of the DSMC program. If it's true... that is extremely important.

 

[Tom Lowe]

Assuming that a person can afford the additional $3,000 required for the jump from S35 Scarlet to FF35 Scarlet, the question to ask yourself is: resolution or dynamic range?

Let's face it. If you're spending $50,000 or whatever on a new camera system, is three grand really going to make a difference?

I think people should wait as long as they can before buying anything. By the time FF35 Epic comes out, there might already be an announcement about an S35 Monstro camera. You can decide what you want to do at that point. I for one would not even consider the Epic X, unless I was very busy with work and could pay it off pretty quickly. But even then, I would sell the brain on ebay or here as soon as FF35 Epic came out. The DR and larger sensor are of monumental importance to many of us.

My guess is that all the people who are skeptical about the FF35 format now will not be skeptical for long, especially as lens options begin to get sorted out. FF35 will move into the mainstream sooner than many realize, and if you are sitting on an S35 Epic, you might be wishing you had waited for the flexibility of FF35. In the meantime, R1 will continue to kick ass and take names.

Think about this: With FF35, we are talking about a 6K RAW digital Vista Vision camera. This is something cinematographers could only have dreamed about two short years ago. We should embrace this new format, IMHO, and do everything we can to signal to the marketplace in advance that we need quality FF35 cinema glass, pronto. As many here have said, some of the high-end "S35" glass, like Master Primes, already cover a fairly large portion of the FF35 frame.

A lot of people like to say that Jim is a visionary, but IMO, his embrace of FF35 is actual living, breathing proof of his visionary outlook.

 

[Stephen Williams]

I think people should wait as long as they can before buying anything. By the time FF35 Epic comes out, there might already be an announcement about an S35 Monstro camera. .

Hi Tom,

Probably even the replacement for Monstro if you wait long enough.

Stephen

 

[Tom Lowe]

Quit being a smartass. :tongue:

I think FF35 Epic and Scarlet will come at just the right time, with just the right specs. Even little red riding hood might agree.

 

[Joe G.]

I'm in disbelief of all this apparent complaining about the FF35. This camera will take ANY lense you want to put on it. It will allow S35 frames all day long with 2 more stops of light recorded. It will also allow full frame images at even higher resolution.

As long as it has a windowed mode, there is no basis for complaining, whatsoever. This is the obvious choice (to me anyway). There is no need for Red to churn out yet another model, when everything is already covered.

 

[Jeff Kilgroe]

mail4joeg wins a cookie. It's been said before, but it doesn't hurt to say it again...

But there is one more solution for those who think they must have an S35 sized Monstro -- Keep your RED One! Jim has stated that there will be an upgrade path for the RED One for the Mysterium-X sensor and later on, the Monstro sensor. Perhaps that is the way to go... Of course, it seems it would be a compromise in which REDCODE levels will be recordable, but I think people are so focused on the newly announced cameras and no one is thinking about the viability of the current RED One and how future upgrades will take place.

RED is not dropping the RED One, they are continuing to support it. It is positioned in their lineup as a camera between the Scarlet and Epic, somewhat of an all-in-one camera without all the modular flair. The One will offer many features that Scarlet will not like ramping and various frame frame sizes, etc.. It will keep its S35 sized sensor, which will eventually be available in the form of Monstro. The catch is that it's not going to have much of the same internal processing power and bandwidth that Epic will have, so that will translate to higher compression ratios. But a good compromise for a camera that fits right into the middle of their line-up. At times I wonder if people are being too eager to trade in their RED One for the "next best thing" when in actuality, the best camera for their needs may already be in their possession.

 

[roryhinds]

while the RED1 takes great images the ergonomics are poor.
In a handheld situation RED1 is awaked to work with.
Its interesting to remember RED marketed the RED1 as being modular to.

The EPIC & Scarlet systems seem to have address ergonomics with their small form factor. Having a camera system that locks together instead of bits hanging off it is worth the upgrade for sure.

I'm looking forward to see what RED bring out with Epic & Scarlet and how it works ergonomically.

 

[David Rasberry]

I'm not sure it is an impossibility -- the Dalsa had a 14-bit A/D processor for their slightly-larger-than S35 sensor. I wouldn't be surprised if the next ARRI digital cine camera had a 16-bit processor with an S35 sensor.

Bit depth and dynamic range are not exactly the same thing. Each pixel of a sensor, whether CMOS or CCD, puts out a linear analog voltage in response to photon density accumulated over exposure time. The DR of the sensor is defined by a logarithmic ratio of maximum voltage response to inherent thermal noise of the sensor and AD circuits. It is an expression of contrast range.

Bit depth determines the number of discrete value levels encoded by the AD converter over the total available DR.

Discrete shades of gray by bit depth:

8 bit = 256
10 bit = 1024
12 bit = 4096
14 bit = 16384
16 bit = 65536

Greater bit depth gives a broader color gamut and a more subtle resolution of details within the available DR.

 

[JonathanF]

I thought it was a Mysterium-X sensor that the RED ONE could be upgraded to, not a Monstro sensor.

I would also prefer a 16-bit S35 Monstro sensor option for the EPIC, preferably 5K, with a 24mm wide area used for picture information, plus a lookaround area, so perhaps a sensor that was 28mm to 30mm wide. I'm sort of assuming that whatever the source is for the Monstro EPIC sensor, it can't accommodate the S35 size limitation for some reason, maybe because it's an adaption of an existing FF35 DSLR sensor. Or somehow make the Mysterium-X system be 16-bit with more dynamic range, like the Monstro, since the size of the sensor is about perfect for S35 photography.

If someone built 2X anamorphic lenses that would cover the 645 sensor area, then there would be some advantage in getting a 2.40 image since the sensor is 4x3, but it seems a better route would be to design some 1.6X anamorphic lenses to cover the 1.50 : 1 FF35 sensor... IF you are going to build new anamorphic lenses in the first place.

Depth of field will be challenging enough with a spherical lens on a FF35 sensor, even harder with anamorphic, and even harder with 645, and even harder with anamorphic on 645...

David,

I could not have said it better. For my $ I'll take additional bit depth and more dynamic range over the additional resolution above 5k any day of the week.

Most of us shooting for theatrical release are going to be mastering to the DCI spec soon enough anyway and for all practical purposes and that's a 4k standard so while it'll be nice to have a 5k camera master I'm pretty happy working 4k and will be for the foreseeable future. In fact, two of the last 3 films I worked on had full DI and the OCN was scanned at 4k and then downrez'd to 2k and output from the recorders at that resolution.

1:1.85 is losing the advantage of the available negative real-estate anyway so the actual resolved resolution of a 1.85 release is already far below 5k in my book after a trip through DI and a 10 bit log grading post process and recording back to film. Thankfully most of the reputable effects houses are doing everything after the scans in floating point anyway so we don't loose too much there anymore given the increasing number of VFX shots in our films.

People forget that a 1:1.85 release print uses only 0.825" x 0.446" of the _available_ area for a Normal 35 Academy aperture at 1:1.37, 0.8661" x 0.6299".

People can go on all they want about resoution, but the real gains in terms of image quality in my book are going to come from improved bit depth and better processing. At 24 fps there's only so much we're going to gain that audiences are going to perceptually appreciate with added resolution, especially given the nature of commercial projection these days.

I'll take stops over more rez at this stage without any question.

One other thing worth noting is has anyone here looked at the price of 62mm Maxi PL glass lately? That's what you're looking at cost wise for good Cine optics at these larger sensor sizes people are going on about.

As a sidebar, I once held a Super Panavisoin 70 lens and I'll tell you now, I don't want to be schlepping a bunch of those around anytime soon... glad we don't use those things anymore (kind of).

You go to these larger formats and people forget that, just like the transition from 16 to 35, it involves a lot more than just the camera.

All the support equipment scales as well and no one's even discussing the availabilty of key camera support equipment for all these wonderful new formats.

My $0.02 for what it's worth.

J