Welcome to our community

Be a part of something great, join today!

Log in
  • 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 :)

2K Soft Focus on all Red Ones

K

keicol

Well-known member
Joined
Oct 13, 2007
Messages
57
Reaction score
0
Points
0
It's most noticeable while shooting Wide Shot VFX elements on stage when focus is critical. I'm usually using Cooke S4 14 to 21mm or the Zeiss Ultra 16 lenses from 6mm to 12mm.

Focusing on objects in the 15 to 20' range is most difficult. I've even tried to loosen up the back focus to see if I could dial in better focus. It might be a result of shooting in the 100 fps range at 2k with a 180 degree shutter that just doesn't hold up like 4k at normal speeds but I doubt it.

It's most noticeable when shooting background plates at distances more than 15 feet do we see this, and it's on every Red we've tried.

Try looking at the same field of view at 2k with a 4k shot at around a 27mm or wider. Try to match the width of the 4K shot as close as can be gotten with the glass you have available for the 2k shot. Focus on the objects at 15 to 25' range and see if they can be focused on sharply. That's my problem when shooting background plates for VFX shots at off speeds which need to go to 2k to reach desired speeds. I'm finding it impossible to sharp focus on objects at those distances when that wide at 2k.
 
I Concur. :)

I Concur. :)

Yep, I can say that I concur with Keith's findings. While I do not do VFX work, the difference is visibly noticeable. I shot a quick test to see it for myself. You can download the stills that I shot at the link below. They are nothing special visually, but they do show what is going on. (Red 18mm - 50mm lens used.)

Focus was set on the middle brandy bottle (The text) which is 15 feet from the camera. The footage was shot in Redcode 36, processed in Red Alert using the highest debayer setting. I shot the 2k footage at 18mm at both 24 frames a second and at 75 frames. To help match exposure I increased the ISO to 800 for the 75 frames. (The lens was at T3 for all shots.) The 4k still was shot at 38mm, and I tried to match the framing as much as possible.

Tiff Frames: (66MB)
http://www.ryanewalters.com/downloads/ResTest.zip

This issue really intrigues me, so I'm going to shoot some more tests later this week to see what is going on.
 
One might think that a statement like this would draw more attention. I hope this situation is corrected in the new models. I'm thinking that if they're not aware of it's existence then it might not be addressed in future releases. Now would be a good time to get a comment from a Red Rep assuring us that this concern is being addressed.
 
2k is a crop of the sensor so Of course its not as sharp.

I noticed this with 16mm Lenses too, With an 8mm set on 4' everything 4' to infinity was sharp. Its completely Logical. Red's sensor is bigger than a 16mm Image Circle,
 
Not this much off, I've seen sharper pictures at HD resolutions on HDV cameras with consumer lenses. Check out Ryan's pictures. They speak for themselves. An unsharp picture is just that. Unusable in a professional situation. When production wants a slomo version of a shot I just did at 4k I don't want to tell them that they will have to sacrifice focus, that they'll have a problem pulling a matte or recognizing a character in the shot. Basically what we have is a really great camera at 24 to 30fps but at off speeds we just have a soft focus camera. I just hope that someone from the higher up's might comment on future versions of this camera not having this issue. I can forgive problems on the first version of a great thing but let me hope that version two of which I'm ready to fork out a bunch of hard earned on will be void of these shortcomings.

I want to shoot sharp off speed, just like film can and does. I want a tool that can deliver what I tell producers it can, not have them call me during dailies with complaints about the soft off speed shots.

Sorry, that's just how it's got to be.
 
I think you should check and be sure BackFocus is set correctly. And be sure Lenses are Checked out.

Also 180º may not be enough for you, Ive had fantastic results at 90º with a 2.8 using Ultra Primes. Shooting Frame Rates of 36-100 fps.
 
Alex Carr said:
I think you should check and be sure BackFocus is set correctly.
Click to expand...

I've even tried dialing in back focus to see if I could get a sharper focus. I've also tried Zeiss S16 Ultra Primes thinking I'd get more focus room but the problem seems to be on subjects at 15' and farther from the lens, the focus sits somewhere between 20' and infinity which is a very small space on the actual lens. Sometimes while shooting wide open on a moving Techno Crane it's just useless to try and follow focus.

You need to do a test like Ryan did and see for yourself, check out his results, there's a link to his photos posted above. If you don't have a camera and lenses to do the test you're welcome to come over to my studio and use any of mine.
 
Alex Carr said:
2k is a crop of the sensor so Of course its not as sharp.
Click to expand...

As I mentioned above, I'll be doing more testing this week on this issue. But the center crop of the 4k image shot at 4k does not seem to have this problem. So it seems to be limited to only the 2k mode. Additionally, the 15' - 20' range seems to be a "magic" number as well as things at that distance go soft in 2k, even though they are in focus in the 4k image.

This doesn't seem to be a back focus issue, or a lens issue. With the tests that I did above, I set the focus in 4k and kept it locked off. So if it were either of these issues it should have showed up in the 4k image as well.
 
You cannot shoot wide open on a 16mm Prime. I've tried it, the focus is way off and nothing seems right. You just can't shoot T1.3 with 16mm Primes. I was using an 8mm and until I was stopped to a T2 1/2 things were getting right. But you cant trust the markings on the barrel for 16mm primes anyway. My focus was set at 4' and everything to infinity is in Focus. Many of the 16mm (wide) Primes will work this way and you cannot check focus on the Monitor. Look at a DoF Calculator to find your DoF and set it to that mark. Mine was 4' on the 8mm at T2.8,

Was your Backfocus set by a technician with a Collimator? If not then you will run into problems. This shouldn't be something AC's can willingly adjust. And I think your problem lies here. This is an extremely Precise setting, You can't set up a chart and set the Backfocus to it at a certain distance, It doesn't work that way. It has to be put on a Collimator, an AC on set can only get about 50% to the mark without the proper tools and time to study it.

This kind of thing happens when backfocus is set incorrectly. Did your 1st AC Check the Lenses and ask to Collimate the body before it left the Rental House?
 
Alex,

I assure you that I never shoot with a badly back focused camera, nor that my tests were restricted to shooting wide open on S16 lenses. I'm sure Ryan whom I've never met is as professional with his as you are with yours.

I'm very lucky to have my studio minutes away from Jacek's Optitek who's one of the best optical engineers around and fortunately for me he's always testing his new inventions on my cameras. In fact his latest invention is his really cool portable backfocus collimator. In fact all 20 of my lenses have all been collimated, projected many times at his studio, it's almost like having an optical engineer on my staff so no, it's not a camera or lens issue. I suggest that you get your camera out and test what we're talking about and see for yourself.

Do a center crop of a 4k image, blow it up and match it to a 2k image with the same field of view. Have your subjects at 15' to 20' and try to get a good focus, enough to match the 4k punch in, it's impossible.

2k is simply soft, softer then a center crop blown up of a 4k image, go figure??

I just hope this problem is solved with the new improved next generation of cameras. I would love to have some feedback from Red confirming this.

Please Alex, join us with the addition of your findings based on actual tests instead of what you'd like and hope to see. I'm with you, I'd hope to see this differently and be able to offer sharp slomo on my gigs, wouldn't we all?

Cheers, Keith
 
As I mentioned above, I'll be doing more testing this week on this issue. But the center crop of the 4k image shot at 4k does not seem to have this problem. So it seems to be limited to only the 2k mode. Additionally, the 15' - 20' range seems to be a "magic" number as well as things at that distance go soft in 2k, even though they are in focus in the 4k image.
Click to expand...

If you shoot 4k at the wider angle (18mm) so the center 2k crop matches the 2k shot FOV is there still a difference? If there is not then you are bumping against the limits of the lens, OLPF, and smaller sensor area resolution or MTF.
 
Alex: If I understand Ryan correctly, this is not a collimation issue -- he's using the same lens in both 2K and 4K; the same error should show up in both cases.

Ryan: Looking at the tiff files, it appears that focus has shifted. It appears that the closer chairs are in focus in the 2K file, and that the rear chairs are in focus in the 4K file. Is it possible that there is a focus shift related to zooming the lens to equalize the frame size?
 
We shoot 2K, 3K and 4K constantly - mostly action sports outdoors, and ahve done exhaustive testing of various lenses, post techniques and variables, collimation variances, different mounts for stills, etc - and our experience mirrors the original poster in essence. We use RED as much for the over-crank capability as anything else - as much as we are deeply dedicated red fans and users, without 100+ FPS at 2K, we'd be shooting S16mm on Arri HS's - something we considered seriously after our initial 2K red testing - see another thread on that.

I too wrote it off ultimately to the simple sensor limitations at 2K, however, we find material focus variability even when other variables are not altered, same glass 3K to 2K, and it does seem to go beyond just sensor/OLPF/downstream issues.

There seems to be some basic technique needed to focus 2K in the near infinity range that we have yet to master. the fact that we are shooting in remote environs, run and gun with little ability to calmly and slowly set everything up makes it, well, challenging! Pragmatically, we get acceptable results sometimes, and then for reasons we can't quite nail down, don't at other times. Often, we know what mistake we made, but not always.
 
keicol said:
2k is simply soft, softer then a center crop blown up of a 4k image, go figure??

I just hope this problem is solved with the new improved next generation of cameras. I would love to have some feedback from Red confirming this.
Click to expand...


Epic was announced 5k @ 100fps. 2K would be way higher.

I've done the same tests comparing a 2k with 4k (and center cropped to 2k) and it looks exactly the same. I've no idea why you'd be seeing something different.
 
Ariana said:
I've done the same tests comparing a 2k with 4k (and center cropped to 2k) and it looks exactly the same. I've no idea why you'd be seeing something different.
Click to expand...

You're probably not looking at subjects 15' or further out.

I'd love to see your tests that contradict our findings, so far Ryan, Anson and myself have all independently come up with the same conclusions.

2K is soft focus on all Red One cameras

Check out Ryan's published tests above.
 
If you shoot a shot in 2k and 4k with the same lens focal length, focus distance, etc. then crop the 4k to 2k in post is it sharper than shooting at 2K? If shooting 2k is softer in this situation then there must be an issue beyond the optical path, maybe in the debayer from 2k or some other aspect of the software.
 
I don't see why it would make a difference whether you cropped 4K to 2K in-camera or in post other than you have less info to compress if you do it in-camera.

But remember that the low-pass filter was optimized for the full 4K sensor.

To compare something remotely similar, you'd have to compare the Red in 2K to the SI-2K to see if a sensor designed for 2K, with an OPLF designed for that resolution, produces better results or not. Otherwise, I can't think of any other single-sensor 2K camera anyway to tell you if there is any way to get a better 2K RGB image when only starting out with 2K RAW.
 
keicol said:
You're probably not looking at subjects 15' or further out.

I'd love to see your tests that contradict our findings, so far Ryan, Anson and myself have all independently come up with the same conclusions.

2K is soft focus on all Red One cameras

Check out Ryan's published tests above.
Click to expand...


It was a while back so I'd have to look for the tests.
We shot both charts and a range of footage including wide landscapes and closeups. 2k equaled a 4k cropped (not scaled) to 2k.
 
David,

Please, if you have the time, come over to my studio and we'll fire up some cameras and I'll show you what we're talking about. I'm at New Deal downstairs from Ken at Filmworks, we've seen each other there a few times while you were timing Manure and Stay Cool.

This is a very frustrating issue. When shooting 2k off speeds with subjects at around 15' or greater using a wide lens the picture is unusable for compositing plate work. This is very apparent when shooting VFX elements at 4K and say a 25mm or a 32mm S4 and having to match that same shot at an off speed in 2K. The difference is more than just blowing up the center punch of a 4k sensor to get the 2k image which I've done to compare the differences. The difference is enough to bring out the film cameras and pull the Reds.

I'm actually shooting with a 3D SI2k rig for the next two weeks on our stage and that might be a fun test but the lenses on the SI2k rig are these tiny Schnider C mount necessary to fit the beam splitter rig and wouldn't be a very good comparative test against my S4s but it would be a good test especially if the SI2Ks were actually sharper. Why don't you come by and join us for a look see?

Cheers!
Keith
 
You must log in or register to reply here.
Back
Top