The FF35 system

[Daniel S Williams]

and you guys are forgetting...after debayering none of this is true 4k or 5k or anything anymore

 

[Stephen Williams]

and you guys are forgetting...after debayering none of this is true 4k or 5k or anything anymore

Look at pictures, resoloution no's are unimportant IMHO.

 

[Joe Walker]

I remember making these exact same points over at cinematography.com over a year ago regarding digital 65mm, Vista Vision or IMAX, and those guys tried to laugh me out of town. Hahaha. They aren't laughing any more thanks to Jim! :wink:

Uuuhhhhhhh.......

 

[Daniel Browning]

Yes, ISO/ASA ratings need to come up a tad to allow for clean DOF control, but that is a given. The new DSLRs like the D3x and 5D2 have seen absolutely massive jumps in light sensitivity,

There have been big reductions in read noise, but sensitivity improvements have been very modest. The D3X and D3 sensitivity is only 15% greater per area than the Canon 1D3, 1Ds3, and 50D. In fact, compared to the old Canon 10D, the D3/D3X is only 1 and 1/3 stop more sensitive.

The reason improvements have been so modest is that we're already near the theoretical limit of physics. Sensors like the D3 are seeing very close to all of the light falling on them (not counting the color filter). There is no way to see more than 100% of the light; you can't see something that isn't there, so sensor designers have little to improve in that regard.

Almost every camera built in the last 4 years is within 1/3 stop of eachother when it comes to sensitivity. What sets them apart in low light is read noise.

Unlike sensitivity, read noise has improved very rapidly. The D3, 5D2, and most other cameras have made improvements of as much as one stop or more compared to previous cameras. The D3X and D90 have made a huge improvement in low ISO read noise, but their high ISO read noise is not as good as the D3 or 5D2.

Low light photography will be improved by reduction in read noise.

so there is absolutely no doubt whatsoever that DPs will be able to stop down enough to control DOF to their liking and their focus-puller's ability.

No need to wait for the future. That has been true for the last five years. Cameras on the market now already prove the fact.

The last hurdle, of course, is glass. Glass will control how fast the move to FF35 takes place.

Agreed.

Looking at the Canon 5D, it's clearly possible.

You don't need to wait for any future possible improvements; it has been a fact in still photography already for at least the last five years. I went ahead and started a new thread about it:

Myth busted: larger sensors have thinner DOF.

 

[Tom Lowe]

Look at pictures, resoloution no's are unimportant IMHO.

Well, if your goal is a True 4K finish, then resolution obviously plays a major role. You sure as hell are not going to get a true 4K finish shooting on an F900 or Viper. To get True 4K from a Bayer-pattern camera, you really need a 5K sensor. On the other hand, as you say, you need to look at the image. If a 4.3K Monstro camera can produce a better image with better DR and color and less noise than a 5K Mysterium X can, then resolution becomes a less-important factor in this S35 vs FF35 Epic/Scarlet debate.

 

[Stephen Williams]

Well, if your goal is a True 4K finish, then resolution obviously plays a major role. You sure as hell are not going to get a true 4K finish shooting on an F900 or Viper. To get True 4K from a Bayer-pattern camera, you really need a 5K sensor. On the other hand, as you say, you need to look at the image. If a 4.3K Monstro camera can produce a better image with better DR and color and less noise than a 5K Mysterium X can, then resolution becomes a less-important factor in this S35 vs FF35 Epic/Scarlet debate.

Hi Tom,

My goal is beautiful pictures, Benjamin Button shows the Viper is a totally capable camera that has been around for 5 years. 27,6000,000 pixels at the capture head has a lot to do with it, even if only 1920 x 1080 are recorded.

Stephen

 

[Tom Lowe]

Hi Tom,

My goal is beautiful pictures, Benjamin Button shows the Viper is a totally capable camera that has been around for 5 years. 27,6000,000 pixels at the capture head has a lot to do with it, even if only 1920 x 1080 are recorded.

Stephen

I didn't say anything about the Viper being anything other than a great tool for filmmaking. I have been very impressed with basically all the big features shot on it, including Button, Zodiac, etc. I simply stated that Viper cannot be used to acquire True 4K, any more than an DVX100 can be used to acquire True 1080p. Red One cannot even acquire True 4K.

 

[Tonaci Tran]

Hi Tom,

My goal is beautiful pictures, Benjamin Button shows the Viper is a totally capable camera that has been around for 5 years. 27,6000,000 pixels at the capture head has a lot to do with it, even if only 1920 x 1080 are recorded.

Stephen

Stephen,

Do you think Viper is more capable than the Red One?

Tonaci

 

[Häakon]

You want Monstro, you buy the FF35 model for $5K more (yep, that's it) and you gain all the added benefits.

The base price of the FF35 camera is only $5K more, true - but as a great deal of us are RED ONE owners already, it means we get a free CF module, battery module, I/O module, and controller unit with the Epic X. I don't know how much all of that is going to cost, but likely another few thousand dollars. Slap some tax on top of that, etc... you're easily in the near $40K range for a comparably packaged monstro camera. That may or may not be a concern to you, but it should be pointed out.

Also, there's also the fact that Monstro debuts later than the Mysterium-X, and with their "Winter 2009" estimate, it's my guess that you're looking at least a year from now (or more) down the road before the FF35 debuts. If the Epic-X comes out in the summer, that could make another big difference in someone's plans. Of course we all want the best camera possible, but we also want to be shooting now and not waiting around for "the next best thing" to come down the line. Also remember that you can only use your RED ONE trade-in once, so if you choose to get an Epic-X it may be a lot more difficult to come up with $40,000 for the Monstro when that time comes. My guess is the market for a used Epic-X will be eroded when "the next best thing" is available.

Just food for thought.

 

[Häakon]

and you guys are forgetting...after debayering none of this is true 4k or 5k or anything anymore

Completely agree. Resolution numbers may not be important to some people anymore, but it still is to me. For instance, the 2K mode on the RED ONE is completely unusable for cutting in with 4K work in a 1080 timeline in my opinion. The difference is very drastic and noticeable.

To me, cropping the FF35 sensor to get a s35 image is a total waste of the chip - but judging by people's comments, my guess is that's how the majority of shooters are going to be using this camera. I'm still boggled as to why they don't just create a slice of the Monstro sensor in a s35 size. Yes, the Mysterium-X variant would indeed have a modest resolution advantage in comparison to a s35 Monstro, but I think most people would trade that for better dynamic range. As Jeff has pointed out, it would leave such a camera at around 4.2K - and while I'm in the minority camp who don't feel that this is "enough" resolution moving forward, I feel that it's probably the best compromise/solution of the offerings that are being made available right now. And sure, one can get effectively get that result by slapping a PL mount on the FF35 camera, but as my previous post pointed out, you're probably going to have to spend an extra $10K and wait around a good deal of time for it - and then you're still wasting half the sensor. Does no one else think this is rather silly?

 

[Jeff Kilgroe]

The base price of the FF35 camera is only $5K more, true - but as a great deal of us are RED ONE owners already, it means we get a free CF module, battery module, I/O module, and controller unit with the Epic X. I don't know how much all of that is going to cost, but likely another few thousand dollars. Slap some tax on top of that, etc... you're easily in the near $40K range for a comparably packaged monstro camera. That may or may not be a concern to you, but it should be pointed out.

Good points... It's true we don't know how much the modules will cost. And the EPIC-X including CF, battery mod, I/O, etc.. plus some extra features at no additional cost, is a pretty sweet deal. I guess I was just looking at it from the perspective of outright purchase cost. I'm not sure just how, when and where I'll use my RED One trade-in just yet. My gut feeling is it will probably go toward EPIC-X, because like you say, I'm expecting the Monstro models to arrive at some much later point down the road.

As for FF35 vs. S35... I see your point in that many users will end up using mostly S35 glass and will never utilize the full sensor. But then again, there are advantages even with PL mount such as increased sensor area for shooting anamorphics. I think what we're seeing here and with what has been said by Jim / RED is that while S35 monstro may be desirable, it just doesn't seem to have any practical advantages on their end. In other words, they're going to sell you the FF35 version as a bit of over-kill because if they produced an S35 model, that is one more brain model to support and production costs would be the same as the FF35 version anyway.

All I do know is that there are a lot of people wanting Monstro in the S35 size. I think if there's a way to make it happen, RED will do this. They listen to their customers. But I'm still betting it won't happen, personally I don't see a need for it to happen. When there's only a $5K difference between S35 Mysterium-X EPIC and FF35 Monstro EPIC, you have to wonder just where the bulk of that $5K cost difference really is. What is involved in producing an in-between model other than a different sensor size. What do most people wanting the S35 Monstro hope to achieve by getting RED to offer such a thing? Are they hoping to save some imaginary and arbitrary amount like $2K or are they doing it just because they feel it's "silly" to buy a camera with a bigger sensor than their glass will cover? Or is there something else that escapes me entirely? I've heard and read all kinds of misinformation and FUD about it, like "there's a loss of quality when you crop" and other BS like that. Are myths like that really widespread? ...I guess I'm just trying to understand what drives the thirst for an S35 Monstro EPIC over the FF35 model. Right now, I think 99% of all people requesting it are doing so thinking it should save them a bit of $$$, based on the smaller sensor == smaller price logic.

 

[Häakon]

As for FF35 vs. S35... I see your point in that many users will end up using mostly S35 glass and will never utilize the full sensor. But then again, there are advantages even with PL mount such as increased sensor area for shooting anamorphics.

No question about that, and I've stated many times that if your desire is to shoot anamorphically, the FF35 is the obvious choice. I'm approaching this strictly from a spherical-shooting perspective.

In other words, they're going to sell you the FF35 version as a bit of over-kill because if they produced an S35 model, that is one more brain model to support and production costs would be the same as the FF35 version anyway.

Well that shouldn't be the case, because they'd be cutting smaller sensors, which means they can stamp out a higher quantity per wafer as compared to the larger version. Also, a larger sensor area means that if you have any dead pixels/defects, you're throwing away a much bigger chunk of wafer real estate - which means you've just increased costs again.

What do most people wanting the S35 Monstro hope to achieve by getting RED to offer such a thing?

Well if there was a s35 Monstro, it would defeat the purpose for having the s35 Mysterium-X camera; my expectation that it would replace that model. And thus, you'd save $10,000 by getting that camera compared to the FF35 version - which is a huge amount of money if you're just going to crop the FF35 sensor for your shooting anyway. And this way you still get the same great dynamic range capabilities. Again, this is with spherical shooting in mind. But in essence, we're having to pay $10,000 more for a bigger sensor we aren't going to use. That's why I would personally want a s35 Monstro, anyway.

 

[Stephen Williams]

Stephen,

Do you think Viper is more capable than the Red One?

Tonaci

Hi,

The last time I tested the Viper had more dynamic range, no rolling shutter motion artifacts, less noise in the blue channel, less fan noise, quicker boot time, lower battery drain & the uncompressed images were better for color correction. As film still seems to have the edge overall, I find the question irrelevant as all are tools that can be used today. Horses for courses.

Stephen

Edit It was Tom that brought Viper into the thread, I would not otherwise have mentioned it, BTW Viper is blown away by F23.

 

[Tom Lowe]

Completely agree. Resolution numbers may not be important to some people anymore, but it still is to me. For instance, the 2K mode on the RED ONE is completely unusable for cutting in with 4K work in a 1080 timeline in my opinion. The difference is very drastic and noticeable.

To me, cropping the FF35 sensor to get a s35 image is a total waste of the chip - but judging by people's comments, my guess is that's how the majority of shooters are going to be using this camera. I'm still boggled as to why they don't just create a slice of the Monstro sensor in a s35 size.

Even if they did this, it would not improve the resolution for you. It would simply limit you from having the flexibility of FF35. If you are requesting Epic-X-style free modules on a cropped-sensor Monstro S35, I can understand the desire to save money on those modules, of course. But then you are getting into a situation where you might be waiting for a year and a half or more for the camera to even come out.

In the end, Häakon, my guess is that someone like you who keeps busy working a lot on cinema-style projects is probably going to want the Top Dog camera in the Red lineup, and that will be Epic FF35. It will, in fact, by the time it comes out, probably be the top cinema camera on the planet. Also, don't forget that there could be nice cine glass like Master Primes that could take advantage of a sensor area larger than S35, even if that means doing some trimming from the FF35 record area. Jeff mentioned the possibility that Epic FF35 might even offer a "custom" resolution, like, say, 5099x2560, that would fit EXACTLY to the measured capabilities of a given cine lens, so that you would not be wasting anything.

I think the place where everyone's combined efforts would be most useful right now would be putting early pressure on lens companies, like Cooke for example, to consider looking at FF35 cine glass designs right away. Either that, or to begin researching options for the rehousing of still lenses for cine use.

 

[Stephen Williams]

I think the place where everyone's combined efforts would be most useful right now would be putting early pressure on lens companies, like Cooke for example, to consider looking at FF35 cine glass designs right away. Either that, or to begin researching options for the rehousing of still lenses for cine use.

Hi,

A small magnifier would do the trick, some stop loss though.

 

[Häakon]

You're right on all accounts, and I guess I'm looking at a couple of different things at the same time. One is a wanting to keep pushing for higher resolution from camera manufacturers (ie, I do not believe 4K to be the be-all, end-all "holy grail") while at the same time stating that as the number of pixels gets larger, it can't happen at the expense of things like like dynamic range (which is just as (if not moreso) important to the quality of the final image). The only way this is realistically going to happen is if we break past the s35-size barrier, and that's exactly what RED is doing with the FF35 camera. Ultimately, I think that will be one hell of a system - I just think the lack of cinema grade glass to cover its entire image area is going to be its achilles' heel. And if that's the case, it's an awfully expensive camera to purchase when you're only going to use half of it. That's really what I was driving at. Believe me, I would love for someone like Cooke to surprise us with a brand new line of high-end, FF35 cine glass (heck, 645 glass for that matter...), but I don't think the rest of the industry moves as fast as RED. RED spoils us, and bless them for it.

 

[Tom Lowe]

There are a lot of industry heavy-hitters who participate on this board, many of them lurkers. If they start talking to lens companies now about this forthcoming FF35 camera system, it might result in something. Who knows. Given enough time, the market will jolt itself into action. But it's the transition time, waiting for new or rehoused FF35 lenses, that is the concern.

I think David mentioned that when Vista Vision was in its heyday, DPs used a real hodgepodge of still and medium-format glass on them. I cannot remember what he said about it exactly. Maybe someone who knows something about the history of Vista Vision and the lenses used with Vista Vision could help to give us all some background and perspective on this FF35 issue.

 

[roryhinds]

I don't see the S35 being a size barrier.
Its a industry standard that works really well with some amazing glass available now.

I just don't see the solution being reinventing the wheel because the new hubcaps don't fit.

 

[Jeff Kilgroe]

IMO, with the rumors circulating over Leica releasing cinema primes (I think it's all but actually confirmed but the specs), I'm hoping that they may provide some FF35 support. There's also the RED primes that are over-due. Perhaps there is an FF35 connection with those? I know it's a lot of speculation, but also there are some real possibilities here. At times I'm half tempted to go ahead and put the Nikon mount on my RED and get that set of modified ZF primes from Duclos. They seem to work really well and produce great images for a good price. And will cover FF35. Quite honestly at this point, if the upcoming RED primes do not cover FF35, I may pass on my reservation for them and continue to rent various primes until I see what FF35 offerings will be available for the upcoming EPIC / Scarlet systems.

 

[roryhinds]

yes the issue is a lot of investment has gone into S35 lenses which the industry is very happy with.

If you're thinking about buying new lenses then its tough to be caught in the waiting game while you see what new lenses get released if they ever do.

Red's own Prime lenses will only cover S35.
Working with still lenses in a cinema production is very problematic.

I think the problem is that RED's current sensor design doesn't allow for a S35 size sensor with more DR. They are being forced into making larger sensors even though most cinema productions will not use the full chip and will have to window down.

Having a Motion camera that can take stills is no new feature. I personally don't see how I'd really use this feature in a production. Sure maybe for my own personal use but not a paid professional production.

If I want to take stills I've have a professional stills photographer on set with a still camera, using an optical viewfinder and still lenses.

I can't see the still photographer taking stills using Cooke lenses and having to look at a LCD eyepiece.

There simply are different tools for different styles of shooting. Stills requires light lenses with an auto focus function.

I can see that RED are marketing that their motion camera takes better stills than the top end Canon or Nikon, but I see this as a testament to the quality of the motion pictures, not me wanting to replace my stills camera.

I have a feeling Mysterium X will yield better results for S35 production than Monstro. Its S35 @ 5k

Problem is we will have to purchase Mysterium X and wait for Monstro, then if it does turn out to be better we'd have to offload our old Mysterium X at a reduced rate as it will not be worth much with a better sensor on the market.

If RED would only offer a sensor upgrade on the Brains things would be so much easier for purchasing and future proofing. Having to buy a whole new Brain is not a upgrade, no matter how you market it.

REDONE is upgradable, for $4500 you can have a Mysterium X installed. I don't see why for a similar price you can't have a Monstro installed in your Mysterium X Brain.