|
|
11-25-2020, 02:53 PM
Hi Ignacio, with the newer RED camera's you can do up to 4 manual blackshade 'captures' in one hit. If you do that, then load each calibration up and shoot some frames with the lens cap on, you can look at each calibration (in REDCINE-X or whatever) and pick which one gives the best 'black'. There usually isn't much or any difference between each calibration/capture, but sometimes there can be.
11-25-2020, 03:03 PM
Yes, I know about the procedure, my doubts are more related to the operating temperature of the camera, not the number of captures.
Anyway I'll post tomorrow my R3d's with the black shadings at 33-38-45 degrees, with the 38º being the best of the bunch, but might be not perfect.
Last edited by Ignacio Aguilar; 11-25-2020 at 03:11 PM.
11-26-2020, 08:03 AM
I've been doing several tests with the black shadings of my camera to try to achieve the best results: https://www.dropbox.com/sh/txci5qv64...IoerQXa8a?dl=0
Camera was calibrated to 33, 38 (two different black shadings, the "alternate" is from today with the newest firmware), factory and 45 degrees.
Then I did these shots at that temperatures with the correct calibration for each one.
11-26-2020, 03:04 PM
Slightly off topic: Ignacio, when you used DSMC1 Dragon with LLO, did you try to compensate for the 1/2 stop less in highlights by physically rating/exposing (shutter, ND, iris) at ISO2000~3200 (while still having relatively clean mids/shadows, because you were using the LLO OLPF) and recovering in post?
I still trying to find the sweet spot, and currently do STND at ~ISO1280, but when I can't ETTR the lows get a little gritty.
11-26-2020, 03:49 PM
It's been more than five years since I last used Dragon I think, but I've never rated it so high. Perhaps 1280 ISO with the LL OLPF, or so. I did most of my work with that camera with the STH olpf and rated it as low ISO as possible, but pushed it to 1000 with good light levels (I mean, histogram evenly exposed at that ISO) and never found an issue, nor overexposure or clipped highlights because that's pretty easy to avoid with the goal posts.
I think that the key to expose these cameras (Gemini in standard mode behaves pretty much like Dragon) is that you can't really underexpose them if you want a dark look with clean & rich blacks. You just have to add more light or open up the iris so you can lower the ISO, but the RAW data must be evenly exposed at 800 ISO, so that the sensor is fed with enough light.
11-27-2020, 12:29 PM
I had a look at your R3D's (mainly that alternate 38ºC one) and found, when pixel-peeping, the blacks were clean up to about ISO1600-ISO3200, when .266 Contrast was applied in Resolve (and also with no contrast added but with minimal tweaking of Lift/Gamma/Gain). That was using REDWideGamut RGB/Log3G10 and Medium Contrast/Medium Roll-Off to Rec.709_BT1886.
That seems normal to me, though I haven't played with much Gemini footage in particular.
The idea behind doing the four captures at the one temperature is to see how much variation there is at any temperature, let alone between different temperatures. As well as quadruple checking the captures were done properly in the first place.
Imo the effect of the blackshade procedure can become a bit theoretical (or reach the point where if it's really that critical, maybe using a different type of camera altogether is a better idea). Not saying that's what's going on here. And obviously you'd want to use something that isn't totally off, but generally speaking, blackshading at any of the available operating temps should give good results.
Like you mention in your previous post, proper exposure plays a big part in clean & rich blacks, as well as post-processing, and staying within the calibration temperature, more so I'd say than blackshading variations (within limits).
As always, people have to decide for themselves what their criteria for a good image is.
Another fun tip is to deliberately let the temp go out of calibration while recording and have a look at what the range really is.
Anyway, good luck with it.
12-21-2020, 09:59 PM
Are these frames shot at standard ISO or for Low Light(LL) ISO?Originally Posted by Ignacio Aguilar
12-22-2020, 02:47 AM
Standard mode.
12-27-2020, 03:42 PM
I've been doing several tests and I'm a little confused (more than before doing tests, LOL) with this. When I was using Dragon, I always felt 800 ISO on Alexa seemed to be brighter than 800 ISO on Dragon. In fact I was kinda shocked when I was shooting night interiors on Alexa at T4 when I used to have the same light levels on Dragon and be around T2.8. So Alexa 800 ISO looked twice as fast as Dragon, or Dragon 800 looked like Alexa 400 (which I think was the case).
If I check the feature "use updated ISO calibration" on my Red Gemini, the images are brighter at the same ISO setting than with the same feature left unchecked. If I'm not wrong, that was an update to make Red cameras i) fully compatible with light meters, which should be more accurate than before and ii) to make Red cameras as bright as others (such as Alexas).
Question is, should I use the updated ISO calibration or not? I did my feature film leaving it checked and exposed mostly with the in-camera tools. But I've been testing my light meter and it gives me very precise medium grays (green video check areas) when I leave the updated ISO calibration option unchecked. And if I use the updated ISO calibration, exposures don't match anymore. I believe it should exactly be the opposite!
Any ideas?
12-27-2020, 04:15 PM
If you are in IPP2, shouldn’t the newer more accurate iso settings already be automatic?
Or are you using the Legacy color?
| « Previous Thread | Next Thread » |
