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 :)

OLPF v1 and v2

J

Jake Wilganowski

Well-known member
Joined
Nov 9, 2008
Messages
2,860
Reaction score
0
Points
0
Location
Dallas, TX
So it looks like I may have nabbed one of the discounted epic dragons (used in the military shoot) that Jarred put up for sale on his Facebook, so I need to decide which OLPF to get. Ive looked at all the tests here, but wanted to ask some simple questions to people who've used both. Ive only shot with version 2 personally and it was an outside shoot.


1) How often does the magenta orb issue rear its head? I shoot a lot of music videos with bright lights/flares in frame, is this going to happen all the time? Does it happen if I'm shooting with a bright a window in frame in a narrative setting for example? Would moving the camera slightly or changing the angle slightly remedy this pretty easily?

2) Does rating the V1 OLPF at 800 produce the same rolloff as shooting v2 at 250ish after processing, or will v2 aways have a superior tonal gradient up in the highs? Meaning, if i expose darker on the V1 OLPF, then correct in post, will I get a nice rolloff superior to MX and equal to that of V2?

3) Do i need IR filtration with V1 OLPF, and after how much ND?

4) Final simple question, is V1 OLPF dragon superior to MX in every way? Or just in every way besides the Orb issue?

THANKS!
 
Jake Wilganowski said:
So it looks like I may have nabbed one of the discounted epic dragons (used in the military shoot) that Jarred put up for sale on his Facebook, so I need to decide which OLPF to get. Ive looked at all the tests here, but wanted to ask some simple questions to people who've used both. Ive only shot with version 2 personally and it was an outside shoot.


1) How often does the magenta orb issue rear its head? I shoot a lot of music videos with bright lights/flares in frame, is this going to happen all the time? Does it happen if I'm shooting with a bright a window in frame in a narrative setting for example? Would moving the camera slightly or changing the angle slightly remedy this pretty easily?

2) Does rating the V1 OLPF at 800 produce the same rolloff as shooting v2 at 250ish after processing, or will v2 aways have a superior tonal gradient up in the highs? Meaning, if i expose darker on the V1 OLPF, then correct in post, will I get a nice rolloff superior to MX and equal to that of V2?

3) Do i need IR filtration with V1 OLPF, and after how much ND?

4) Final simple question, is V1 OLPF dragon superior to MX in every way? Or just in every way besides the Orb issue?

THANKS!
Click to expand...

1. Magenta Highlight Orb, It's a reflected effect that dangles off of in frame hot sources like headlights, flashlights, and any similar source. It can also produce an interesting "glow" to highlights.

2. No. They are mapped differently and calibrated the same (though you can shoot and process however you want). Meaning the image you get at ISO 800 is correct for exposed Middle Gray for both OLPFs. The difference falls in approximately 1.3 stops of improved overall sensitivity for the V1 (how far it sees into the shadows and cleaner higher ISOs). Whereas the V2 OLPF pushes that 1.3 stops into highlight retention. That said, both versions of the OLPF have nice highlight roll-off. V2 is more or less "more film-like"

3. Not necessarily. The V2 does a better job at cutting IR and is superior to the V1 in terms of color science and accuracy. If you look at the example here (the shadows on the shirt) you can see where IR comes into play at ND 2.7 in the shadows and that's more or less the visual difference:
phfx_redDragon_NewOLPF_ChartDaylight.jpg


4. I would say the V1 OLPF is more or less similar to Mysterium-X as there are similar artifacts in terms of higher aperture reflected sensor dots. The cleaner available higher ISOs could be considered superior however.

5. If you haven't already look at this post here with the examples between the Version 1 and Version 2 Dragon OLPF:
http://www.reduser.net/forum/showthread.php?112230-the-New-Dragon-OLPF

6. Congratulations!
 
Jake Wilganowski said:
2) Does rating the V1 OLPF at 800 produce the same rolloff as shooting v2 at 250ish after processing, or will v2 aways have a superior tonal gradient up in the highs? Meaning, if i expose darker on the V1 OLPF, then correct in post, will I get a nice rolloff superior to MX and equal to that of V2?


THANKS!
Click to expand...
From my way of thinking about exposure, I believe you are correct in thinking that the "rolloff" (what i would call highlight latitude) would be the same. If you expose dragon v2 at 250 iso you are moving your mid gray point higher up the grayscale. As you do so you lose highlight latitude between your new mid gray and clipping point. Since your new mid gray point is 1.6 stops higher up the scale (compared to mid gray at 800 iso) you have effectively chopped off 1.6 stops of latitude above your new mid gray, which makes your highlight latitude about the same as v1 dragon. Except now you get that same highlight latitude at 250 iso. Assuming that Phil is correct and that v2 dragon has 1.3 stops more highlight latitude than v1 dragon, exposing v2 dragon at 320 iso should give you the exact same highlight latitude that v1 dragon gives you. I think you are also correct in that you could go the other way and expose v1 at a higher iso and extend the highlight latitude that way. Theoretically you would have to expose v1 at 2000 iso to match the highlight latitude of v2 at 800 iso. That would be an interesting test for someone to make to see if noise is the same between v1 at 2000 and v2 at 800 iso.
 
Yes this is what I'm talking about. Anyone else care to weigh in? Phil?
 
I get the theory and it does kinda work when it comes to exposure but you are leaving out important aspects of the other improvements with v2 so when you try to extend the highlight latitude of the v1 then you are taking it into a different direction in terms of visual artifacts.

Plus the optical path physically changes between v1 and v2 so that has an overall effect on color fidelity, etc, in relation to how the calibration of the camera is set relative to the OLPF.

I imagine red is devoting most of its calibration effort towards the "shipping" OLPF and not towards the one that has been replaced in favor of highlight latitude. (That's just my thinking, I don't know if that is true)

so while this argument might hold true for light levels, there is more to consider than just the measured light because the optical path does physically change and that changes things like color.

Battistella
 
But, BESIDES color improvements and flare improvements, version 2 is basically remapping middle grey so that 800 on both olpf's looks the same brightness correct? But in version 2 the sensor is actually receiving less light so it's is noisier.
Am I getting it?
 
Jake Wilganowski said:
But, BESIDES color improvements and flare improvements, version 2 is basically remapping middle grey so that 800 on both olpf's looks the same brightness correct? But in version 2 the sensor is actually receiving less light so it's is noisier.
Am I getting it?
Click to expand...

The bias is toward the highlights in V2, I don't think its an ND thing as much as when you have to introduce elements into the OLPF that eliminate big problems (like the RED Dot noise so prevalent in Phil's excellent tests). This requires filtration of specific light spectrums.

For example V2 contains an excellent built in IR filter (remember when everyone was bitching about IR problems on RED and what it does to skin tones.

It's a whole bunch of things so it would be oversimplifying it to say that one has more ND over another. There is also a moire filter in there as well. What an OLPF seeks to do, and what this one does quite well, is eliminate problems in the optical path (moire, IR contamination) before it hits the sensor. Then calibration of the sensor is the next part of the equation.

There are many people here who understand this much better than I do (John Marchant, Phil, etc) but the fact is that it is a science and there are many pieces to the puzzle, everything you move creates a new set of problems. So, to think that it's just a bit of ND is oversimplification because it's a triangulation of things that produces a desired result.

There might be many who think that this is some kind of simple thing, but I think if it were simple the solution would be quite rapid.

FWIW, i think RED have a very solid OLPF in V2 and combined with the SW solutions happening this makes it FAR superior and more flexible imagining device than the V1 which was demonstrating problems that were carrying over from the EPIC MX, that big list of complains, Skintone, sensitivity, color accuracy, red dot noise, etc, that this OLPF just flat out eliminates.

OK, red dot shadow noise was a by product, which shows up at specific exposures, but there is a tool to turn that ON or OFF depending on your needs. Some might argue that makes the system weak, id say it makes it more flexible and allows you to extend creativity. Some want it "in camera", its clear its not needed on every shot. Many argued for sharpening in camera on the RED One and RED did not move on that (thankfully) so I sorta see this the same way.

Some want it for the video monitoring path (you are already getting a FULL debater of a 6K frame there so there shouldn't be too much "noise" leftover.

The modular idea is going to give people the chance to:

1. Install one OLPF and stick with it.
2. Swap to a specialized OLPF when called for.

I don't think many will be swapping OLPFs daily, but some people will.

As far as lowlight goes, there is a difference between V1 and V2, but that is becoming increasingly small as the SW tools and calibrations are quickly catching up and surpassing the V1.

Battistella
 
Without trying to define it in specific, V2 just plain has better color rendition. Full stop. To me, V1 was like a sneak preview of what Dragon is capable of, but with too many "veils" like the magenta blooms and the subtle color crosstalk they created, etc (plenty of threads about the specifics if you're on the due diligence track).

If RED comes out with the proposed swappable OLPF system and it includes a low light biased option, let's call it V1 with a few tweaks - V1.1 perhaps? - I'd love to have it in my kit for those times when the situation makes under-exposure unavoidable. Other than that, V2 all the way. I shot some footy yesterday that had 3 subjects in frame, all of whom had distinct skin tones, and each one was faithfully reproduced with the same global settings. Moreover, the transition of the skin tones into the rim light held hue beautifully with no discernible color casts.

V2 is the droid you're looking for.

Cheers - #19
 
I really wish people would stop referring to the loss of 1.3 stops due to OLPF 2 as "Highlight Retention". It's rhetoric employed to sound like an increase in dynamic range is achieved. It's only highlight retention if there's too much light on set to deal with, and someone forgot to bring the ND filters (sorry - I mean Highlight Retention Filters).
 
Roger Viloria said:
Jake, here there are r3d's from this quick test, nothing fancy. Similar situation/T stops.

OLPF V1
https://www.dropbox.com/s/km1td9dnyc558v5/OLPF V1 A001_C007_082354.0000961F.R3D.zip?dl=0

OLPF V2
https://www.dropbox.com/s/6w1nnk7mow4p2yh/OLPF V2 A001_C004_08142W.0000000F.R3D.zip?dl=0
Click to expand...

Thanks. Thats definitely a significant difference in noise. Ive shot the v2 OLPF in a dark barn and got clean blacks, but I rated the cam at 250.

This is a hard decision. The DP in me says get V2 no question. But the practical business side of me says 1.3 stops cleaner is SO useful in lower budget stuff or practically lit environments.
 
Phil Holland said:
you can see where IR comes into play at ND 2.7 in the shadows and that's more or less the visual difference:
phfx_redDragon_NewOLPF_ChartDaylight.jpg
Click to expand...

Nine stops of ND added and vulnerable clothing in daylight is needed to spot the improvement in color?
For now I'm with Roger.
The reduction of ability to dig into shadows with OLPF 2, unless you have tons of light (setting ISO low) is the real issue here, for the majority of challenging shots. Roger's example above makes that pretty clear.

I'm guessing there'd be an improvement in the new OLPF in rendering red traffic lights or a red neon source.
A post from John Marchant showing various OLPFs illustrated the strong cyan cast of the new one.
I guess that means the red channel clips less easily.
I'd love to see a comparison of the two OLPFs with subjects like these two.
 
For what it's worth. I'm thinking of going back to V1. Everything was beautiful back then. I've had some struggles with V2, but they are my struggles, not the OLPF. I don't have expensive lights, I'm shooting in the darkness which is why I'd consider it.

If you give the Dragon enough light, it's a killer camera.
 
Will Keir said:
For what it's worth. I'm thinking of going back to V1. Everything was beautiful back then. I've had some struggles with V2, but they are my struggles, not the OLPF. I don't have expensive lights, I'm shooting in the darkness which is why I'd consider it.

If you give the Dragon enough light, it's a killer camera.
Click to expand...

There has been a few people saying the V1 is not available right now. This might be a sign a replacement is coming. I cant believe Red would manufacture more V1`s with the orb issues. Fingers crossed.
 
Really? I'm a two camera Dragon shooter and I'm sick…….20K worth of upgrades and I've got a camera which was advertised as being the end all to end all. What I've got is two cameras which I need to sell and defend to producers (in the know) and DIT's (in the know). Red team, please help…Lets get this right. What we have here is a experimental camera with potential. We're a long way from it being what was hyped.
 
You must log in or register to reply here.
Back
Top