|
|
This topic comprises 3 pages: 1 2 3
|
|
Author
|
Topic: Imax Forrest Gump ad
|
Bobby Henderson
"Ask me about Trajan."
Posts: 10973
From: Lawton, OK, USA
Registered: Apr 2001
|
posted 09-16-2014 03:53 PM
quote: Marcel Birgelen
I guess I didn't make myself clear. Sure, you need to be able to change lenses, as one lens obviously doesn't fit all. I was pointing at interchangeability between current 35mm lenses and some future "digital only" design with a longer focal length and possibly even other sensor dimensions.
Making a lens system work well with differing sensor sizes is, I think, a more difficult issue. It's going to involve some level of compromise. And if the different sized sensor is larger than a 4-perf 35mm frame or an 8-perf 35mm still camera frame it's going to need a different lens system to cover that extra negative area.
People using more affordable DSLR camera bodies deal with sensors that crop into the view of a 35mm still camera lens at factors of 1.3x or 1.6x. Those smaller sensors have smaller pixels, which can reduce image quality. Those smaller pixels and tighter pixel grid will do more to show off limitations in a given camera lens, like if the lens is a tad bit soft. Diffraction can start to take a toll. A VistaVision sized full frame DSLR sensor won't be as prone to those problems because of the larger pixels spread across a much greater view from the lens. Instead it will be more prone to vignetting and showing how good or bad a lens is near its corners.
quote: Marcel Birgelen
While I do understand the concerns for semi-pro gear, we're talking about pro cameras here that have traditionally had a price tags in the range of luxury cars. And while it's nice for the semi-pro market that high-definition gear is getting more affordable, there's no need to cut those corners in this high-end gear.
Yeah, but going the CCD route will add a LOT of cost to the equation. And if you're talking about adding prisms, beam splitters, 3 CCD arrays, etc., as well as having a much more versatile lens system, you're going to end up with one huge, bulky, very expensive video camera.
quote: Marcel Birgelen
Electronics manufacturers will start pushing 4K and content owners will see this as an opportunity to resell your existing content at a "remastered" quality so they will follow suit. This will push 4K further into consumer markets. Exhibitors that don't want to be stuck with "sub par" quality will need to upgrade. And after 4K, the next upgrade will surely be around the corner, maybe a focus on a broader color gamut, maybe 8K, maybe HFR...
I think the upgrades are going to go only so far. Aside from the problems with consumers growing ever more hesitant to adopt new things (like Blu-ray 3D, 4K UHDTV, Home Dolby Atmos, etc.) these higher resolution camera systems are going to be a tough sell to some movie production people.
Actors are very age/youth conscious. A higher resolution camera will do more to show any crows feet, wrinkles and other signs of aging that peek through all that makeup. A higher resolution camera could reveal more flaws in movie sets. Not everyone wants a much sharper image. Us movie fans do. I've been wanting a whole lot more movie productions to originate in 5-perf 65mm for the longest time, but that hasn't been happening. I think there has been more than the somewhat higher cost of shooting in 65mm preventing the use of 65mm.
| IP: Logged
|
|
Marcel Birgelen
Film God
Posts: 3357
From: Maastricht, Limburg, Netherlands
Registered: Feb 2012
|
posted 09-16-2014 05:25 PM
quote: Bobby Henderson
Making a lens system work well with differing sensor sizes is, I think, a more difficult issue. It's going to involve some level of compromise. And if the different sized sensor is larger than a 4-perf 35mm frame or an 8-perf 35mm still camera frame it's going to need a different lens system to cover that extra negative area.
No, there should be a consensus about a sensor size to avoid suboptimal use of lenses and cross-manufacturer compatibility. That shouldn't be based on legacy film formats, but on technical constrains, usability and costs.
quote: Bobby Henderson
Yeah, but going the CCD route will add a LOT of cost to the equation. And if you're talking about adding prisms, beam splitters, 3 CCD arrays, etc., as well as having a much more versatile lens system, you're going to end up with one huge, bulky, very expensive video camera.
Sony has been building 3CCD DVCAMs for years and those aren't any more bulky than the average film camera, quite to the contrary. Sure, they're probably not using 35mm sized sensors, but I doubt those cameras need to be much bulkier than current common "digital film" cameras. Many professional digital cameras still use a mechanical shutter, you can get rid of this component when using CCDs.
quote: Bobby Henderson
I think the upgrades are going to go only so far. Aside from the problems with consumers growing ever more hesitant to adopt new things (like Blu-ray 3D, 4K UHDTV, Home Dolby Atmos, etc.) these higher resolution camera systems are going to be a tough sell to some movie production people.
This reluctance is already fact of life. You already see this with computers, that have become a total commodity. Mobile phones seemingly still enjoy some kind of special status, but that won't continue forever either.
Most consumers weren't motivated to upgrade to 3D but practically all newer TVs support it one way or another. I don't see anybody getting all excited about 4K TVs either, but most people will get those upgrades "for free" when they buy their next TV.
Once Dolby cannot market their Atmos system as something special and novel anymore, many people will end up with something supporting "Dolby Atmos" and will maybe just install three speakers or use some kind of soundbar with "virtual surround".
| IP: Logged
|
|
|
|
|
|
Marcel Birgelen
Film God
Posts: 3357
From: Maastricht, Limburg, Netherlands
Registered: Feb 2012
|
posted 09-17-2014 01:50 AM
We're really drifting away here, but since Terry has already insulted everybody in this topic and we all know Digital IMAX is mostly snake oil and we've done this discussion many times over, we probably can do that. ![[Razz]](tongue.gif)
Well, the Sony CineAlta HDWF900R for example, uses a 3CCD design. Unfortunately, only with a 1080p resolution and yes, you're right, there doesn't really seem to be any production camera that uses a 4K or higher CCD sensor. There are a few 4K CCDs out there, most of them in the 60mm+ range, but they seemingly didn't make it into production camera designs just yet. Also, at that odd sensor size, it will be hard to get matching lenses.
If you look at the technical details, you'll see that CCDs are still outperforming CMOS in most cases. Also, what will absorb more light? A beam-splitter setup or a color filter? I guess the latter one will effectively absorb more usable light.
So, we don't see such sensors, primarily because the high-end CMOS sensors are most probably made in the same foundries as the lower end sensors. It costs an enormous amount of money in both research and initial investments in production lines, to introduce something like a production quality 4K CCD sensor.
Tweaking existing sensor designs without the need of new production lines is therefore cheaper and just like with e.g. TFT panels, the cheaper sensors, with smaller surface areas can probably produced from larger sensors with small amount of production defects.
quote: Mike Blakesley
That's me right there. We have two large TVs at home...we got the first one about 4 or 5 years ago, and the other one last summer.
My last TV also came with "3D" support, although it was being pushed as the next-big-thing a few years ago, I've only used it once ever since (I don't think that 3D on small(ish) screens really makes any sense) and now it's just an extra logo on the box.
| IP: Logged
|
|
Bobby Henderson
"Ask me about Trajan."
Posts: 10973
From: Lawton, OK, USA
Registered: Apr 2001
|
posted 09-17-2014 12:45 PM
quote: Marcel Birgelen
If you look at the technical details, you'll see that CCDs are still outperforming CMOS in most cases. Also, what will absorb more light? A beam-splitter setup or a color filter? I guess the latter one will effectively absorb more usable light.
I don't know about that performance claim of CCDs. Red's Epic Dragon sensor, a CMOS chip, is the current performance champ at DxoMark, scoring over 100 points. Some people insist the previous champion sensor, the Nikon D800, should remain at the top because Red's proprietary Redcode RAW format doesn't allow direct access to the Bayer pattern data. At any rate, they're both CMOS-based imaging chips.
If CCDs were really a better solution I think companies like Arri, Red, Sony & Vision Research would be using them in their d-cinema cameras. Even Panavision's Genesis camera, which only does 1080p resolution, uses a CMOS chip. I think the fundamental problem is the beam splitting prism. If you want to use CCDs in a professional camera you need three of them and the image has to be filtered through that prism. At least some penalty with image quality must occur in that stage. And then you have the cost. Three CCDs, the size of a Super35 sensor, would substantially increase the cost of a d-cinema camera. One with VistaVision sized CCD sensors would cost quite a lot more.
One of Vision Research's high speed cameras, the Phantom v7.3, has a CMOS sensor with a true global shutter where every pixel has its own parallel read-out. Unfortunately the sensor has a native resolution of only 800x600.
Over the next few years I expect CMOS sensors in many product lines, from d-cinema cameras to GoPro cameras, to incorporate global shutter functions within the sensor. There's just too much demand for them to put off doing so. Meanwhile, some current d-cinema cameras are offering expensive, mechanical solutions like rotary shutters to counteract rolling shutter artifacts.
I'm watching this situation mainly from the still photography end of things, because global shutter sensors are needed just as badly for still use. It just sucks having flash sync speeds of no better than 1/200 or 1/250. The combination of the sensor's top to bottom linear read out and the old 1st curtain/2nd curtain shutter design creates that serious limitation. Flash sync speeds of 1/1000 to 1/1600 is a big selling point for those medium and large format digital still cameras (they use things like leaf shutter lenses). They'll let you use controlled studio lighting outdoors in broad day light. You can open the camera's aperture to get a more artsy looking depth of field and controlled focus and greatly reduce ambient light from the sun. It would make a huge difference on things like wedding and event photography gigs.
quote: Marcel Birgelen
My last TV also came with "3D" support, although it was being pushed as the next-big-thing a few years ago, I've only used it once ever since (I don't think that 3D on small(ish) screens really makes any sense) and now it's just an extra logo on the box.
And that's really the problem with selling new things like 4K. Stock investors want these new "the next big thing" items to generate huge sales windfalls like what DVD did for home theater. Too many people now have digital devices that work fine and are good enough. Even though most people have HDTV capable flat panel TV screens, plenty of them are still feeding those screens standard definition signals from DVD players and stretching standard definition feeds from a cable or satellite TV box. We're now in a cycle where people are only adopting new technology standards when their old device finally dies. I know people who own Blu-ray players yet are still buying DVDs because they don't understand the difference. With those factors present at the consumer level and what the movie studios are doing to gut their home video distribution departments I don't see things like 4K creating a huge boost in the consumer electronics business.
Likewise, the lethargy taking place on the consumer end will allow commercial movie theaters to drag their feet on upgrading to better projection and sound systems.
Overall, I think ground-breaking movies, ones that turn into a pop cultural phenomenon the way Star Wars did back in the late 1970's, do more to drive sales of commercial movie theater gear and consumer electronics gear than anything else. The problem is we don't have any of those kinds of chance taking movies being made anymore. Hollywood is too busy playing it safe in the content it creates.
| IP: Logged
|
|
|
|
|
|
Marcel Birgelen
Film God
Posts: 3357
From: Maastricht, Limburg, Netherlands
Registered: Feb 2012
|
posted 09-17-2014 03:13 PM
quote: Bobby Henderson
I don't know about that performance claim of CCDs. Red's Epic Dragon sensor, a CMOS chip, is the current performance champ at DxoMark, scoring over 100 points. Some people insist the previous champion sensor, the Nikon D800, should remain at the top because Red's proprietary Redcode RAW format doesn't allow direct access to the Bayer pattern data. At any rate, they're both CMOS-based imaging chips.
I've not seen any pictures shot with the Red Epic Dragon camera, at least not that I actively remember, so I will withhold judgement. What I do know is that I generally dislike the looks of their previous cameras, like the Red One and the Epic. It's hard to assign any non-subjective terms to it, but I would describe it as flat and lacking color depth. Also, they tended to be rather noisy at low light levels, noisier than other CMOS based cameras.
quote: Bobby Henderson
I think the fundamental problem is the beam splitting prism. If you want to use CCDs in a professional camera you need three of them and the image has to be filtered through that prism. At least some penalty with image quality must occur in that stage.
The only real problem besides focal length that I see with beam splitters is convergence. Convergence is also a problem in 3DLP and 3LCD projectors. But a beam splitter will be more light efficient than a bayer filter. A filter effectively blocks out most of the light per pixel. A prism splits the beam into three separate beams and doesn't just throw light away. (Yeah, sure, nothing is 100% efficient, but you get the point ![[Wink]](wink.gif) ).
Besides that, the larger size of the individual photocells in a CCD avoids the usage of a micro-lens array, like many CMOS cameras use.
quote: Bobby Henderson
If CCDs were really a better solution I think companies like Arri, Red, Sony & Vision Research would be using them in their d-cinema cameras. Even Panavision's Genesis camera, which only does 1080p resolution, uses a CMOS chip.
Canon openly admits the primary reason for CMOS imagers is power consumption and costs. Effectively, CCDs still deliver a far better image quality in almost all aspects, they are more light sensitive, less prone to noise and don't need any hacks for universal shutters. For the market Canon is in, cost and power consumption is very relevant, for a professional cinema camera, those factors are far less important.
The reason why CCDs are a dying breed in even high end video cameras? I guess it's just because main-stream has gone the CMOS route and it's cheaper and less risky to improve upon that technology than go the CCD route.
CCDs are still being commonly used for high end science imaging applications, like digital telescopes for example. The primary reasons are both light sensitivity and less noise.
And this noise really is quite a large, remaining issue for CMOS. For still photography using a real physical shutter, this is not really a problem, as the camera actually takes two pictures, one with the shutter closed and one of the actual image. The "black" picture is then being used to cancel the noise in the actual image. This isn't perfect, but "good enough" for most applications, even most professional applications. For video cameras without a physical shutter, this is absolutely impossible and those who still have a mechanical shutter, it will still be hard to squeeze an additional black frame in-between each recorded frame.
quote: Bobby Henderson
Overall, I think ground-breaking movies, ones that turn into a pop cultural phenomenon the way Star Wars did back in the late 1970's, do more to drive sales of commercial movie theater gear and consumer electronics gear than anything else. The problem is we don't have any of those kinds of chance taking movies being made anymore. Hollywood is too busy playing it safe in the content it creates.
Maybe movies have become far too expensive over time. If you look at the initial budget of Star Wars, which was about $11M, that would be about $43M today. That's still a lot of money, but a big budget blockbuster nowadays starts somewhere around $100M. All this excludes marketing costs, which can easily double the actual costs of a movie. You only need a handful of those $100M+ projects to explode in your face and it can bring the future of your entire business into peril. So no wonder everybody is playing it safe...
| IP: Logged
|
|
Bobby Henderson
"Ask me about Trajan."
Posts: 10973
From: Lawton, OK, USA
Registered: Apr 2001
|
posted 09-22-2014 09:28 AM
It's kind of funny how we (or at least I) was suggesting the need for larger sensor sizes. This weekend Arri officially announced a new camera they were rumored to have been developing, the Arri Alexa 65.
There's several articles about it online already, like this one at the No Film School web site.
Arri Rental Group also has this page available now.
The new 6.5K resolution sensor covers slightly more area than a frame from a 5-perf 65mm film negative. The sensor's active image area measures 54.12mm X 25.58mm, with an "open gate" maximum resolution of 6560 X 3102. Obviously the image will get cropped in just a little. The sensor is CMOS based.
Some useful size comparisons: a "full frame" DSLR still camera sensor, like that out of a Canon 5D body, has a 36mm X 24mm image area, similar in size to a VistaVision frame. "Super35" sized sensors are smaller, only 24.9mm X 14mm.
The photosites on this new 6.5K Arri sensor aren't significantly any bigger than those from the Arri Alexa XT camera, due to the much higher resolution level. It's almost like Arri took the same sized pixel grid and added a lot more pixels to it to flesh out a 65mm film frame sensor. I wouldn't expect a huge improvement in terms of color & contrast quality, but the picture will be a whole lot sharper.
Just as I expected, the development of a new camera like this also required development of a new series of "XPL" mount lenses. Arri has a wide variety of them with optics made by Hasselblad & IB/E. One obvious question I have: will this camera be able to work with existing 65mm film camera lenses, like those from Panavision's System 65 line? It would be a shame if Arri didn't already have some kind of adapter ready to fit such lenses.
There are no commercial grade 6K resolution d-cinema projectors. We probably won't see such projectors for at least a few years. The movie industry is gearing up for 4K 6P Laser projection.
However, if 5-perf 70mm film projection can be revived that would provide at least some way for 6K cameras from Arri and Red to get that extra detail onto the screen.
| IP: Logged
|
|
|
|
|
|
All times are Central (GMT -6:00)
|
This topic comprises 3 pages: 1 2 3
|
Powered by Infopop Corporation
UBB.classicTM
6.3.1.2
The Film-Tech Forums are designed for various members related to the cinema industry to express their opinions, viewpoints and testimonials on various products, services and events based upon speculation, personal knowledge and factual information through use, therefore all views represented here allow no liability upon the publishers of this web site and the owners of said views assume no liability for any ill will resulting from these postings. The posts made here are for educational as well as entertainment purposes and as such anyone viewing this portion of the website must accept these views as statements of the author of that opinion
and agrees to release the authors from any and all liability.
|
Home
Printer-friendly view of this topic