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 :)

RELEASED: Beta Build 30 Firmware (v30.4.1)

What Milan is seeing is exactly what Leo and I saw.
 
Build 30 (Mysterium-X and Mysterium)

Build 30 (Mysterium-X and Mysterium)

After the GREAT daylight tests with B30, I'm now having similar problems as Leo when attempting to shoot under tungsten.
Click to expand...

We have shot ton of stuff under tungsten, so can you share the settings of your post production software Milan?
 
Some clarity, please.

Milan's shots are exactly what we saw on the monitors from the camera.

Stuart, are you telling us we must apply a look to the camera rather than shoot everything at nominal?

When you did your tests did you have a custom look applied to the camera, or was everything at default?

Best regards,

Leo
 
Stuart English said:
We have shot ton of stuff under tungsten, so can you share the settings of your post production software Milan?
Click to expand...

Hi Stuart,

I opened the clip in Redcine-X (B104) and took a snapshot. Haven't touched any settings, it's what came out of the camera.

After originally having problems in B30 I reverted to B21, then back to B30 to take that snapshot. Other than the ISO and color temperature, I haven't touched any other settings in camera before recording.

I really tried to make it work, though looking back at my daylight samples this morning, they seem to have also suffered from reduced DR.
I'm really liking B30, hope it's an easy fix or even better, an user error. ;)

Cheers,

Milan
 
Last edited:
I am having absolute joy with build 30 However not used it with Tungsten lighting.
I have shot 2:1 for the first time on a shoot last week (always shot 16:9 or 1:85 due to Agency and client requests) I was really impressed with the 120fps has anyone else noticed it looks a lot slower or is it just my imagination. I have shot a lot of 435 at 150fps and red looks slower. I am no sycophant but it really does look slower and my camera crew all feel the same. I will do a test, but shooting falcons at 120 fps and a 45 degree shutter angle the footage was spectacular
 
Leo Ticheli said:
Some clarity, please.

Milan's shots are exactly what we saw on the monitors from the camera.

Stuart, are you telling us we must apply a look to the camera rather than shoot everything at nominal?

Leo
Click to expand...

Stuarts asking for more info so we can figure out what's different in your case.

Can you send us r3ds from the two builds?
 
Milan Spasic said:
After the GREAT daylight tests with B30, I'm now having similar problems as Leo when attempting to shoot under tungsten. Severe loss of DR and overall brightness at any ISO plus inability to balance for tungsten. Looked like the blue channel was completely absent. But the loss of DR and overall brightness was the most worrying. :(

I don't know if this is related to tungsten or something simply went wrong between this morning tests and right now (I haven't changed any settings). Hope it's just a simple bug.

Just reverted to B21 and everything looks great again. :shrug:

edit: I performed the black shading in B30 several times to no avail.



B_30.4.1 / 3200 / ISO 800 / T3
4431579426_9b3657017c.jpg


B_21.4.1 / 3200 / ISO 800 / T3
4431601414_c66281e2b7.jpg


Cheers,

Milan
Click to expand...

Stick a chart with an 18% grey up in that scene and send me an R3D please. Most likely cause is that light is not 3200k.

Graeme
 
To add on Tungsten colour temperature. Calibrating accurate correlated colour temperatures is non-trivial, so I've been putting extra work into that for B30 calibration across M and M-X.

I have a calibrated tungsten light source of 2856K (no tint as it lies on the black body radiator curve). Under B20 the camera thinks it's 2553K, but under B30 it correctly gets within a few kelvin of the correct 2856k. However, my Sekonic colour temp meter (C500) thinks it's 2490K. Similarly my Lowel Tota is 3156, B20 though it 2843K, but now with B30 it reads as 3209K, and the Sekonic thinks 2770k.

The visual effect of colour temperature changes very rapidly in the tungsten end of things. 100k makes a big difference at 2800k, whereas it makes a minor difference at around 7000k.

Graeme
 
Graeme Nattress said:
Stick a chart with an 18% grey up in that scene and send me an R3D please. Most likely cause is that light is not 3200k.

Graeme
Click to expand...

I will do if I get some free time, I'm currently back in B21.

Even if the light is not 3200k, why would it look OK in B21 at 3200k but not in B30?

I have pointed the camera at several different bulbs in the house with the same result. The color looks fairly accurate in the highlights but as the light falls off (and it falls off quickly!) it seems to be going orange.

But like I said, the most concerning to me was the loss of DR and general muddiness of the image.

BTW, the Auto WB reading off of that white wall behind the lamp was around 2100k and the image still looked terrible. Really green in RedColor and super orange in RAW.
 
Send me the R3D as is for now, and if you get a chance, shoot one with that light and a chart is shot - doesn't matter on the build as I can decode it here as if it were different builds to see what is going on.

Graeme
 
Graeme Nattress said:
Send me the R3D as is for now, and if you get a chance, shoot one with that light and a chart is shot - doesn't matter on the build as I can decode it here as if it were different builds to see what is going on.

Graeme
Click to expand...

I sent it through the support page for your attention, thanks Graeme.

Cheers,

Milan


EDIT: The file upload failed, let me know if I can send it some other way.
 
Last edited:
Attributing this extreme contrast to color temperature is interesting to say the least. My shot was using my standard Mole Richardson 5K with a Chimera, a lighting setup I've been using for decades with cameras ranging from 35MM film, an HL79, and most modern HD cameras.

I've never seen anything like this before and we're not talking something trivial; on my camera with everything set to default, Build 30 is simply not usable. I wouldn't dare let a client look at the monitor with Build 30 on my camera with the default settings.

If the studio is clear tomorrow, I'll shoot tests with both builds and send them along. If we're booked, I'll shoot the test as soon as I can.

I still want to know if you are shooting with settings other than nominal. Answer please!

Best regards,

Leo
 
Leo Ticheli said:
Attributing this extreme contrast to color temperature is interesting to say the least. My shot was using my standard Mole Richardson 5K with a Chimera, a lighting setup I've been using for decades with cameras ranging from 35MM film, an HL79, and most modern HD cameras.

I've never seen anything like this before and we're not talking something trivial; on my camera with everything set to default, Build 30 is simply not usable. I wouldn't dare let a client look at the monitor with Build 30 on my camera with the default settings.

If the studio is clear tomorrow, I'll shoot tests with both builds and send them along. If we're booked, I'll shoot the test as soon as I can.

I still want to know if you are shooting with settings other than nominal. Answer please!

Best regards,

Leo
Click to expand...

Remember that this is beta firmware and if you're seeing something different than we are, we want to figure out why. Figuring out why you're seeing something different is the goal rather than the drama about it. What we'd like to get from you are the log files, and sample R3Ds from both builds and a grey or color chart if you can. This is why we put out betas.
 
Last edited:
Yes, Deanan will give you the upload coords and we'll be both looking at the file to figure it out.

Graeme
 
I've got the same poor results as Leo, and it's got nothing to do with color temp. It's under daylight.

As soon as upgrading was done I noticed a total lack of sensor noise, with the cap over the sensor.

Black Shading was repeated several times

Simply, shadows appear crushed, and are un-recoverable in post. Raising the ISO setting in Redcine X or Redalert, even to 6400 leaves the shadows down at zero.
The Lows need to be raised up to give anything like an attractive curve, and sensor noise becomes evident.

This is the first time I've ever upgraded the firmware and had negative results with image quality. And the difference here from the previous build is dramatic. I'd strongly advise against using this build, and I think it should be withdrawn from the downloads page.
 
Just with M as far as we know. We have the team working on this issue.

Graeme
 
You must log in or register to reply here.
Back
Top