Welcome to our community

Be a part of something great, join today!

Register 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 :)

Dragon with no OLPF

Thank you for all the kind comments on my short piece, shot on Epic.

As I gain more experience I am learning that the camera capabilities are important but the locations and subjects are even more so.
If you want to see some incredible cinematography take a look at "These Birds Walk" ( on netflix instant now ) , a documentary shot in Pakistan by two new filmmakers. They used a rather striped Canon 5dmkII . There are aliasing artifacts and other tech issues, but my god, the beauty of the shots is jaw dropping and frankly the nit picking tech details are quickly forgotten. I want to shoot more like those guys in the future.
 
Les Dittert said:
Thank you for all the kind comments on my short piece, shot on Epic.

As I gain more experience I am learning that the camera capabilities are important but the locations and subjects are even more so.
If you want to see some incredible cinematography take a look at "These Birds Walk" ( on netflix instant now ) , a documentary shot in Pakistan by two new filmmakers. They used a rather striped Canon 5dmkII . There are aliasing artifacts and other tech issues, but my god, the beauty of the shots is jaw dropping and frankly the nit picking tech details are quickly forgotten. I want to shoot more like those guys in the future.
Click to expand...

I really like your film Les. But ofcourse difficult to see or judge the detials on a vimeo link. would low to see a r3d snap from it just to see how it actually looks.
 
Björn Benckert said:
I really like your film Les. But ofcourse difficult to see or judge the detials on a vimeo link. would low to see a r3d snap from it just to see how it actually looks.
Click to expand...

Björn, here are some stills from those or similar shots. I sized these to 4096x2160. Worst case thin wire 'color error' situations on a couple of them.

https://dl.dropboxusercontent.com/u/61349012/misc-RED-test/Tondo_4k_stills/63__2.jpg
https://dl.dropboxusercontent.com/u/61349012/misc-RED-test/Tondo_4k_stills/63__7.jpg
https://dl.dropboxusercontent.com/u/61349012/misc-RED-test/Tondo_4k_stills/100__0.jpg
https://dl.dropboxusercontent.com/u/61349012/misc-RED-test/Tondo_4k_stills/105__00.jpg
 
Thanks Les!

If taking off the olpf in Dragon can make a comparable difference then we will be talking about a new game here. And I think we will.

I still like the r3d's better because someone can do so much with those but I thank you for sharing these exceptional images.
 
This is my first post here, hello to everybody! I am a beginner in using RED cameras and I am from Germany.

I want to get high quality 4k footage for demo purposes.

I planned to buy a scarlet. So I rent one first to gain some experience. But I was really disappointed seeing the first results. My expectations were to get a picture quality like a DSLR with the REd cam.
Then I found the postings an picture examples of Les Dittert. This was exactly what I was looking for.

Now what should I do? I do not find here an explanation how to get these results. Does anybody know whether RED will come out with some new components that allow us to shoot on that quality level?

My results btw were not better than what I saw for example from a panasonic GH4 in terms of sharpness.
 
Juergen Schloss said:
This is my first post here, hello to everybody! I am a beginner in using RED cameras.
I want to get high quality 4k footage for demo purposes.
I planned to buy a scarlet. So I rent one first to gain some experience.
My results btw were not better than what I saw for example from a panasonic GH4 in terms of sharpness.
Click to expand...
Really?
Interesting first post.
My advise go and get a GH4 then.
Edit: Or check this thread http://www.reduser.net/forum/showthread.php?112235-shut-up-and-shoot-that-Scarlet
Good luck and happy shooting. ;)
 
>>Now what should I do? I do not find here an explanation how to get these results. Does anybody know whether RED will come out with some new components that allow us to shoot on that quality level?

There's no doubt about it in my opinion. But also you could have gotten these results yrs ago if you have followed the same olpf procedure Les did.

I think your judgement has not taken in consideration the complete picture. Maybe is a good idea to rent the Scarlet again and learn it better. Also use the rcx and read the manual before using it and while using it.
 
I recently got a GH4 to try. So far I am pleased with it. I resolves more detail than a factory MX camera. It shoots 4096 x 2160.
I held off getting one because it was only 100 megabits data rate but when I saw some 4k Youtube videos doing side by side comparisons with the Red I decided to give it a look.
The response curves in it are very adjustable letting you get about 10 or 11 stops dynamic range. The video from it is more color correctable than I would have guessed.
My damaged back is telling me to use it, but I like the high frame rates of the Red. For my developing world shooting projects the light weight and insane battery life make it very seductive. The camera weighs 600 grams, about the same as one of my v-lock batteries for my Epic. Getting shallow focus shots is a challenge. I have an f 0.95 lens for that, but it's optical performance wide open is not that great.
 
If it was intentional then expert
But if he just forgot the lens cap was on...
John Marchant said:
Definitely not a guess hehe. Appreciate that for many it would be of course :) The lens cap test is the proof though, even if it were a guess.
Click to expand...
 
Les, I fully understand your position to not explain in detail your procedure to achive these really amazing results in image resolution. I am totally enthusiastic about it.
Anyhow please allow me to ask a question. When I look only to the aliasing aspect and not to the IR and other color requirements then I do not understand why the resolution in 4k is so much influenced by the OLPF in a system that is able to resolve 5k or 6k. The cut off spatial frequency of that OLPF should fit to the 5k or 6k anti aliasing requirements and therefore it's influence on 4k should be neglibigle when 4k is not a crop but a downscaling. Am I wrong?

If this discussion does not fit to this thread, then please ignore my post.
 
Since this is kind-of an OLPF thread, let me answer:

The OLPF causes the light going through it to get directed in four different directions by a small amount. This amount is determined by design. The reason for this need to spread the image around is the color filter array on the sensor and the subsequent debayer software can get tricked by very thin details in the image, such as thin wires and hair strands. By spreading the image around to the neighboring color filtered pixels the debayer software can be more confident that the color for the final color pixel the user sees is more true to the real image before the camera. A side effect is that those very same tiny details will appear to be softer, again because the detail information has been more spread around.
Because this is a movie camera the issue is of more concern because on details that move across the screen the errors can come and go, making the errors more visible than just looking at a motionless still image.
It's by belief that the OLPF spread of the MX was chosen to be ultra conservative and will never make debayer color mistakes. This is why people looking at DSLR raw images against the MX images are seeing less detail.
On Dragon I have ( with V2 OLPF) been able to get it to make small visible color rainbow errors on a black and white test charts. It seems tighter.
I have learned that advanced debayers can minimize these problems as you move towards less and less spread from the OLPF.

Juergen Schloss said:
Les, I fully understand your position to not explain in detail your procedure to achive these really amazing results in image resolution. I am totally enthusiastic about it.
Anyhow please allow me to ask a question. When I look only to the aliasing aspect and not to the IR and other color requirements then I do not understand why the resolution in 4k is so much influenced by the OLPF in a system that is able to resolve 5k or 6k. The cut off spatial frequency of that OLPF should fit to the 5k or 6k anti aliasing requirements and therefore it's influence on 4k should be neglibigle when 4k is not a crop but a downscaling. Am I wrong?

If this discussion does not fit to this thread, then please ignore my post.
Click to expand...
 
Les Dittert said:
Since this is kind-of an OLPF thread, let me answer:

The OLPF causes the light going through it to get directed in four different directions by a small amount. This amount is determined by design. The reason for this need to spread the image around is the color filter array on the sensor and the subsequent debayer software can get tricked by very thin details in the image, such as thin wires and hair strands. By spreading the image around to the neighboring color filtered pixels the debayer software can be more confident that the color for the final color pixel the user sees is more true to the real image before the camera. A side effect is that those very same tiny details will appear to be softer, again because the detail information has been more spread around.
Because this is a movie camera the issue is of more concern because on details that move across the screen the errors can come and go, making the errors more visible than just looking at a motionless still image.
It's by belief that the OLPF spread of the MX was chosen to be ultra conservative and will never make debayer color mistakes. This is why people looking at DSLR raw images against the MX images are seeing less detail.
On Dragon I have ( with V2 OLPF) been able to get it to make small visible color rainbow errors on a black and white test charts. It seems tighter.
I have learned that advanced debayers can minimize these problems as you move towards less and less spread from the OLPF.
Click to expand...

Thank you Juergen for this great question and Les for an even greater answer.

It appears that if I want to film black and white shooting night or late hours it'll be extremely beneficial to take the olpf off. The pic will be sharper and get more iso?

The problem with changing olpf's or taking it completely off is also dust. I think once people start changing it often we'll see lots of problems with dust.
 
Two quick things:

1> The camera needs some glass in there in place of the OLPF to maintain backfocus. It is so far off without it that you can't focus on much at all.

2> The ISO stuff is really more effected by the IR cut filter that is sandwiched with the OLPF filter. To cut the IR out even better, the V2 IR filter eats into the red spectrum just a little more with a sharper cutoff slope , and it also passes the remaining useful color spectrum a little less as a side effect. This is why the older filter is prefered for low light, as far as I can tell. That combined with a different characteristic sensor curve that allows the more film like bright roll-off by underexposing the rest of the range, or as some call it, highlight protection. It will be interesting to see what the new firmware does to change OLPF metadata settings without actually changing the actual filter !


You don't get very sharp pictures shooting no IR cut because the IR that is then allowed through focuses at a different image plane and will be out of focus compared to the color part of the image. This can look like a bit of blurriness and fogginess to your image, with awful color as well.

Lakis Amarantithis said:
Thank you Juergen for this great question and Les for an even greater answer.

It appears that if I want to film black and white shooting night or late hours it'll be extremely beneficial to take the olpf off. The pic will be sharper and get more iso?

The problem with changing olpf's or taking it completely off is also dust. I think once people start changing it often we'll see lots of problems with dust.
Click to expand...
 
Les,

I played tonight with several lenses at high iso with the new v2 and v5.2.28 and I see a much better much less noisier pic.

Les,
I understand what you say exactly and all makes sense.

What I don't understand is how you managed to take That footage without olpf and look so good! I understand the need to put some glass instead of the olpf to maintain backfocus.

I took the olpf out of one Epic and shot some test a yr ago and all was bluish and unacceptable to put it kindly. No matter what I tried in rcx was still unacceptable so after a quick and unsuccessful shot I put it back disappointed...

Your footage looks fantastic and very sharp especially as the lenses you used I own and they are not that sharp in my Epics or Dragons.

Probably I miss something? Is it possible to put some r3d's?

Very repectfully
Lakis



[h=1][/h]
 
When you shot your Epic test with no OLPF, what were you cutting the IR with ?

Lakis Amarantithis said:
I took the olpf out of one Epic and shot some test a yr ago and all was bluish and unacceptable to put it kindly. No matter what I tried in rcx was still unacceptable so after a quick and unsuccessful shot I put it back disappointed...

Your footage looks fantastic and very sharp especially as the lenses you used I own and they are not that sharp in my Epics or Dragons.

Probably I miss something? Is it possible to put some r3d's?

Very repectfully
Lakis
Click to expand...
 
You must log in or register to reply here.
Back
Top