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 :)

First Day with Dragon

I think that's a fair point, Gavin, I should have said shoulder rolloff instead of clipping.

Obviously, it's a very unscientific frame grab. Different lenses, different brightness on the back wall, etc.

But that aside, the noise on the Dragon is a little rough and the color cast on the wall is noteworthy. Something the Alexa has always done so nicely, is capture balanced color in the shadows and highlights.

But thank you for pointing that out.
 
Last edited:
Mark Wilkinson said:
But that aside, the noise on the Dragon is a little rough and the color cast on the wall is noteworthy. Something the Alexa has always done so nicely, is capture ban aced color in the shadows and highlights.
Click to expand...

I think its good feedback for Red with samples they can look at. Thank you for sharing. The more example we can give them, the better things get.
 
Mark Wilkinson said:
I think that's a fair point, Gavin, I should have said shoulder rolloff instead of clipping.

Obviously, it's a very unscientific frame grab. Different lenses, different brightness on the back wall, etc.
Click to expand...

There's certainly a lot of interesting comparisons to be made in the responses of the two cameras, but I wouldn't be too hard on the shoulder rolloff since the Alexa is doing its Low-con action that it's famous for. The largest problem seems to be (and I saw this in the dynamic range testing) RED is clamping the channel on Dragon in the shadows, Alexa does a much better job of the noise gradating to black instead of being ON/OFF. This is creating very choppy noise. I think it's a calibration issue. In the Alexa there is actually a pretty good Green tint in the very darkest shadows. I think the red coloration in the shirt is actually a lens flare from behind his head. Once you apply so many stops of gain, even imperceptible lens flare become apparent.

That being said, it's certainly possible to match grade somewhat closely, a "recalibration" if you will. So hopefully RED can take a look at doing just such a thing.



Any improvements in color science from RED would be welcome by me.
 
All pretty much agreed, Gavin.

I think the parade tells us a lot about the detail we're losing in the noisy dark on the Dragon.

It did seem less an issue in the last calibration, as many have mentioned. Let's hope it goes away.

And the cast has gotten very minimal and well within the range of todays Grading options. Still. You, know. Better if they just fix it.
 
The choppy noise in the blacks is a total killer with the dragon especially with high shutter speeds above 1/120, i'm sadly missing my MX sensor right now :-). Hope it a firmware fix and not something we need to send back for a recalibration.
 
Wow - that's an interesting idea - why would shutter angle effect noise?
 
Matt Ryan said:
I've noticed on the new olpf/calibration that there appears to be a lot of red noise in underexposed footage. I'm betting that can be fixed by firmware/software update. I do agree though that Alexa has the upper hand on the low end. Lot less noise with Alexa...
Click to expand...

What's funny is, when testing Dragon with OLD OLPF, didn't Geoff Boyle specifically say that Dragon was noticeably CLEANER than Alexa? Has something happened in transition from old OLPF to new?
 
Nick Morrison said:
What's funny is, when testing Dragon with OLD OLPF, didn't Geoff Boyle specifically say that Dragon was noticeably CLEANER than Alexa? Has something happened in transition from old OLPF to new?
Click to expand...

He did say that. Actually he said Dragon was a lot cleaner. No idea what happened calibration wise but I miss the old calibration to be honest. Low end is substantially noisier now and has a red tint to it vs. the old which was neutral in color. Hoping RED will come up with a new calibration or firmware update to fix this. At least it seems to be that way with everyones camera's so it's most likely calibration and not a fluke thing.
 
Nick Morrison said:
What's funny is, when testing Dragon with OLD OLPF, didn't Geoff Boyle specifically say that Dragon was noticeably CLEANER than Alexa? Has something happened in transition from old OLPF to new?
Click to expand...

I remember that - another moving target, I think. And hope.
 
Mark Wilkinson said:
Wow - that's an interesting idea - why would shutter angle effect noise?
Click to expand...

It's possible RED is off a bit with the black shading math. The black shading map is an image of pixel intensities over a number of exposures. So if you blackshade at 1/48th of a second and the Fixed Pattern Noise is 0.001 intensity then while filming if you read 0.002 you'll get a pixel intensity of 0.002 - 0.001 = 0.001. Presumably if you change the shutter angle to 1/96th of a second your FPN only is half 0.0005. But it's possible that the FPN is slightly mischaracterized by RED and the blackshading is actually subtracing too much. So now instead of 0.0005 - 0.0005 = 0.0 your Fixed Pattern Noise might be 0.0001 at 1/96th of a second and when it subtracts the extrapolated FPN value of 0.0005 you end up with -0.0006 and the noise looks crushed and chunky. No idea if it's the reason, but it's a possible explanation of how shutter angle could result in crunchy noise.

Whatever the cause, I would like to see RED find a fix.
 
Been digging around the web and have found the old calibration w/ the old OPLF stills that people provided are WAY WAY cleaner then what i'm seeing now. Could someone from red chime and clear this issue up with us, i have a shoot coming up and really want this camera to kick ass like i promised, but at the moment our client and am debating about just renting a MX.
 
Last edited:
are guys that got their camera upgrade from the old to the new calibration seeing these issues as well? Could it be just a bad batch of sensors? or is just not a lot of people shooting high shutters or not seeing the noise. Sorry to be so loud about this but i've spent the last 48hr with no sleep trying to figure this whole thing out.
 
jacob.schwarz said:
Been digging around the web and have found the old calibration w/ the old OPLF stills that people provided are WAY WAY cleaner then what i'm seeing now. Could someone from red chime and clear this issue up with us, i have a shoot coming up and really want this camera to kick ass like i promised, but at the moment our client and am debating about just renting a MX.
Click to expand...

That's why I'd love to have the old calibration with the new, non flaring olpf. To me the new calibration is too noisy on the low end.
 
What ever they do they need to come out with a solution fast! I need to get this camera battle ready ASAP! I've been talking to red, but haven't gotten far with them.
 
Matt Ryan said:
That's why I'd love to have the old calibration with the new, non flaring olpf. To me the new calibration is too noisy on the low end.
Click to expand...

I can confirm what you say Matt, the old calibration OLPF and blacks had the cleanest.
But I can also say that now the new calibration and OLPF is better in the high lights ..
I'm sure Red will improve this over time. :)
 
luigivaltulini said:
I can confirm what you say Matt, the old calibration OLPF and blacks had the cleanest.
But I can also say that now the new calibration and OLPF is better in the high lights ..
I'm sure Red will improve this over time. :)
Click to expand...

Yeah, highlights are about a stop better. But I protect my highlights when I shoot so I didn't have a problem with the old calibration haha. I just like cleaner blacks personally.
 
You must log in or register to reply here.
Back
Top