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 :)

Overly Noisy RED One MX

Shawn Seifert

Shawn Seifert

Well-known member
Joined
Jun 9, 2009
Messages
50
Reaction score
0
Points
0
Location
Vancouver, BC
Hi Red community.

I've been working with a particular Red one (MX) which I'm thinking is overly noisy. Like a malfunctioning sensor. This became noticeable to me when trying to grade some footage and was having a really hard time trying to grab an individual colour. This camera was not nearly as noisy with with old M sensor. When shooting with this camera we would use ISO 320 with the M sensor and ISO 800 with the MX as recommended.


I have taken a few test shots for your consideration. I can also provide a couple of r3ds of this footage if need be.


Sorry for the massive size of the pictures. I wanted to give you guys the full size and I'm not sure how to attach them as a gallery.

So am I crazy, or is there an issue? Because the footage from the last project was a nightmare to colour grade.


ISO 320
ISO 320.jpg


ISO 800
ISO 800.jpg

ISO 1280

ISO 1280.jpg
 

Attachments

  • ISO 320.jpg
    ISO 320.jpg
    60.8 KB · Views: 1
Would have been better to have single frame r3d's...

But It doesn't neccesarily look wrong at 800/3200K

I think it is often not a good idea to develop at 800, for a number of reasons. This being one of them...

Shooting at 800 is a good plan, though...

Edit:
How gradeable the shots are, depends highly on how you developed them, though...
 
Those were snapshots out of RedCine-x, ungraded. Originally outputted as .tifs and then compressed to .jpg (at the highest settings) for you guys to see.

I guess to me I am just seeing a lot more noise then I'm used to.
 
Or you could try to go redlogFilm at 320 before grade, and check that you don't clamp any of the channels with the White Ballancing...
 
1st off... the name of this thread suggests there is a problem. And you are really asking if there is a problem?

A couple of suggestions.

1. Make sure to run the Black Shading calibration.

2. It appears that you are using REDcolor and REDgamma. Use REDcolor2 and REDgamma2. Your color grading experience will be better.

3. There is a pretty good amount of tint in your lighting. As long as you know.

4. These images look good to me. Even if I set FLUT to 6 on the ISO 1280 image (ISO 80,000) the images look reasonable.

Jim
 
Hi Jim,

Thank you for your response. It's great to have the owner of a company responding himself.

Firstly, I guess I should have put a question mark at the end of my thread name. As in, "is this footage looking overly noisy? And if it is, is that an issue with the camera?" So yes, I guess I was asking if there was a problem... (or had I just been staring at the footage for too long and seeing things that are not actually there?)

In my opinion, the footage has been looking more noisy to me since the camera was upgraded. It's not my camera, but one I work with frequently, so I just wanted to get a few other peoples opinions on the matter.

As to the rest of your questions...

I ran a black shading right before this was shot.

I will be grading in Redcolor2 and RedGamma2. I really like what they do with the image and will no doubt lead to a final product I am very happy with.

I'm not sure about the tint, but it is probably spill from some florescent lights on the other side of the studio. (Although I have noticed, that I often have to muck with the colour temperature settings a bit to get a image that doesn't appear to have a slight tint.)

I guess a more specific example of what I am talking about will help solve the issue (or point out that all is fine with the camera).

This is a 200% blow up of footage shot on the M sensor (320 iso).
M- 320iso.jpg


200% from the MX at 320 iso
MX- 300iso.jpg


200% from the MX at 800 iso.
MX - 800iso.jpg


I know there is lots of variety in these images due to them being shot months apart and with different lighting subject matter, etc.
However, to my eye, the image looks cleanest on the M sensor at 320 iso. Pretty close up would be the MX at 320 and finally the MX at 800.

I realize that 320 iso should be cleaner then 800 on the new sensor (but that exposing at 800 iso is best due to it being the sweet spot in the noise floor), but I was under the impression from those on this forum and within the industry in my city that 800 iso on the MX should give better results than 320 iso on M. Is this correct? And if it is, am I interpreting the footage examples above incorrectly?

I don't want to cause a stir, I just want to know if this is normal looking footage and that all is good with the camera.
I don't post much here, but trust me I spend lots of time reading on this forum.

Thank-you Jim, and everyone else, for taking the time to help. :smiley:
 
The images look like they're under completely different light sources and exposures making it hard to compare.
 
It seems like you are comparing Daylight shot M sensor @320 to Tungsten MX @800. In this case, I'd expect the M sensor to look a bit cleaner. I've always found that shooting daylight makes a bigger difference then shooting low ISO.

From my testing I would say that Daylight M @320 will match Tungsten MX @500 or Daylight MX @800.
 
Deanan, you bring up a great point. Unfortunately, I have limited footage on hand from the Mysterium days of the camera.
Here is a cropped shot from the very first thing I shot on this Red One. It is indoors, using tungsten light at 3200. So it is as close of a match as I can get to the new MX footage. I would love to be able to provide the same shot with this camera, before and after the upgrade, but unfortunately I don't have that. It not being my camera and all.

I am purposely choosing parts of the image that are out of focus because that is where I am seeing the noise the most prominently.

M sensor @ 320iso at 3200 colour temp.
M 320iso 2.jpg


Sam, I had not really taken the differences in color temp into account, because I had heard that with the new colour sciences, that the differences between daylight and tungsten, were all but gone. I'm guessing I really need to go and do some more specific tests like your own to figure out what works really best. Thank you for your helpful notes.
 
Shawn... you said you "will" be grading in REDcolor2 and REDgamma2... my bet is that these images are not. Please regrade these images with those settings before continuing this thread.

Jim
 
Looks like overdue black shading, or badly done black shading to me.
 
You must log in or register to reply here.
Back
Top