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 :)

The toe and the shoulder Dragon DR

Björn Benckert said:
There are quite a few parmeters in the mix here. Iso, different sensors, underexposure and and color science and olpf's...


The dragon sensor is far more light sensitive than MX. Or at least dragon sensor with olpf1 is far more light sensitive than MX with MX olpf (I have not tested the cameras without olpf, so it could be so that the MX olpf just eats more light than the dragon olpf1, but I dont think so I think it's the photosites that are more sensitive in dragon) . I find it very accurately when RED state that: native / mid gray MX @ 800ISO = native / mid gray Dragon olpf1 @ 2000ISO.

To me underexposure is photosites that did not get enough light to reach above the noise floor. Dragon olpf2 needs more than twice the amount of light compared to olpf1 to reach that level.
Click to expand...

Björn, I only can talk about Dragon with OLPF2.

But your findings are interesting. Perhaps OLPF2 ist not the optimum and we will see an OLPF3?

Hans
 
Hans von Sonntag said:
Björn, I only can talk about Dragon with OLPF2.

But your findings are interesting. Perhaps OLPF2 ist not the optimum and we will see an OLPF3?

Hans
Click to expand...

Yes I think so and I hope there will be quite a few more olpfs to come.
 
Its clear the DR has to be moved toward highlights or shadows with the OLPF. Phil's number of 1.5 stops seems about right. So an OLPF that lets in 3/4 of a stop with all or most of the characteristics of the V2 OLPF retained would probably make a lot of people happy. I don't know what all the ramifications of the color science would be though.

The base rating would still be a range with 640 probably being a very sweet spot.

I just am so loving the rolloff though... :)

Battistella
 
David Battistella said:
Its clear the DR has to be moved toward highlights or shadows with the OLPF. Phil's number of 1.5 stops seems about right. So an OLPF that lets in 3/4 of a stop with all or most of the characteristics of the V2 OLPF retained would probably make a lot of people happy. I don't know what all the ramifications of the color science would be though.

The base rating would still be a range with 640 probably being a very sweet spot.

I just am so loving the rolloff though... :)

Battistella
Click to expand...

But with olpf1 and ND .3 you got pretty much the same room for rolloff as the olpf2 the sensor range is not increased or decreased with the olpf, only moved up/down in sensitivity.
 
Björn Benckert said:
But with olpf1 and ND .3 you got pretty much the same room for rolloff as the olpf2 the sensor range is not increased or decreased with the olpf, only moved up/down in sensitivity.
Click to expand...


I'm wondering if the calibration would have to change a bit though. If it was just an nd thing it would probably have been done. Then, it won't match Alexa highlights (not that I care, but that will be the "talk") I think we might be making this a bit more simple than it is. I don't think adding a 1 stop ND to OLPF 1 is the same as OLPF 2, there are quite a few micro layer combinations going on in that OLPF. Not sure there are even many facilities who can even produce them as it's a fairly precise science compared to in front of the lens filtration.

Battistella
 
David Battistella said:
I'm wondering if the calibration would have to change a bit though. If it was just an nd thing it would probably have been done. Then, it won't match Alexa highlights (not that I care, but that will be the "talk") I think we might be making this a bit more simple than it is. I don't think adding a 1 stop ND to OLPF 1 is the same as OLPF 2, there are quite a few micro layer combinations going on in that OLPF. Not sure there are even many facilities who can even produce them as it's a fairly precise science compared to in front of the lens filtration.

Battistella
Click to expand...

I'm no olpf expert but yes I think it's far more than a nd filter. as lot of different things goes on in the olpf. But the little I know about sensors my assumption is that the falloff is not in the sensor or the olpf it's in the color science same with alexa.

So for olpf2 the colorscience is set to digitally gain the input with 1.3 stop to match the olpf1... that gives an extra 1.3 stops in the high end which is then can be sued for dynamic rounding.

this is again me guessing around. Graeme could probably explain it all in a 3 word sentence :)
 
It's not the same as ND 0,3, according to John Marchants pictures of them. They also have quite different color caracteristics.
And if I have understood Phil right, the IR cut is different, too.


The matrix has to adjust for that. And that's probably "new calibration.

I guess, though, that the hardware/firmware for re-color-calibration might be in the box now, as part of the opening for exchangeable OLPFs.
Another guess, is that OLPFs mostly will be swapped on a per production case, more than a per scene...
 
Agree.

Graeme could probably explain this pretty simply, don't know if that would mean giving away trade secrets though. :).

In any case, it will be interesting to see what is offered in the various OLPFs. I wrote this before but a lowlight OLPF should be about 4000 or 6400 ISO clean, then there would be a real reason to switch out OLPFs, because for 95 of the time, V2 will absolutely do the job.

I want a lowlight OLPF that is going to let me set the lens to 4.0 or 5.6, get the DOF I want and still provide me with clean sharp images. That would give me creative options for night/dark filming as well, stuff we really have never seen much. That would feel like going beyond rather than just playing around at the edge.

If it's an extra stop with a 1.4 lens already wide open, what is the point?

Battistella
 
Open question to all on this thread, If you were to be receiving 2 dragons in the next few weeks, what Olpfs would you choose? Does it make sense to have one of each?
A problem in buying into Red is that there is so much stuff "coming soon" but not knowing when.

I really appreciate the comments/concerns.
 
Gunleik Groven said:
It's not the same as ND 0,3, according to John Marchants pictures of them. They also have quite different color caracteristics.
And if I have understood Phil right, the IR cut is different, too.


The matrix has to adjust for that. And that's probably "new calibration.

I guess, though, that the hardware/firmware for re-color-calibration might be in the box now, as part of the opening for exchangeable OLPFs.
Another guess, is that OLPFs mostly will be swapped on a per production case, more than a per scene...
Click to expand...

Yes when I say I think it's just like ND .3 I only mean it eats about that amount of light. It for sure does a lot more than eat light as the red blobs are gone and picuture feels a lot cleaner. The red noise issues in olp2 (that deb deals with) I guess is a result of the red channel beeing cut more in the olpf2. The olpfs are for sure different different, most likly the color filtering is different in bettween the two one starve the red channel more than the other etc...

So All I'm satying is that the olpf2 has a lot of goodies that the olpf1 does not have but the olpf1 lets more light tough to the sensor.
 
scott devitte said:
Open question to all on this thread, If you were to be receiving 2 dragons in the next few weeks, what Olpfs would you choose? Does it make sense to have one of each?
A problem in buying into Red is that there is so much stuff "coming soon" but not knowing when.

I really appreciate the comments/concerns.
Click to expand...
It all depends on your needs and desires, of course. But if I were making that choice, it would be a no-brainer to choose the OLPF 2. I actually have two Dragons and sent the last one in for upgrade during all the fuss over losing DR in the shadows. Reason was because I already had a Dragon with OLPF 2 that had been sent in for changing out the OLPF 1. The color science is worth even giving up more DR in the shadows in my opinion.

And in my case, I use the cameras together so it is important to me that they match.

Oh, one more thing. I was working on some low-light footage last night and today that I pushed in on (crop and zoom.) By blowing it up even more, I could see tiny red specs that led to RED developing DEB.

Since I was working in Adobe CC with no DEB yet, I applied some Neat Video noise reduction. I then compared the two image treatments at normal viewing size and concluded I liked the footage better without the Noise Reduction. But that's just my way of looking at things. Others may have a totally different perspective.
 
Hey bjorn,

i know now thats what you meant. I think that little OLPF is a complicated bit of physics in its own special way.

Battistella



Björn Benckert said:
Yes when I say I think it's just like ND .3 I only mean it eats about that amount of light. It for sure does a lot more than eat light as the red blobs are gone and picuture feels a lot cleaner. The red noise issues in olp2 (that deb deals with) I guess is a result of the red channel beeing cut more in the olpf2. The olpfs are for sure different different, most likly the color filtering is different in bettween the two one starve the red channel more than the other etc...

So All I'm satying is that the olpf2 has a lot of goodies that the olpf1 does not have but the olpf1 lets more light tough to the sensor.
Click to expand...
 
You must log in or register to reply here.
Back
Top