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?

They both are and will be good. Both are moving targets. Red has an extreme price advantage, but that's not a primary concern on some productions.

Today, Panavision's Genesis has a bit better way of dealing with motion- much less skew, and somewhat better system integration. Red has a better EVF with look-around, smaller possible form factor, and a very clever post path.

Image wise they are a bit different, but in the same class.

Matt Uhry
www.mattuhry.com
 
5d5w74.jpg


Genesis looks really nice with this flash mag.

The thing I worry about with an HDR camera is not capturing the bracketed images, it's how the software chooses to meld them. That's the real trick. Perhaps the imaging brains at Adobe could be brought in to help with the software side of an HDR camera.
 
Probably size and weight advantages too on the RED.
 
One more very important thing. The F35 seems to have Sony's very distinctive over saturated color look. You see it across all the Sony cameras and is somewhat in movies shot using Sony cameras. I don't know how of this problem can be corrected in post but there are only 10 bits instead of 12 to work with. Over saturation is a bad thing for preserving image detail.

it's true that if you are recording in rec709 you get the "special sony candy look" - but with the f35 you can switch to s-log, preserving more information for the postproduction.

regards,
martin
 
Actually you're missing a few numbers. Sony's 10 bit uses three photosites per pixel instead of 1 like RED. So it works out to 30 bits compared to 12 bits.

That's not exactly the correct math. Sony does use a 14 bit sensor, but it doesn't allow you to save the entire 14 bits into a raw negative. The Red uses a 12 bit sensor and you get a 12 bit raw file. The Sony does produce a nice image, but it isn't as adjustable as the Red One. The image processing on the camera down converts all these pixels into 1080. You don't have control over how the down sampling is done. Anything baked into the image in this process is there to stay.

If you like the image of the F35 then there's nothing to worry about. It's a great camera, but you lose a lot of control over what happens to the image. That may be an advantage for some people and a disadvantage for others.
 
5k is sufficient, with a decent OLPF and demosaic to have a full measured 4k. 6k just adds gravy :-) Or more room to move. How short sighted is it to think that the acquisition resolution can be as small as the delivery resolution?

Graeme

For best delivered image quality your acquisition resolution/dynamic range should be twice or better the delivery resolution/dynamic range. Like you say Graham, you need room to maneuver through the post process. Whatever you start with will always get degraded to some extent by the time it is delivered.
I was reading an article on HDR photography the other day. The whole point of the process is to take a 14 stop or better acqusition DR down to a print DR of about 5-6 stops with maximum creative control over the final image. The same is certainly true of both digital video and film presentation. Despite the contrast range spec games of the projector manufacturers, you will do well to wind up with an on screen visual contrast ratio of 100:1.
 
it's true that if you are recording in rec709 you get the "special sony candy look" - but with the f35 you can switch to s-log, preserving more information for the postproduction.

regards,
martin

When stepping through the gamma curves on the F35, the s-log does look less candy like. But there's still a lot of color space thrown out. The gamma curves are user adjustable, but the camera controls all the image down sampling and rounding error. From what I saw from the sample videos, there's still this very Sony like signature burned into the image. Very video like and less film like. It's not a bad camera by any means, though.
 
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 levels of gray = 16,777,216 possible RGB color values per pixel
10 bit = 1024 = 1,073,741,824
12 bit = 4096 = 68,719,476,736

Greater bit depth gives a broader color gamut and a more subtle shading within the available DR. Your color resolution is limited to the aquisition bit depth. A Bayer sensor will have lower spatial resolution in the red and blue channels, roughly equivalent to 4:2:2 sub sampling on a 3-chip camera, but greater bit depth produces finer tonality in the greyscale and broader color gamut per resolved pixel in either case. I'll take greater DR and bit depth over higher spatial resolution any day once we have enough pixels avialable at acquisitiion to fully resolve an image at the limits of display resolution.

Of course then we get to watch it on a sucky 8 bit consumer digital TV and it all goes for naught.:ranting2:
 
Of course, even in the worst possible case, RED One resolves 2k in RGB 4:4:4 and best possible case (because of optical low pass filtering) 3.2k. Worst case still out-resolves any HD camera.

Graeme
 
Of course, even in the worst possible case, RED One resolves 2k in RGB 4:4:4 and best possible case (because of optical low pass filtering) 3.2k. Worst case still out-resolves any HD camera.

Graeme

Any images taken with a 2/3" Scarlet chip yet Graeme?
 
Guys..... Did you read this article? It appears to me they are DEVELOPING a chip that will enable the same effect as High Dynamic Range stills.
The weight of the camera, or the price, or the signal/noise(Brooke) of the camera are not in the same ballpark as far as competitive importance.....
What this guy is developing is a camera with unlimited dynamic range!!!
Clearly RED needs to have something in the same ballpark, 2 extra stops will not do this.
The new Epic sensor will need to do the same thing to anything like compete with this. THis guy is not talking in the future tense they are already in hardware development!!! Do you really think he would be publishing it the were not pretty close to an announcement?
 
From the interview:

So if you had true 4K resolution in your local theater, everybody would have to sitting in the first 6 rows. Otherwise they wouldn't see any extra detail. Their eyes wouldn't LET them see it.

This is not true.

I was sitting somewhere in the middle of the IBC theatre during RED presentation and one scene made me think about a poor make up artist, since I could see layers of powder on the actresses face.
Clarity of the footage was breathtaking.

I doubt that watching from the last row would make that clarity/resolution unnoticeable to anyone who makes a living using his eyes, unless being shortsighted without glasses. Even for this "marketed" 4K, as referred.

If this statement is not intentionally misleading maybe I'm just gifted with better sight than "Panavision Senior Vice President of Advanced Digital Imaging".
Cool.
 
Well, the trouble is that John Galt is speaking about true 4K measurable resolution, not necessarily the quality produced by a Sony 4K projector. Does a 4K projector put an image on the screen that is measurably 4K?

This is the whole problem of throwing the term "4K" back and forth. We assume a "4K" projector is showing us 4K line resolution of 4K material -- but we already know that the RED cannot produce 4K line resolution, more like 3K.

A 4K projector may simply be sharper a 2K projector, which may not measure out to 2K, hence why you can still see a difference in quality even from the middle of the theater.

I saw part of "Walk Hard: The Dewey Cox Story" digitally projected using the 2K DLP technology -- it was shot on the Genesis, which is 1080P HD... and I could see make-up powder on the actress' face even from the middle of the theater. So I don't necessarily think that seeing make-up powder is "proof" that you must be seeing 4K line resolution. Thinking that the image has a lot of "clarity" is not scientific proof that you can spot 4K from 2K from the middle of a theater.
 
It's hardly a scientific test, but I second Omen's observation in a slightly different way. When watching a 2K projection (from more than 6 rows back in the local multiplex), I can still (at times) see the individual pixels that make up the image. Of course, screen size may have a lot to do with it, and obviously, sitting waaaay back in the nosebleeds, I can't tell a difference.

Could be that 4K is slightly overkill and that 3K is closer to the eye's resolvable limits in the local multiplex, but 3K's not a delivery spec, and I'd much rather see people err on the side of slightly too much resolution than slightly less.

All that said, of course, delivery is one thing, but acquisition is quite another, and obviously, oversampling is the way to go for acquisition.
 
I don't think this new Panavision sensor will have "unlimited" range like an HDR image - for one thing, something with a huge number of stops of luminence would creating banding artifacts unless you gave it a really high bit depth.

The F35 / Genesis sensor already has one extra row of photosites that are ND'd to capture extra highlight information. It sounds like they are adding one more row with additional ND to capture another stop or so more information. So perhaps their current sensor's 12-stop range will become more like 14-stops on the new sensor, which is closer to color negative film's dynamic range, the holy grail for digital photography. But that's not the same as an HDR image, which is a unique process that can potentially give you detail over an extreme range. I've seen HDR photography that captured some 30-stops or more of information, from the burning filament inside a 20K to the deepest darkest shadow in a dark hole.
 
Yes, but again, you're assuming that a 2K projector is showing you 2K line resolution and a 4K projector is showing you 4K line resolution. The fact that you can see pixel edges from halfway back in the theater may just be proof that the image you are seeing is not quite 2K.

Saying that you can see the quality difference between a 4K projector and a 2K projector is not exactly the same thing as saying that you can see the difference between 4K and 2K.
 
BradWright wrote:
"From what I saw from the sample videos, there's still this very Sony like signature burned into the image. Very video like and less film like"
BradWright wrote:
"Because the camera does show color fringing on very fine grey patterns"


Hi Brad

I wasnt at the demo so cant speak to what you saw.
From my experience as a DP and owner of a rental company with 3 Reds ( more than anyone else in SF ) and a F35 I dont understand how you are reaching these conclusions. If these are all from one demo I suggest you may need to spend some time researching these conclusions

Firstly the F35 in all my tests shows less chromatic aberrations than any other camera I have ever looked at with like lenses. We have a particularly nasty backlight security grill that can conjure CA out of most everything we point at it. The F35 has the least CA of any camera I have seen including the Red. We have experienced some moire issues but nothing significant. Issues that can also be found when downrezzing Red footage

Secondly the F35 color is the least Sony looking camera ever. Sure you can make it look like a Sony, but you can also make it look somewhat like a Red. Its very flexible.

We bought the F35 for our clients that found the Red workflow unsuitable for their way of working. Not because it was better or worse, more expensive or cheaper. Mostly because 1080P still rules in the commercial world and the F35 is a gorgeous looking 35mm camera and very very format flexible in a traditional way. We bought it only after realizing Red were not going to provide a camera that would satisfy these clients in the short term. Believe me we had nothing to prove spending this much money. Only a business to feed.

Brad if you live in the Bay Area feel free to swing by our office anytime to take a closer look at the F35. Youll have no obligation to rent anything, but if you do please choose the Red, it has a much better ROI :-)

John
Chater Camera Inc
San Francisco
 
HI David welcome aboard the HDR ride. (unlimited is certainly hyperbole)
My reading of this suggests they are looking at more levels than what your are saying here, near the bottom of his article he talks about the new in development, Dynamix sensor in terms of 6 levels of exposure. So lets say they are three stops apart? doesnt that make 6X3 or 27 stops? How much bit depth? is there any other way is this physics or design? Lots here I dunno, lots of good points in the article, do you see a hole in his logic about beyer sensor actual rez?
 
Back
Top