M240 vs M10 High ISO

[RayD28]

Advertisement (gone after registration)

I have an M240P and took some indoor shots over the holidays.  I shot ISO 3200 and noticed some noise (not bad, but it was there) and wonder if any M240 users have compared M240 high ISO shots to an M10 and noticed much practical difference.  I'm posting here versus the M10 section to get the opinions of those of us that continue to use the M240.  All comments welcome.  Thanks in advance.  

[pixelman]

My limited testing gives the M10 about .5-.75 stops advantage around iso 3200.  The best data I have seen is from Sean Reid of Reid Reviews. You should subscribe and browse a bit.  If you aren’t familiar with Reid’s site, you should know that is subscription based (no ads) and that if you try it you should, you need to read the instructions... if you use it on an iPad you need to hold the iPad in landscape mode. 

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[Jon Warwick]

I’ve also found Sean Reid’s reviews to be very useful. It’s subscription only. I found it very good value, given his tests are the closest thing I’ve found to show the differences across Leica lenses and cameras in what seems an objective, unemotional, test-based set-up. That is pretty useful, given forums like this one inevitably have people that have paid $ for a new camera, and may have quite a degree of (often involuntary) bias in their opinions!

 

[DandA]

Sometimes the differences seen between these two cameras say at ISO 3200 isn't just the noise, but also the possibility of banding and sometimes faint patterns on certain fabrics of surfaces contained within the images. Simply with a few limited test files between these two, I found in very general terms, more or less a 1 stop advantage for the M10 at ISO 3200 under low light conditions. Others with more extensive use of these two cameras can elaborate much more. As an aside Sean Reid's site is extremely informative and I identify with his testing methodology.

Dave

[adan]

One point, of course, is that DxO measured the M10 ISOs at about 1 stop less than rated. Factor that in, or not, as you choose.

I got the beginnings of shadow noise banding at ISO 2500 when I tried an M240. No leeway for recovering more shadow detail. With the M10, that happens at ISO 12500 (which DxO would call 6400). So a nominal 2.33 stops, but perhaps only a real 1.33 stops gain for the M10. I think I'm getting more than 0.5-0.75 stops, myself, but not 2 stops - based on practicalities (shutter speeds and/or apertures under equal light levels).

I don't pay much attention to absolute noise (Hey, I'm used to Tri-X pushed to 1600/3200!) so long as it is evenly sized and spaced (somewhat filmlike), but I quit where digital bands appear.

However if we are talking "equivalent to a scan from color slide film" I'd stick to 1600 or below with either.

The M10 seems to have a stronger default contrast curve written into .DNGs, (darker shadows, brighter highlights) which punches up the color saturation to be more like the CCD M8/9's. But may also help suppress shadow noise a bit - so long as you don't then try to rescue those shadows.

(Rescuing the shadows at ISO 200 is really important, however. It's the only way to really dig out the M10's quite amazing dynamic range, which is hidden by the default contrast applied).

So one's personal experience may vary with whether one habitually shoots higher ISOs in: contrasty light (night with patchy street/stage lights), flat light (cloudy days), full-spectrum daylight, or limited-spectrum artificial light.

FWIW - M10, ISO 5000, yellow (2300°K) indoor track-light bouncing off a gallery ceiling/wall (moderate contrast), no exp. comp. no sharpening, slight color (but not luminance) NR. Crop and full picture. (ignore the fuzzyness - old 21 Elmarit pre-ASPH, 1/90 sec. f/2.8)

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Edited December 28, 2018 by adan

[DandA]

Adan, in your posted photographic example above, you mentioned it was shot at ISO 5000.  I assume that figure ( 5000) was what the M10 ISo dial was set to but in DXO parlance/ terms., they would consider it ISO 2500?

Dave

Edited December 28, 2018 by DandA

[jaapv]

Advertisement (gone after registration)

Don't start that hare running again... ISO in the digital world is a very fluid concept. In the end, it is what the manufacturer defines that it is.

[RayD28]

11 hours ago, pixelman said:

My limited testing gives the M10 about .5-.75 stops advantage around iso 3200.  The best data I have seen is from Sean Reid of Reid Reviews. You should subscribe and browse a bit.  If you aren’t familiar with Reid’s site, you should know that is subscription based (no ads) and that if you try it you should, you need to read the instructions... if you use it on an iPad you need to hold the iPad in landscape mode. 

Forgive me if I get this wrong, but this sounds like the M10 at 3200 ISO is about the same as an M240 at 2400 ISO (assuming .50 stop).  I had a subscription to Reid's site so I might renew.  Thanks for the feedback. 

9 hours ago, Jon Warwick said:

.... That is pretty useful, given forums like this one inevitably have people that have paid $ for a new camera, and may have quite a degree of (often involuntary) bias in their opinions!

 

Agree with you Jon, which is why I posted to the M240 section.  

8 hours ago, DandA said:

Sometimes the differences seen between these two cameras say at ISO 3200 isn't just the noise, but also the possibility of banding and sometimes faint patterns on certain fabrics of surfaces contained within the images. Simply with a few limited test files between these two, I found in very general terms, more or less a 1 stop advantage for the M10 at ISO 3200 under low light conditions. Others with more extensive use of these two cameras can elaborate much more. As an aside Sean Reid's site is extremely informative and I identify with his testing methodology.

Dave

Good point about the banding.  

1 hour ago, adan said:

One point, of course, is that DxO measured the M10 ISOs at about 1 stop less than rated. Factor that in, or not, as you choose.

I got the beginnings of shadow noise banding at ISO 2500 when I tried an M240. No leeway for recovering more shadow detail. With the M10, that happens at ISO 12500 (which DxO would call 6400). So a nominal 2.33 stops, but perhaps only a real 1.33 stops gain for the M10. I think I'm getting more than 0.5-0.75 stops, myself, but not 2 stops - based on practicalities (shutter speeds and/or apertures under equal light levels).

I don't pay much attention to absolute noise (Hey, I'm used to Tri-X pushed to 1600/3200!) so long as it is evenly sized and spaced (somewhat filmlike), but I quit where digital bands appear.

However if we are talking "equivalent to a scan from color slide film" I'd stick to 1600 or below with either.

The M10 seems to have a stronger default contrast curve written into .DNGs, (darker shadows, brighter highlights) which punches up the color saturation to be more like the CCD M8/9's. But may also help suppress shadow noise a bit - so long as you don't then try to rescue those shadows.

(Rescuing the shadows at ISO 200 is really important, however. It's the only way to really dig out the M10's quite amazing dynamic range, which is hidden by the default contrast applied).

So one's personal experience may vary with whether one habitually shoots higher ISOs in: contrasty light (night with patchy street/stage lights), flat light (cloudy days), full-spectrum daylight, or limited-spectrum artificial light.

FWIW - M10, ISO 5000, yellow (2300°K) indoor track-light bouncing off a gallery ceiling/wall (moderate contrast), no exp. comp. no sharpening, slight color (but not luminance) NR. Crop and full picture. (ignore the fuzzyness - old 21 Elmarit pre-ASPH, 1/90 sec. f/2.8)

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

The noise in that photo is acceptable to me.  I don't pixel peep for noise, but I do for focus sharpness and that is what started me asking about all this.  Given what you describe, I might get by with a little less camera shake.  

1 hour ago, DandA said:

Adan, in your posted photographic example above, you mentioned it was shot at ISO 5000.  I assume that figure ( 5000) was what the M10 ISo dial was set to but in DXO parlance/ terms., they would consider it ISO 2500?

Dave

Good question.  I took it to mean ISO 5000 per camera setting but Adan can confirm.

1 hour ago, jaapv said:

Don't start that hare running again... ISO in the digital world is a very fluid concept. In the end, it is what the manufacturer defines that it is.

Agree Jaapv.  I'm just looking for practical experiences and observations from others.  My hands are not as steady as they once were and anything I can do to stay above 1/60 and still get decent shots is great.  I don't have a nearby dealer to do an actual comparison so practical advice from here is helpful.  

Edited December 28, 2018 by RayD28

[adan]

Yes, that picture was shot with the M10 set to "its own" ISO 5000. That's the camera setting that was recorded in EXIF. I won't chase the hare of what DxO would call it.

Note 1: As chosen by Auto-ISO, in this case. I don't generally use Auto-ISO, but in galleries/events with patchy light, and where I am shooting rapidly, and swapping lenses often (21-135, max. apertures f/1.4 to f/4), and want the shutter speed to track 4X the focal length, I've decided it's useful.

Note 2: The M10 ISO knob/dial does not include ISO 5000 as such - it has only "whole stop" engraved ISO steps: 100 - 200 - 400 - 800 - 1600 - 3200 - 6400. Plus "M" (for "menu ISO selection") and "A" (for Auto ISO selection by the camera)

5000 is only available via the menu (like a 240) or via Auto-ISO deciding it is the best to use.

[adan]

Just for the record, this is the M10 at its own (manual, not Auto) ISO 12500, but with -0.7 exposure comp (effective ISO 20000 - to get a bare-minimum hand-held 1/125th sec with a 135 Tele-Elmar at f/4 in the light available). Low-rent low-wattage stage "spotlight" with ambient fill, 2700°K, no luminance NR but about +10 color NR), no sharpening.

And my own "secret-sauce" high-ISO processing in Adobe Camera Raw (darkening the shadow slider, minimal increase with the exposure slider, using mostly highlights, whites and contrast to bring up the brighter areas with minimum-possible banding). There is ever-so-slight banding in the central gray background shadow - about as far as I'm willing to take the M10.

Being able to use a 135 f/4 in more situations was one driving force for me in moving from the M9 to the M10 - higher mag. rangefinder; cleaner, better high ISOs. It has achieved those goals.

Normally I process this picture to monochrome, but restored the color here for "example" purposes.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Edited December 28, 2018 by adan

[jaapv]

As a side remark, I found that Topaz Denoise is quite good at removing banding. I would add that crushing the shadows by setting the black point can be helpful, if the subject allows it is another question. Masking and blurring uniform and/or OOF parts (skies!) is another trick.

[RayD28]

1 hour ago, jaapv said:

As a side remark, I found that Topaz Denoise is quite good at removing banding. I would add that crushing the shadows by setting the black point can be helpful, if the subject allows it is another question. Masking and blurring uniform and/or OOF parts (skies!) is another trick.

+1 regarding Topaz.  

[mike3996]

On 12/28/2018 at 3:02 AM, adan said:

One point, of course, is that DxO measured the M10 ISOs at about 1 stop less than rated. Factor that in, or not, as you choose.

Not wanting to fuel the reported-vs-real-ISO flamewars but M240 doesn't go at the rated speed either. At reported base ISO of 200, M10 goes at 100 and M240 goes at 134... The comparisons between the M's should probably be done within 1/3-1/2 stop range instead of a full stop...

Edited December 31, 2018 by mike3996

[sblutter]

This is why I took a pass on M10.  Am hoping the eventual M11 is real upgrade.  The technology is there - and I want it!