Welcome to our community

Be a part of something great, join today!

  • 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 :)

Creative Cow Article: Panavision, FUD and...HDR?

Not impugn your credibility Michael, I'm sure what you say is true to some extent, especially on larger theatrical projects but I do have first hand experience of a series of Sony promo shoots that started out Red and when a Sony exec found out demanded that Sonys (F900s being the only similarly priced rental camera) be used. The look was significantly altered and compromised by this.
 
That 2K DLPs can show 2K resolution is easily verified with test patterns.
regarding whether 2K DLPs show 2K resolution:

For practical purposes, I would just assume that a 2k projector delivers 2k resolution. From a production + post standpoint, it is practical to make that possibly faulty assumption.

2- Suppose the projector projects an image that has very blocky looking pixels. (You might have seen this on some projectors.) You're getting a lot of resolution since the image isn't blurry. If the projector is projecting a test pattern with black and white lines, then you would say that all the black and white lines are being resolved.

-But the problem is that the pixels look blocky. To get rid of that, now we want to blur the edges of the pixels/blocks. And we don't want any dark gaps between the pixels.

If we are projecting black and white lines, these lines would now be more grey-ish.
One might say that we are losing resolution. How much resolution you "lose" depends on how you want to measure it. By some measurements, you lose no resolution. By other methods of measurement, you might be losing resolution.

-What it comes down to is how the picture looks like when it is projected. We run into the problem of either blurriness and/or blocky looking pixels. There is also the human visual system + psychology to take into account... both are very complex.

You can try to put a number to "resolution" and how the pixels look (e.g. it can be analogous to aliasing). From a practical perspective, you should just pick a model that suits your goal.
If you are a theatre owner, then you may realize that people probably won't notice the difference between projectors (digital and film). To increase quality, it may be more effective to have a clean theatre and to have big threatres that look the part. (Uh... much like how many post houses have very nice furniture.) From a business perspective, this would probably be the model to use.
If you are a DP, I would just assume that a 2k projector puts out 2k resolution. A well-setup projector would have slightly blurry pixels and the pixel structure won't be overly objectionable. We can get the projectors looking pretty sharp without objectionable pixel structure. What matters is whether or not the image looks good. If some engineer comes up with numbers that says that the image 'ought' to look bad (or good), well... I think that would be applying the wrong model to the situation.

3- If you compare digital projectors to film projectors, then film projectors handle things a little differently.

Film doesn't suffer the problem of an annoying blocky pixel structure since film doesn't have pixels. (Ignoring the DI process, where film is digitized into pixels. Assume photochemical here.) But because of how the film prints are made, the final prints are blurry compared to the original image. And it has grain, judder, different contrast characteristics (which makes stuff look sharper).
It's different.
 
Not impugn your credibility Michael, I'm sure what you say is true to some extent, especially on larger theatrical projects but I do have first hand experience of a series of Sony promo shoots that started out Red and when a Sony exec found out demanded that Sonys (F900s being the only similarly priced rental camera) be used. The look was significantly altered and compromised by this.

SONY is a HUGE company with many divisions ... and many executives. I have personally experienced BOTH sides of this coin - so while I am sure it's not a "mandate" by the corporation - it IS happening - I dealt with this dynamic less than two weeks ago with a significant person from SONY on the west coast suggesting to the producers the Canon 5D as their B camera. I am also aware of a mini-series backed by Sony shooting entirely on RED. It is no secret that several companies have sold, and rented, a few less cameras after the "rise of Red" - and I actually expect a lot more "anit-Red" pressure this year.
 
Guys,
It's a little disappointing to see the spin inserted into a "whitepaper", but let's not forget our own Red Team is happy to boast when and where they can (and rightfully so!) So let's cut them a little slack.
Panavision is a top-notch firm, skilled in design, machining, and optics. The HDR concept looks really interesting. The world of cinematography would be a lesser place without this company. Let's not be too eager for Red to "kill" them.
On the topic of the "Sony Directive":
Yes, happened to me last year too.
It was to be a 2-camera Red shoot and went F-900R with Pro-35s.
I don't really want to say that it's a top-down edict, but I think there are some overly aggressive execs in the middle somewhere who push the agenda.
I also had a 2-camera Red shoot go F-900 with Pro-35 last week, but that was a post fear factor on either the client or agency side.
No big deal. I can usually put all my lenses and accessories on these jobs, so I don't care. It's just that sometimes Red is clearly the better camera, and rental house politics, misguided perceptions about imaging and post, and fear often make for another (poor) choice in camera.
Of course, sometimes' Red's hype has led to rentals for me where it was seemingly overkill for the project (think Web ads etc.) But that's OK with me! As cinematographers, we should always keep an open mind and use the tool that is the right choice for that project.
Cheers,
Harry
 
I don't really want to say that it's a top-down edict, but I think there are some overly aggressive execs in the middle somewhere who push the agenda.

And the problem with that is..???

I don't even understand why this is being discussed. If Red were also a production company, I would strongly suspect that their productions would, at the very least, be "strongly encouraged" to shoot with their equipment. Sony is a company that consists of many different parts. What would be the point of having all of those parts and not encouraging cooperation and synergy between them? And why is that wrong? To me, it's no different than a sound man going on a job and "encouraging" the production manager to let him use his own equipment on a box rental rather than get a different kit from a rental house, even if the UPM had a better experience in the past with the particular kit from the rental house than the one the sound man is offering. Conglomerates exist, in part, to take advantage of the ability to self source things that smaller companies have to procure from outsiders. That's sensible business logic, not some evil conspiracy.

Now, having said that, I also know of many, many Sony productions that didn't have any Sony equipment on their sets other than the occasional monitor.
 
First off, I am not saying that it is not with in their rights to say, "If it's my money then it's my camera." All I said is it happens. Albeit my first statement was too wide sweeping. I also never said that film wasn't still being used.

I am not here to bash Sony, I have many of their products and like what they do. I was simply illustrating that this is often a factor in what camera system gets used and used Sony as an example. They are by no means the exception. Most companies work deals and use leverage to put things in place. It's business. Now wether or not I am operating on urban legend has been, put to bed I feel.

Back to the white paper:

Again, it's Panavisions paper it will be from their perspective. Fine. No problems.
I don't expect everyone to make the same assertions and decisions I have when it comes to camera systems.
 
I also never said that film wasn't still being used.

Again, it's Panavisions paper it will be from their perspective. Fine. No problems.
I don't expect everyone to make the same assertions and decisions I have when it comes to camera systems.

Hi,

You never mentioned film, which is odd as the majority of Sony films are filmed with it.

Whilst Panavision is Sony's Largest customer for F900 series cameras, they make the electronics for the Genesis but also rent RED cameras in London, probably elsewhere too. They may soon be RED's largest customer too.

Stephen
 
HA!
David has it right. Wayyy to much organization for a large company.
If it were a top-down edict then it might make sense.
And Mike, I don't really care; as I said, I still loaded up the rental ticket with my lenses and accessories. And as I've said MANY times before on this forum, we should choose the tool that's appropriate to the job.
And as I indicated in my post, it's happened to me, but one commercial in 20 years does not a trend make...
If I had any beef at all, it would be some mid-level executive's misguided effort to find "synergy" when an F-900 is a poor choice. It should be up to the production company; a technical and artistic choice involving the director and cinematographer. Remember, I'm talking commercials here. This is where Sony has employed an ad agency, who in turn has selected a director and production company, based on their craft and experience. Why should a guy with an MBA insist that they use an F-900, or 35mm, or Super-8 for that matter? Sure it's their money and their right, but how many of us would tell our electrician to use a certain brand of Romex over another when wiring our house? At what point do you stop being an impressive corporate drone and let the craftspeople you've hired do their job in the best way they know how?
But in case you're wondering... like I said... I don't really care! ;)
Cheers,
Harry
 
I agree. It was never my intention to turn this into a camera conspiracy thread. I was only discussing my view points as to many of the factors that lead to which Digital Cinema Camera is chosen.

The more PC way to put it is: "Sometimes the directive as to which camera is chosen is handed down by an executive with in the production with ulterior motives to the earnest wishes of the artistic team."

I didn't mention film because it wasn't part of the conversation. I felt we were discussing the industry comparison of digital formats.

For the record, I don't believe the same tool is right for every job. Each has their strengths and weaknesses. They are all tools that can be used to make handsome product in the right hands.
 
Well for what it's worth, the Japanese Panavision agent is now renting Red cameras along with the Panavision catalogue so it'll be interesting to see how that goes.....
 
-But the problem is that the pixels look blocky. To get rid of that, now we want to blur the edges of the pixels/blocks. And we don't want any dark gaps between the pixels.
Exactly, we don't want to see samples/pixels. We want to see the analogue signal the samples are describing. In theory these samples need to undergo a DA conversion and should not be looked at directly. The DA conversion is done by the optical system of the projector and the eye. If the pixels are rendered pin sharp you need to sit so far away that the screen door effect becomes invisible and the content must be frequency limited to less than half the sampling rate or aliasing is still visible. You are not going to get full 2K resolution from 2K projectors beyond test patterns that align with the pixel grid. Your film credits are either somewhat soft or sharp and aliased on a big cinema screen watched from a distance where you see 2K resolution. 2K is somewhat soft or coarse. For sharp and analogue looking (smooth and not coarse) you need simply more pixels.
 
Maybe Red can Buy the Sensors and Make their own HDR Cameras?

Maybe Red can Buy the Sensors and Make their own HDR Cameras?

Maybe RED can buy the sensors from Panavision and put them into cameras, so that they can have a HDR camera too?

As I understand it, the sensor is available for sale from Panavision SVI.
 
According to their site: http://www.panavisionimaging.com/2d_arrays.htm they neither have engineering samples, nor are in production of their sensor. It's pretty hard to buy such a sensor, never mind use one. I wonder how you turn a diagonal pixel array into a square one though?

Graeme
 
The Magic Pixies are working over-time at the moment here. I'd love to tell you what they're up to, but I had to sign an NDA with the Magic Pixies, and they do read Reduser and would know if I spilled the information. I'm confident you'll find their latest work highly interesting when you get to see it though...

Graeme
 
According to this whitepaper http://www.motion.kodak.com/motion/uploadedFiles/Wheeler_Rodriguez_SMPTE_J-Oct07_Web.pdf
the overall quality is better from a bayer sensor then a striped sensor. Best is a 3 chip system, no mystery there.
I guess both arrays have their merits. The striped array might be better for low light tungsten and chroma key.
What i hope for in next generation Red sensor is improved sensitivity in the blue channel, not only by improved signal processing but some improvements in the dye formula for blue filtration in the CFA array.

Jesper Holmström
Dagsljus Filmequipment
 
I actually started this thread (mine is the first post up to the line "Dynamax is new to me..." - Interesting how this thread got merged with another from later in the day, and now my comment is mis-attributed to Cal Brunker)

Anyway, one of the reasons I wanted to start this discussion (other than to share that Panasonic's party line is now reaching a wider audience through the Cow) is that HDR seems to be a possible new frontier for motion picture imaging, but nothing I've read so far has made a cogent argument for a workable theory behind it, much less a plan for an actual system/sensor.

Look, we already have Red and others pushing the boundaries of what is needed for a true 35mm film replacement. Red has proven time and time again that the Red One is a perfectly viable tool for motion picture photography. The Epic and Scarlet roadmap will continue to slam the door shut on that argument.

It seems that one area left to argue is that film still has the edge in dynamic range. Whatever kind of Magic Pixies are (will be?) at work in the Dynamax sensor, I still can't wrap my head around the idea of non-colocated pixels of varying exposures working well. Don't even get me started on diagonal pixels!

I understand it's hard enough to create good code to interpolate for Bayer-pattern non-colocated photosites. Graeme is truly an artist, in that regard. The naturalistic images from the Red One seem to me to be full of the necessary compromises in current technology being handled always in service of the image.

Why wouldn't it be easier (if your goal was to capture HDR) to have a 3-CCD like image block with a 3-CMOS configuration, have 1 sensor at underexposure, have another at overexposure, and expose for the middle (neutral) sensor?

The on-camera software would compare pixel values for the 3 sensors, and generate "superwhite" and "superblack" values for those pixels that were above clipping or below a minimum charge on the neutral sensor.

I'm sure I'm being naive. I'm not a sensor designer, or an engineer. Is the problem in the additional, massive onboard horsepower needed to process such a pixel comparison? Is the problem that 3 sensors would basically triple the price of such a camera? Is it that you'd essentially have to record 3 streams of data instead of one, then offload the HDR processing of those images into post? Is the light loss from a 3-CMOS image block too much to design for currently?
 
Back
Top