Jump to content

M9 ISO comparison


tashley

Recommended Posts

Advertisement (gone after registration)

C'mon! Who are you kidding?!? A cell phone? There's some noise there, but more than usable and a decided improvement over M8 files at that speed.

 

Matt, it seems to me to look at M8 pictures here. It happened to me to get better results with my M8, I was waiting for this camera a lot, but what I've seen so far is not convincing me to buy one.

Link to post
Share on other sites

  • Replies 125
  • Created
  • Last Reply

But then, Enrico, it took about a year for users to figure out how to get any decent high ISO files out of the M8, and the M9 is only a day old, hardly time to get beyond "out of the box" results. And if these are already visibly better than the best M8 files (and they are) I think the claim of 1 to 2 stops improvement is not an empty one.

Link to post
Share on other sites

But then, Enrico, it took about a year for users to figure out how to get any decent high ISO files out of the M8, and the M9 is only a day old, hardly time to get beyond "out of the box" results. And if these are already visibly better than the best M8 files (and they are) I think the claim of 1 to 2 stops improvement is not an empty one.

 

I agree and honestly, this is the main reason I am waiting on one for at least the immediate future. I think one stop is a given, two would be more along of the lines of what I hoped for.

 

The problem with the need to "Tame" noise is that a lot of us simply don't have time to "Massage" the files to get them cleaner, I hate the look of noise ninja - I don't like the idea of losing acuteness and edge transitions in rolling the proverbial dough too much.

 

I think Sean Reid is going to have the best take on the noise factor due to all the obvious reasons, IE, an actual professional who uses the gear with talent and quality in mind. I don't think much can be done in the way of noise control through firmware upgrades, but this camera is fresh off of the line, so anything is possible I suppose.

 

But this is what we expect, right? This is the day after the launch of what I believe is a killer new camera from Leica and it has a lot of ground to cover before it can be summed in up in one or two threads let alone a single post.

 

Off to Reno to get my new home...:-)

Link to post
Share on other sites

But then, Enrico, it took about a year for users to figure out how to get any decent high ISO files out of the M8, and the M9 is only a day old, hardly time to get beyond "out of the box" results. And if these are already visibly better than the best M8 files (and they are) I think the claim of 1 to 2 stops improvement is not an empty one.

 

Jaap - while the noise characteristics are visibly better than the M8 at higher ISOs, I think there are a few of us that are so far underwhelmed by other things we're seeing in the DNG files: color characteristics, overall sharpness, and residual IR sensitivity.

I'm not going to jump to any conclusions yet - I was just into a local dealer and agreed to bring in my M8 and Noctilux on Monday or Tuesday to make some direct comparisons for myself - but if the only criterion to judge the quality of a camera is how noisy it is at high ISO, then none of the Leica cameras come out on top quite frankly.

 

However, on the plus side:

1. that baby is full-frame - which I'd really like

2. it's the quietest M I ever tried (including the M7 - which has a nicer but louder 'thwock' to the shutter)

Link to post
Share on other sites

Mani, I reserve judgement until I have received and used mine for a bit, and although I am pleased at the thought of some, or hopefully, significant noise improvement, it is not more than one of the considerations for getting the new camera.

Link to post
Share on other sites

Advertisement (gone after registration)

Mani, I reserve judgement until I have received and used mine for a bit, and although I am pleased at the thought of some, or hopefully, significant noise improvement, it is not more than one of the considerations for getting the new camera.

 

Indeed - took me a long while to get the best out of my M8 (well, I probably haven't achieved that yet, actually) and I was even considering selling it before I switched to a better raw developer, but opening the DNGs that are online, I'm really bothered by the general look of the files alongside the M8 files.

But my tastes are not universal - I actually shoot the M8 at 320 and accentuate the grain because I do think it makes the files resemble Portra, so noise is pretty much a non-issue for me.

We'll see. I'm sure you're gonna enjoy the new camera.

Link to post
Share on other sites

Sure, for you and Jamie look at this one Toke posted - look at the top left area. Clear banding IMHO...

 

 

 

http://www.l-camera-forum.com/leica-forum/attachments/leica-m9-forum/161100d1252536136-i-have-m9-l9995255.jpg

 

Actually just looks like color noise to me... not typical high-ISO banding, though I know that's perhaps splitting hairs.

 

Something though, is messed with the RAW converters and color so far, which isn't surprising since none of them support the M9 yet (or haven't been tweaked).

 

But I do see what you mean.

Link to post
Share on other sites

wouldn't it be nice if some unassuming photographers currently using canon/nikon high end DSLR's bought M9's and, disgusted that they don't best their current cameras in high iso pot-plant tests, sold them off cheap... to us!

 

What will you give me for my D3 and lenses: 35 f2, 85 1.4, 135 f2.0....? :D

Link to post
Share on other sites

wouldn't it be nice if some unassuming photographers currently using canon/nikon high end DSLR's bought M9's and, disgusted that they don't best their current cameras in high iso pot-plant tests, sold them off cheap... to us!

 

The pot plant test is notoriously unreliable. Everybody knows that. Ashtrays are the only way to go for scientifically valid results.

Link to post
Share on other sites

Actually just looks like color noise to me... not typical high-ISO banding, though I know that's perhaps splitting hairs.

 

Something though, is messed with the RAW converters and color so far, which isn't surprising since none of them support the M9 yet (or haven't been tweaked).

 

But I do see what you mean.

 

Jamie perhaps you may be able to explain to me exactly what banding is and where it is shown on the photo of the guitarist? In what region do I look?I have looked and I cannot see any banding( I presume lines running through the photo?) I would like to get to the bottom of this phenomenon, yet another Leica lesson for me to learn:o

Link to post
Share on other sites

Jamie perhaps you may be able to explain to me exactly what banding is and where it is shown on the photo of the guitarist? In what region do I look?I have looked and I cannot see any banding( I presume lines running through the photo?) I would like to get to the bottom of this phenomenon, yet another Leica lesson for me to learn:o

 

Well, for me, banding due to sensor noise (and not RF interference or something weird like used to show up on the 5d with certain AF lenses) looks usually like a relatively wide monochromatic (gray) streak accentuated by color noise.

 

Because it has both a monochromatic character and a color character, it's notoriously hard to correct. It's easier to correct in BW, but it depends on the noise character too.

 

You can see a BW version here, where the sensor at ISO 2500 is just starting to band:

 

highiso - L9941797

 

See the horizontal gray variations there? (I suspect this is an effective ISO 5000 or more, actually).

 

Don't worry that this is typical; it isn't unless you've underexposed by a lot. Here are more typical ISO 2500 shots from the M9 (also taken by Jono) where I don't see any banding whatsoever (just scroll through his presentation to see more):

 

highiso - L9230175

 

It's difficult to see in "Toke's" post Tim linked to because it really only has the color component. In this case if you look really closely at the upper left-hand corner you will see yellow splotchy "bands" running into the frame. But yellow splotchiness is running through a lot of ACR M9 conversions right now, so I wouldn't worry about them right now.

 

I really do think that's just a RAW converter not getting the color response of the sensor correctly, and not managing color noise, which is an easy thing to do (once you have access to the camera, that is, and that's why color often sucks for so many digital cameras when they're first released--often no-one gets a production model until after the camera actually hits production!).

 

So we need to wait for 1) more firmware development and 2) better RAW converter development, which according to Scott Kirkpatrick, should happen starting tomorrow with a point release of C1 :)

Link to post
Share on other sites

Actually just looks like color noise to me... not typical high-ISO banding, though I know that's perhaps splitting hairs.

 

Something though, is messed with the RAW converters and color so far, which isn't surprising since none of them support the M9 yet (or haven't been tweaked).

 

But I do see what you mean.

 

I agree with you, Jamie; this is blue-channel noise. It's a big problem with M8 files in tungsten lighting even at ISO 640. My M8 also exhibits banding especially at 1250 and 2500, but the two phenomena are different.

Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...