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 :)

How high do you rate MX?

R

Rob Ruffo

Well-known member
Joined
Dec 28, 2009
Messages
2,493
Reaction score
0
Points
0
Location
Montreal
A survey...

1 - With Neat video or other
2 - Without Neat Video or other.

From what I see, I think ISO 2000 is quite safe, but I'd love to hear your thoughts. Output will be 1080p/broadcast. (I know, broadcast is 1080i, but it's really 1080p, split into fields for transmission and then reassembled into progressive frames by the user display)
 
i'm quite pleased shooting at 1600 and in some cases even at 2000. in fact shot at 3200 a few weeks ago and i'm ok with it. Every situation of course is different.
Here's a link to some tests i shot: http://vimeo.com/14346399 keep in mind that they are compressed by vimeo and only one pass.
 
Most of my DP clients still choose to rate the camera at 320 ASA under normal circumstances. For some night exterior shots we have frequently rated as high as 800-1000 ASA. I recently used 1600 ASA in a very low light exterior situation and found the "noise" comparable to shooting Kodak Vision 800T film. That is to say, noisy, but certainly not intolerable. If anything, the noise at that speed was slightly more subtle than film grain.

- Jordan
 
PageImage-483134-1732715-C007_C081_0827P9_S0000000871.jpg


That is rated at around a thousand, shooting tungsten. I find that I tend to rate around 800 to a thousand rather than 320 because the highlight latitude is better - especially for something with practicals etc in the frame.

PageImage-483134-1714596-4903352027_6349fe9226_b.jpg


This is at 2000 iso. Notice that the bounce on my face is coming from the white wall on frame right, but there is tons of detail in the wall because of the increased latitude at 2000. True, its verging on noisey at that iso, but in a high contrast image where you might crush the noise into the black such as this, its not really an issue. This image could be crushed further and look even cleaner.

I havent tried neat video yet, but Im downloading as I type. Ive been very sceptical of noise reduction programs, but heard good things about neat video. Ill post again with findings.
 
Those shots look great. I don't see the noise as a problem at all. Interesting that the highlight latitude is better higher up.
 
exactly. And if you want to subtly blow out something like a lamp, it looks so much better if it isnt clipped in the raw, but merely with a curve in post. You can build in a filmic shoulder more or less.

I tried Neat video, btw. It makes things scary clean without making them too plasticky. I actually missed the noise when it was gone. The craziest part - I was hunting through my harddrive for some MX footage that would be a good candidate, and I realized I didn't have a single shot so far that had objectionable noise. Wow.
 
On the show I'm working on now we shoot 300 gb to 1 tb a day at 800, 1000, and sometimes 1600 ISO. Never lower than that as there's no noise problem with a full debayer on 800 or 1000. We use light noise reduction above 1000 ISO.
 
Neat Video Sample

Neat Video Sample

Here are some low light samples- I went extreme and went with ISO 4000 and 6400 and then cleaned it up with Neat Video: http://www.vimeo.com/13552888
 
This is probably a stupid question, but what do you mean, "meter" and "histogram" on these different tests? Exposing via a light meter versus exposing via the internal histogram?

The tests are very interesting, I would like to know more about them. . .

What do you think of neatvideo???

Thanks.

Stephen
 
First, thanks to all - great survey! I'm glad I started this thread!

Oe thing for all to remember is that ISO is not the only factor to increase noise in darks - a high gamma curve will do the same.

To Stephen: Neat Video is the best noise reducer I've found, at any price, and at $100 U.S.D. it's basically free.
 
Can someone explain to me how rating the RED at a higher ISO (800, 1000, etc.) gives you more latitude? I could be completely nuts here, but it was my understanding that ISO had exactly NOTHING to do with the noise in the RED's RAW image. The ISO is ALWAYS 320, regardless of what you set it at.

Thus, all you're doing is seeing a brighter image, which, to my way of thinking, anyway, has absolutely nothing to do with latitude.

Or am I nuts? (Again.)

Again, what possible latitude gain can come from rating a native 320 sensor at a 1200 ISO?

BTW, great, great, shots here! I'm really envious!!!

Stephen
 
Stephen Pruitt said:
Can someone explain to me how rating the RED at a higher ISO (800, 1000, etc.) gives you more latitude? I could be completely nuts here, but it was my understanding that ISO had exactly NOTHING to do with the noise in the RED's RAW image. The ISO is ALWAYS 320, regardless of what you set it at.

Thus, all you're doing is seeing a brighter image, which, to my way of thinking, anyway, has absolutely nothing to do with latitude.

Or am I nuts? (Again.)

Again, what possible latitude gain can come from rating a native 320 sensor at a 1200 ISO?

BTW, great, great, shots here! I'm really envious!!!

Stephen
Click to expand...

Yes, the sensor is always recording the Raw at around 320. By "Rating" it at 1280 though, you will stop the lens down two stops to get the same exposure as 320. These two stops prevent things (especially things over which you have minimal control over) from clipping.
 
When you rate the sensor at 1000 ISO, for example, you're going to 'underexpose' relative to what you would expose for at 320 ISO. When you underexpose, your highlights are left intact and you can build a very nice filmic curve that blooms the highlights out organically. Because the sensor is extremely clean in the shadows, bringing up your lows and mids will not bring out noise. Leaving you with a very nice image with lots of latitude.

The key is that you can 'underexpose' relative to 320 much more with MX than you could with the old sensor. Underexposing to retain all your highlight detail is key... if that's the look you're going for.
 
Call me stupid, but I still don't get it. We can look right at false color and know exactly what's blowing. . . or look at the histogram or goalposts or stoplight and know that something, somewhere is blowing. Aren't you guys just fooling yourselves into trying to tell yourself something that your brain already knows isn't true?

Apparently, for this to work, you are strictly exposing from the LCD or EVF APPEARANCE rather than the metering. I, personally, have never been able to trust the visuals that much. What my eyes see as "fine" I sometimes find out later was "blow." But the meter never lies. It's either out of bits or it's not.

Or so it would seem to me.

Stephen
 
The image on the screen is irrelevant- one doesnt even have to set the iso on your camera at anything other than 320. But set your meter to something higher, 800, 1000, what have you. This way you are underexposing and biasing your middle gray lower down. The only reason to set your iso on monitor/in camera is so you can see and be sure that the information isnt buried too deep in some problem area. Its not tricking your brain. There is no reason not to underexpose a stop or so on the mx - it is so clean you might as well underexpose and keep a little extra headroom incase the clouds get a little thinner and suddenly everything goes up a stop or what have you.
 
You must log in or register to reply here.
Back
Top