D700 vs. M9, high iso.

I guess each person should test this on their own. My RFs give me two extra stops of acceptable shooting in low light (I go to 1/8 versus 1/30 on SLR), so my personal test would probably be comparing the D700 at 3200 to the M9 at 1600 or 800.

I am surprised at how slow the progression of sensor capabilities has been.
 
gnarayan said:
People presumably want to shoot an M9 with a 35mm so if it needs ferider's 1-2 stops of correction (I'd be stunned if it was that high - particularly since again you can do this comparison at the center of the image) and it applies it to the raw images (ughhh) then it is perfectly fair to compare it against a D700 at 35mm. Of course the OP was shooting a 50 ...
Click to expand...

Both should be real easy to test, Gautham.

- Put a wide on an M9 and have it auto-recognize it
- shoot a picture
- tell the camera via menu it's a 50 or even better 90 Summicron
- shoot the same picture and compare

I don't think anything will happen to the picture center. But I would expect 1-2 stops for a 35 and maybe 2-3 (?) for a 21. Would be cool if one of the first owners could try ....

Cheers,

Roland.
 
M9 retains more information in the image, and is of course sharper. The D700 records less information, which is then pushed even more with noise reduction. Go M9!
 
Hi Roland

Hi Roland

I agree the test below would be most interesting, same with examining colors for cyan drift corrections.

A couple of things from that video with Stephan reassured me about a few things in this thread -

The vignetting correction won't attempt to dial out all, leaving some in, just like many older lens naturally vignette.

Leica is aware (not so much as some of the reviewers) that the aperture guessing of the blue dot is not accurate at all, it's not TTL, it could be +/- several stops of actual aperture selected by say someone using filters or compensating for backlight, so over-corrections that exist in the M8.x today will be less likely.

and finally, that it allows you to just turn the lens detection off.

Also, the goal for the M9 is to do all corrections (JPG and raw) in-camera, unlike the 4/3rds and some P&S raw cameras that apply massive amounts of lens corrections and other post capture mods.

C1 (their competitor) has been provided an M9 for profiling and will still be offering a profile/developer in addition to the included Lightroom one, giving the M9 customers a choice of raw developer.


ferider said:
Both should be real easy to test, Gautham.

- Put a wide on an M9 and have it auto-recognize it
- shoot a picture
- tell the camera via menu it's a 50 or even better 90 Summicron
- shoot the same picture and compare

I don't think anything will happen to the picture center. But I would expect 1-2 stops for a 35 and maybe 2-3 (?) for a 21. Would be cool if one of the first owners could try ....

Cheers,

Roland.
Click to expand...
 
I watched the two load as large images and the Leica has more resolution and noise.

The Nikon is obviously flatter in contrast and has less rez but smother more uniform noise.

I prefer the M9, but the Nikon is competitive.
 
Follow up on my previous post:

At minute 29, in his recent interview by Reichmann & co (http://www.vimeo.com/6344220) Stephan Daniel is very clear about cyan shift and vignetting correction to be done in the M9 prior to DNG generation. Lens profiles are not shared with Lightroom.

Roland.
 
Which makes sense. Now Leica has something proprietary to favor the use of Leica lenses on the M9. Digital gave them something the M mount never could.
 
Pickett Wilson said:
Which makes sense. Now Leica has something proprietary to favor the use of Leica lenses on the M9. Digital gave them something the M mount never could.
Click to expand...

Hi Pickett,

I don't understand how this favours Leica's lenses.

You can still mount any M or LTM lens and take a picture. If there is a worry that the corrections are not appropriate for that lens in camera then I think you could still set the lens to something like a 50 cron asph that required little correction and then apply whatever processing you like in post with PTlens or something. Or even better if there is a real "no correction" option.

Indeed, it'd be interesting to see how strong the corrections are by using an effectively "no correction" option and to compare lenses from CV and Zeiss this way.

Cheers,
-Gautham
 
Does the M9 have a "no correction" option? That would be surprising. Any examples out there with VC lenses. I know about the problems with the 15, buy anything else?
 
Pickett Wilson said:
Does the M9 have a "no correction" option? That would be surprising. Any examples out there with VC lenses. I know about the problems with the 15, buy anything else?
Click to expand...

It has an option for no correction.

Also neither the M8 or M9 perform any correction on RAW images.
 
It does seem the M9 could be better than the M8 but clearly there is no M8 here.
It is not as good as a D700 as this goes up to 6400 with ease
This test is like handicapping an Olympic athelete and then just finishing close to them in a race.

Richard
 
Am I the only one who thinks the tests pictures are not ideal for the purpose? Almost all digital cameras have good noise rendering at high iso, provided there is even light in the frame and the picture is properly exposed. This is the case of the images tested here. I would be more interested in images with deep shadows or underexposed. That is also more like real shooting conditions.

Arturo
 
The reality is that the M9, pushed to iso3200, printed to 20x14 inches, looks just as good, even with more detail than the d700 at native iso3200. The d700 is the high iso classleader. That is.. great! And this in color at that, and without any noise reduction. I think I could safely underexpose the M9 with 1.5 or 2.5 stops, and push it to 6400 or 12800, without it looking miles behind the d700. I guess that will be my next test, since the sceptics are so very, very sceptical for some obscure reason.. :)
 
ulrikft said:
The reality is that the M9, pushed to iso3200, printed to 20x14 inches, looks just as good, even with more detail than the d700 at native iso3200. The d700 is the high iso classleader. That is.. great! And this in color at that, and without any noise reduction. I think I could safely underexpose the M9 with 1.5 or 2.5 stops, and push it to 6400 or 12800, without it looking miles behind the d700. I guess that will be my next test, since the sceptics are so very, very sceptical for some obscure reason.. :)
Click to expand...


Just curious - if you guys are serious, why are you comparing the Leica to the D700 and not the D3? The D3 (not D3x) is the real "iso classleader", costs $5000 or so, and is considered a "pro" camera, much more so than the D700 and D300.

I expect that the M9 at lower iso settings will provide better resolution than the D3, but that's not what the D3 was designed for - if you're concerned with resolution, we should be talking about the D3x.

The D700 is a very nice camera, and produces great pictures, and many people enjoy it very much, but it's no D3, nor does it cost as much - far from it!

(I've got a D3 and an M8, and while the M8 seems to capture more detail than the D3, the M8 doesn't even come close when it comes to high speed operation and high iso - which is perfectly acceptable, as each was designed for a different purpose...)
 
After looking at the image comparison and glancing at DP Review, I've noticed a few interesting differences:
1. There is only the slightest amount of purple on dark subjects with the M9, something that I think might explain why an earlier poster felt the D700 "looked miles more natural". It's not nearly as objectionable like the M8 but it does show up when compared to the D700. The good news is that the IR filters you had to buy for lenses with M8 use completely eliminates this minor problem.
2. The M9 image appears more saturated but I think it partly has something to do with the exposure being about 1/3 stop less exposed than the D700.
3. Less "grain" or noise with the D700 but better resolution and contrast M9. I assume this is partly due to the 18 MP with the M9 vs. 12 MP with the D700.

Overall, these image differences demonstrate that the M9 compares nicely to the D700 given it's superior resolution, smaller size and well known rf advantages over an slr WRT faster better performing normal and wide angle lenses and stealthy handling. The D700 excells in a more comprehensive supply of lens options, superior flash options, etc. If the M9 doesn't have any bugs like the M8, I think the M9 would make a great digital camera for commited rf enthusiasts and given that it's a Leica with a full frame sensor and 18 MP, I don't think 7K is a surprise, if anything I'm surprised it's that "cheap" for Leica full frame high MP digital camera. After all, look at what people were shelling out for an M7, something over 2X the price of it's Japanese competator (ZM). I own a D700 with several Nikkors and an MP with several M lenses. If I had the money and the past extreme enthusiasm for that type of shooting, I'd certainly consider adding the M9.
 
Last edited:
You must log in or register to reply here.
Back
Top