CSI: Colorspace 35mm / 16mm Digital Cinema Cameras

[Wayne Morellini]

Pity, that would not put them in good stead against the Red. I think the majority of people that would like to buy an extra quality HD camera don't want to pay more than 10K for it, and maybe a majority of them far less than that.

It is a shame everybody gave up on a low cost, machine vision, open sourced, camera once the first commercial projects were announced. I would be happy if it turned out different, but I don't think any them are.

[Levan Bakhia]

Well, even thou I am number 178 in RED list, I still am looking forward to TRUE35.

[Jason Rodriguez]

It is a shame everybody gave up on a low cost, machine vision, open sourced, camera once the first commercial projects were announced.

Wayne, how low cost is low cost??

If you mean $3000-$5000 for the camera head, recording software, lenses, etc., then I think you have no idea how much even a cinema-quality sensor like the 3570 or whatever Colorspace and RED are using costs just by itself.

I can't tell you specific numbers (of course), but just assume that the prices of good sensors are not even remotely "cheap" (and that "cheap" would still probably not be "cheap" to you) unless you're buying in quantaties of 10,000. If you buy in lots of 100, you pay very hefty premium. The only sensors that are "cheap" right now in the megapixel market are slow cell-phone and camera sensors (not fast enough for full-motion, high-def video).

Now take the price of the sensor, and add on all the engineering that goes into making a high-quality, noise-free camera head. Then add software (or firmware). Then the price of engineering and manufacturing the enclosure (go get some prices on CNC, sand, or die-casting). Then all the electronics and firmware for on-board recording and image-processing. Then consider in the end, nobody's going to be purchasing 10,000 of these things to keep the costs low like consumer/prosumer Japanese HDV technology.

So anyways, just keep in perspective on what is "low-cost" right now . . . you're looking at start-ups here (well, we're not a start-up, but the SI-1920HDVR is our first cinema-directed product) who are making first-to-market products . . . and for what you're getting, they are a VERY good deal.

[Wayne Morellini]

Originally Posted by Jason Rodriguez

Wayne, how low cost is low cost??

If you mean $3000-$5000 for the camera head, recording software, lenses, etc., then I think you have no idea how much even a cinema-quality sensor like the 3570 or whatever Colorspace and RED are using costs just by itself.

Jason, you know I have an very good idea, but you have no idea of what I'm talking about, and this market reality is coming.

I shouldn't answer anymore of this:

not be "cheap" to you) unless you're buying in quantities of 10,000. If you buy in lots of 100, you pay very hefty premium. The only sensors that are "cheap"

Exactly, the finale camera price is proportional to how many you plan to sell. So, putting away big cinema camera preference, and realising cinema means a lot less than an Altasens (and I do take it that pricing has increased over the previous sensor model) I am interested in something better for the average HDV professional, and everything cheap or not is relevant to the best 'valued' competitor, and that one we know and who is behind it (apart from the DVCPROHD quality Micron Elphel experiment currently going on in Alternative Imaging). I am more interested in something at least the quality of the XDCAM HD 1/2inch, that has cinema use, but is also primed for normal video jobs, documentary and and Eng. This subtle dual use has much more market potential. There was a death to open source plus $2K of of the shelf pre-designed pre-marketed volume equipment, but that is credible. And there has been a few more commercial projects lurking out there, then you may realise.

To the rest, $500, different cameras for different levels of quality and market segments.

Now back to the issue, the lower market segments rather than everybody battling it out with too many competitors, against a Megalithic competitor, at once for a limited market space. Unless you have a camera to match it's appeal, or outstrips it in some needed way (SN, Range, portability, handling, use for non cinema purposes) can be not good for business. If the megalithic competitor was a steep double the price, then $20K for the best 1080p camera would be very different in terms of value (please note, out of courtesy, I am not naming names, except for the one I did so you can find the thread). My point was that (even though for everybody that can afford these cameras it is good value) for the rest of us a pity.

Give True some room to decide to do whatever they are going to do. I personally, would like to see everybody succeed in their own space. I prefer to sit in the sidelines for now and encourage them into areas that I know they can be the most successful, and see what happens.

Re-edit:
I forgot to add, for "low cost", I always estimate on suing low cost quality secondhand SLR lens through a low cost triplet relay lens). Talking about normal cinema lens and low cost are not the same thing, they are a luxury extra.


Good luck Adam.

[Wayne Morellini]

Ironically, I had some links from a post open in the background, they say everything:
http://dvinfo.net/conf/showpost.php?...2&postcount=68
http://dvinfo.net/conf/showpost.php?...postcount=2071

They are from Rai, the professional engineer from our Cinema Camera projects that made more than one Drake camera. Whom I have had a bit to do with behind the scenes. We both think a like, and he says virtually exactly what I would say in the links above. You also notice the original prototype they used to film the movie has PC ports out the back, before they went to the FPGA path, so he did both a quick software and after hardware versions. I also had casual input 'guidance', into another completed cinema camera project who are doing what I advised them to do.

So I don't think I have no idea about these things, hanging with people in the computer world, in times past, that could do a non-production prototype cinema camera (it works) in two weeks to two months, rubs off (actually, two days might be possible, but I'm not that good to be able to confirm it).

[David Mullen]

I know the answer is "marketing"... but when is a 2K Bayer-filtered sensor equal to 35mm in any way other than the size of the chip, and since when is something less than 1920 x 1080 resolution equal to 16mm other than the size of the chip (especially considering that a Super-16 negative is slightly physically larger than a 2/3" sensor)?

Not to be argumentative myself, but calling something "True35" is only going to provoke a lot of arguments at the NAB booth, mainly with the people who already were skeptical that a 4K Bayer-filtered camera was truly "4K" and thus really equivalent to 35mm resolution. Now they are being told that 2K Bayer-filtered images are "true" 35mm equivalents?

It's the same problem I have with the 4K "Phantom 65" camera, which has a 65mm-sized sensor but 35mm resolution (again, assuming you want to say that a 4K Bayer-filtered image is equivalent to 35mm resolution) -- yet the size of the sensor is allowing the company to make some sort of implied equivalency to 65mm resolution!

I think the use of the word "true" in the name of the camera is unfortunate.

It's funny because today I asked an Arri person why the D20 is called that, and he said it was because, with a sensor that was about 2800 pixels across (more than 2048), they felt it was somewhere between 16mm and 35mm in terms of resolution, even though the size of the sensor is 35mm -- hence the number "20" as if it were a fictional 20mm film format in terms of resolution.

You would think that the 2K Bayer-filtered "True35" camera was the one closer to the "true" resolution of Super-16.

Please rethink the use of the word "true" in the name of the camera if you don't want to anger a lot of the cinematography community, unless the product isn't aimed at professional cinematographers.

[David Mullen]

I just want to reiterate that I recognize that 2K and HD are used all the time nowadays for cinema release material with excellent results... my objection to the name "True35" is merely that it is inaccurate in that it implies equivalency with the resolution of 35mm color negative origination.

[Adam Burtle]

Below [this paragraph] is a copy and paste of my reply to this question on the CML (where it was sortof cross-posted, and I happened to reply there first-- although CML emails take a while to show up in everyones' boxes). Just to add a bit to what Matthew said-- as I've said many times, all these different products (ours, Sony's, etc) are just tools for the toolbox.. ultimately the story is what matters. While I am not a fan of tools being built to progressively lesser and lesser standards (we should be going upwards in quality as well as upwards in flexibility), ultimately resolution matters a lot less than the story itself. I can think of plenty of awesomely shot films where I have gotten lost in the story and completely forgotten they were shot on MiniDV, Beta, etc. If a tool lets a cinematographer tell a story that otherwise wouldn't be possible (because existing tools are too expensive, too heavy, can't record uninterrupted for long enough, whatever)-- that is what is most important, in my opinion.

(edit-- I said "would" when I meant "wouldn't" ...edited to be correct now)
---

Mr. Mullen-- I suspect you're reading a bit too much into what was meant to just be basic marketing. The point of the "True35" name was simply to connote that you can work with the camera in a way more "true" to how one would work with film. For instance, offering (basically) a 35mm frame size, in 4:3 aspect, so you can choose to hard or soft matte, etc. The name WAS NOT meant to imply that digital [in a 2k bayer implementation] is a match for 35mm or even 16mm film. Every format has its strengths and weaknesses. In my opinion, 2k bayer is pretty good, and if you preserve the image fidelity throughout the post chain, what you see in a theatre can be pretty comparable to 35mm film in a theatre (if non-DI generational loss is taken into account). Again, that isn't to say that a bayer 2k digital "negative" is going to have all the same data as an equivalent 35mm original film negative-- it won't. Part of the reason I liked the idea of a 24x18 2k sensor was the incredible pixel size (10+micron), which means better dynamic range and sensitivity to light. It's not that we don't think 4k is better... but the data wrangling issues inherent to 4k, plus the reduced dynamic range of a greater pixel density, etc.. led us to feel that 2k was a "sweet spot" for 35mm digital, based on where the industry was at the time (and basically still is).

As Mike Most pointed out, we've run into supply chain issues with some aspects of the camera, and we're re-thinking some of its development. Our development model has always included the idea of building products that filmmakers want to use, and we're taking this opportunity to improve the specs of the product to make it more desirable to cinematographers. I know our project doesn't get the attention of some similar digital cinema ventures; we can't pretend to have the same level of marketing machine, but at the same time, we really are working to build products that the users want-- and since the beginning have been listening directly to anyone who will offer us input or ideas. Sometimes this input is a reassurance, and sometimes it feels a bit critical, but either way its valuable to us. I think we now have a pretty interesting sensor-design answer to overcoming the limitations of bayer2k, but that's the extent of what I can say publicly. If you (and anyone else) want to offer your input, I'd be happy to speak privately.

The recorder end of the camera system has progressed with much less inhibition, and we'll be at NAB talking about where we are in development, and what we can offer people. A while ago we decided to develop the recorder to work with as many cameras as possible, rather than locking it into only working with our camera-- again, the idea of a modular platform that is just another tool to the user; I don't want to coerce people into using our products.. if the ICON recorder is the right tool for a particular user, but only when attached to another company's camera, that's fine. The recorder is designed to be a "toaster-sized" on-board 4:4:4 recorder with swappable media packs. I mention the recorder here to reiterate the idea that we're trying to empower filmmakers-- right now there are some good data-centric recording options on the market, but I feel that they limit the flexibility of the filmmaker by requiring a tether to a heavy box.

I know there's a lot of marketing hype in the industry (talking about digital being as good as film, etc)-- it's definitely easier to hype a lesser product than to develop a better one ;) ...but I want to make the point that the name of the cameras wasn't to hype a lesser product, or try to claim that digital matches film, or even that digital should try to replace film. The names were innocent enough in their origin, and we simply wanted to imply that we'd tried to take some cues from the film making paradigm. In my opinion, many digital products force filmmakers to adapt the techniques in ways that are unnecessary, and the, "true" name was simply a way of trying to convey that we were working to stay "true" to some of the elements of the film paradigm. I apologize for any confusion, and for what it's worth we'd been considering going with a new naming scheme, since the development path has changed so much ;)

[David Mullen]

Thanks for your reply, Adam. Your explanation makes sense. I still think that the word "true" is a loaded one that you may wish to avoid, just to save yourself pointless semantic debates with nitpickers like myself! Good luck on the camera; I'm always looking forward to any 2K/4K digital camera that will allow me to use standard 2X anamorphic lenses to get a scope image.

[Wayne Morellini]

Originally Posted by David Mullen

It's funny because today I asked an Arri person why the D20 is called that, and he said it was because, with a sensor that was about 2800 pixels across (more than 2048), they felt it was somewhere between 16mm and 35mm in terms of resolution, even though the size of the sensor is 35mm -- hence the number "20" as if it were a fictional 20mm film format in terms of resolution.

You would think that the 2K Bayer-filtered "True35" camera was the one closer to the "true" resolution of Super-16.

Please rethink the use of the word "true" in the name of the camera if you don't want to anger a lot of the cinematography community, unless the product isn't aimed at professional cinematographers.

I think there is a clash of cultures here. To cinema people 35mm might imply much more then it means (the 35mm cinema size) for budding digital people around here, it means it's original connotation, the frame size. There is an common movement around here, in making 35mm SLR (and film) lens adaptors for video cameras, which make very film like images with shallow depth of field and larger latitude, worth checking out. Maybe the camera should be viewed as equivalent to an true 35mm film camera with an more grainier, lower resolution, 35mm film stock.

Adam, it doesn't look like I will be working on that other cinema camera, seems to have disappeared, so feel free to email me.

[Tyson Perkins]

what are we look at price wise for this cam?

[Brian Drysdale]

Originally Posted by Wayne Morellini

I think there is a clash of cultures here. To cinema people 35mm might imply much more then it means (the 35mm cinema size) for budding digital people around here, it means it's original connotation, the frame size. There is an common movement around here, in making 35mm SLR (and film) lens adaptors for video cameras, which make very film like images with shallow depth of field and larger latitude, worth checking out. Maybe the camera should be viewed as equivalent to an true 35mm film camera with an more grainier, lower resolution, 35mm film stock.

.

The resolution is a key part of using 35mm: it gives you the power of the wide shot. Unfortunately, this gets lost if you're using formats with less resolution. Shallow DOF is only a part of shooting 35mm and this element can change with fashion or the demands of the story.

[David Mullen]

The problem is the "true" moniker -- it's a loaded term and I think should be avoided. It suggests things like "it's truly 35mm!", "more like 35mm than previous digital cameras!", etc. If it were called the "Colorspace-35" it would be a little less misleading/confusing/controversial than to call it "True-35". The addition of the word "true" suggests something much more than just a match in the size of sensor.

I mean, who wouldn't want to use a digital cine camera if it were "true" to being like 35mm as the name implies?

These sorts of new digital cine cameras will be used by both film and video people, so why piss off half your market with a loaded name like that?

[Charles Papert]

David:

Respectfully, don't we have enough issues to worry about as it is without being concerned over the semantics of a product name? Will it really piss off the entire film market, or just those who like to gnash their teeth on certain cinematography message boards (that's not aimed at you, by the way)?

The industry has a long history of hyped-up monikkers (what exactly is so Ultra about Ultravision?!). For me, if a digital camera is able to deliver the goods, I don't care if it's called Red, Pink, Chartreuse, True35 or Howdy Doody Cam...but then again, I don't hang out on the aforementioned forums.

That all said, I think that you are doing these folks a service by alerting them to the fact that our industry does tend to be skeptical of newcomers, and it's probably a useful bullet to dodge if at all possible.

[David Mullen]

Obviously if the camera is good, I don't really care what it's called... I'm just trying to make them aware that they will be causing marketplace confusion with this name. It comes off as promising more than they can deliver since it won't be 35mm quality, but some poor souls will assume it with a name like "True 35".

Since there's no reason to cause this controversy unless they actually want to be misleading for sales reasons, I think they will be better off finding a new name. Otherwise they will be setting themselves up for years of being scoffed at by various cinematographers at trade shows, etc.

It's a similar problem (though less technical) to heavily advertizing a 4K Bayer-filtered camera as being "true 4K" as opposed to just saying that it has a 4K Bayer-filter sensor and letting people draw their own conclusions regarding the resolution. If the name of the RED camera was "the RED True-4K" camera, you can bet every other geek would be wanting to call them out on this. There's a point where being too aggressive in the marketing can backfire among professional cinematographers, who get more skeptical the more something is "sold" to them.

You and I may know this is all part of the marketing game, but some producer who hires us may be sold a bill of goods by some clever marketer and we'll find ourselves trying to educate them, hence why these things matter.