M8 and Canon 5D Comparison Pix and High Isos

[sean_reid]

Advertisement (gone after registration)

Sean,

They were all jpegs.

 

JC

 

They looked that way but I didn't see it mentioned on the site. Thanks.

 

Cheers,

 

Sean

[KJB]

I agree Sean that there is a great comparison test between the two cameras that was done recently that takes all of these variables into consideration and then some Looking forward to your next installment on the M8.

 

Regards,

Ken

If only there was already a comparison like that for these two cameras...hmmm <G>

 

I'm with you on 9 out of 10 and factor them into my own comparison tests. #6 is a matter for debate because using two different focal length lenses with the same subject distance introduces a (sometimes significant) difference in depth of field even at F/8. My own tests of the M8 and 5D (over a month ago) kept field of view roughly equivalent for the first set and focal length equivalent for the second set. I'm tending the favor the latter now, having worked with the former for years. The problem, that some people seem not to understand, is that unless one controls the kinds of variables you list, the results will be anecdotal but not necessarily a valid comparison. I'm emphatically in favor of "real world" evaluation but a comparison test has to make sure that the variable is the camera (or camera/lens combination) and not other factors. If one doesn't control the confounding variables (to the extent possible) the results are of limited usefulness. There's a degree of basic scientific method in comparison testing. Ideally, one gathers both kinds of info.: controlled tests and results from normal use.

 

By the same token, any test of dynamic range that doesn't consider and factor in lens contrast also has limited usefulness. It call one what to expect with a given camera and a given lens...and perhaps *only* with that given lens.

 

Sean

[xrogers]

According to Tina's LUG mail, the jpgs she posted are from C1 converted raw files.

 

I'm almost afraid to point it out to a community of grumpy old men (not hyperbole, just see the age thread plus any M8 thread to prove it), but after reading Tina's LUG mail, I did a similar quick-and-dirty test for myself. I put it on a web page as another take on the same sort of test, take a look if you care to.

 

http://homepage.mac.com/xrogers/page1/page1.html

 

I'm pleased with the M8's low light performance.

 

Thanks,

 

Clyde Rogers

[xrogers]

Oh, and Sean's test is the best out there. Anyone serious about the M8 will find his site a terrific bargain.

 

Clyde Rogers

[AlanG]

I can't say there was enough information provided to draw too many conclusions from Tina Manley's tests. For instance, we have no idea if both cameras were set to the same f stop and shutter speed. Maybe someone should find out.

 

I will put forward this talking point:

 

If those ISO 1250 comparison images were shot at the same f stop and shutter speed, and the lighing remained the same in the room, then the Canon is shooting at a higher ISO than the Leica. In other words either the Canon is more sensitive than 1250 when set at 1250, or the Leica is less sensitive than 1250 when set at 1250. Or they are both off. So what will the noise look like from the Leica when you turn up its ISO to get the same exposure that the Canon 5D produces at 1250?

 

The only real way to make a sensible noise comparison is to set both cameras at the same shutter speed and f stop. Shoot with one camera at a gven ISO and then shoot with the 2nd camera. Keep adjusting the ISO of the 2nd camera until the tone curves of each camera match. Then convert the files without any noise processing in the converters and compare the files.

 

Additionally we might want to try taking these same raw files and try to optimize each one with the best conversion programs, sharpening and noise reduction techniques we can employ. This will give what we might consider the optimal image from each camera taken at a given nearly identical exposure.

 

Now some people like testing things and some don't. Some want to know objective results and some don't. It has always been this way with photogrphy. When I was a kid I shot everything with Tri-X and processed in D-76. I really didn't care if another film was better or if Accufine or Diafine or Rodinol was a better way to go. I didn't like testing, was happy with my results, and just wanted to keep using what I was comfortable with.

 

When I started shooting professionally, I started doing a lot more testing.

[sean_reid]

Oh, and Sean's test is the best out there. Anyone serious about the M8 will find his site a terrific bargain.

 

Clyde Rogers

 

Hi Clyde,

 

Thanks. I'm in weird position sometimes. If I mention something that was part of an article I've written there's a contingent who will insist that the posts are all just PR for Reid Reviews, etc. Yet, since RR is (an increasingly large) part of what I do, a large part of what I know about the M8, etc. comes from my work on those articles.

 

So, for example, I read about these various M8/5D tests, internally note my concerns with some of them, and think to myself "but my own testing shows XYZ". So, then, do I post or not? If I post, there's going to be a contingent who think its just a commercial for RR. So, very often, I say nothing because it would be impossible for me to respond in any depth without dicussing an article. It's almost like a doctor trying to answer a question about knee surgery while avoiding any discussion of operations he's done.

 

Here come the trolls...

 

Cheers,

 

Sean

[rvaubel]

Advertisement (gone after registration)

According to Tina's LUG mail, the jpgs she posted are from C1 converted raw files.

 

I'm almost afraid to point it out to a community of grumpy old men (not hyperbole, just see the age thread plus any M8 thread to prove it), but after reading Tina's LUG mail, I did a similar quick-and-dirty test for myself. I put it on a web page as another take on the same sort of test, take a look if you care to.

 

Canon-Leica

 

I'm pleased with the M8's low light performance.

 

Thanks,

 

Clyde Rogers

 

Clyde

 

I took a look at your test and I too am impressed with the results you are getting. Once again, even your 2500 ISO shots cleaned up easily in Neat Image. Actually they were pretty noise free to begin with.

I've looked at a lot of high ISO shots and they all were much better than Tina's shots. It's hard to say what happened as her results are awful.

 

Rex

[sean_reid]

When I started shooting professionally, I started doing a lot more testing.

 

Very interesting comment. Sometimes, one does not have the luxury of *not* testing. Better to test first than to find out too late, during or after a shoot.

 

Cheers,

 

Sean

[KJB]

Shooting manual exposure in RAW (at equivalent ISO) to get the two exposures as close as possible on the camera's histogram and then doing fine adjustments in C1 EC to get the histograms as close as possible is in my opinion the best practical method. Compensating exposure by adjusting ISO in my opinion would defeat the purpose of this test that is comparing high ISO performance.

 

I can't say there was enough information provided to draw too many conclusions from Tina Manley's tests. For instance, we have no idea if both cameras were set to the same f stop and shutter speed. Maybe someone should find out.

 

I will put forward this talking point:

 

If those ISO 1250 comparison images were shot at the same f stop and shutter speed, and the lighing remained the same in the room, then the Canon is shooting at a higher ISO than the Leica. In other words either the Canon is more sensitive than 1250 when set at 1250, or the Leica is less sensitive than 1250 when set at 1250. Or they are both off. So what will the noise look like from the Leica when you turn up its ISO to get the same exposure that the Canon 5D produces at 1250?

 

The only real way to make a sensible noise comparison is to set both cameras at the same shutter speed and f stop. Shoot with one camera at a gven ISO and then shoot with the 2nd camera. Keep adjusting the ISO of the 2nd camera until the tone curves of each camera match. Then convert the files without any noise processing in the converters and compare the files.

 

Additionally we might want to try taking these same raw files and try to optimize each one with the best conversion programs, sharpening and noise reduction techniques we can employ. This will give what we might consider the optimal image from each camera taken at a given nearly identical exposure.

 

Now some people like testing things and some don't. Some want to know objective results and some don't. It has always been this way with photogrphy. When I was a kid I shot everything with Tri-X and processed in D-76. I really didn't care if another film was better or if Accufine or Diafine or Rodinol was a better way to go. I didn't like testing, was happy with my results, and just wanted to keep using what I was comfortable with.

 

When I started shooting professionally, I started doing a lot more testing.

[denniswong]

I can't say there was enough information provided to draw too many conclusions from Tina Manley's tests. For instance, we have no idea if both cameras were set to the same f stop and shutter speed. Maybe someone should find out.

 

In fact she did, and I am puzzled by the information she posted under the individual images.

 

These are the information she posted

ISO 1250

Canon f5.6 1/4s 24mm

Leica f1 1/8s not mentioned of focal length, I presume it was 50mm

ISO 1600

Canon f5.6 1/4s 24mm

ISO 2500

Leica f1 1/11s

ISO 3600

Canon f2 1/60s 24mm

 

Please forgive my ignorance but I have this question : the lighting condition in the images doesn't seem too much different but the Leica required greater exposures, does sensor size have a effect on physical aperture?

 

Dennis

[Jamie Roberts]

A little bit of objectivity please. We are being asked to ignore a series of comparison shots between two cameras and instead to accept the result of one shot with one of the cameras along with an unsubstaniated claim that the other camera "would do no better". Does that seem reasonable to you?

 

George Deliz

 

Hah!!

 

George, of course it seems reasonable! If you go back to my original post, look at the comparison. It's flawed. The Leica is underexposed, end of story. What's unreasonable is actually ignoring the evidence as posted!

 

The fact that I own and shoot the 5d professionally (and up till very recently, the 1ds2) also makes my suggestion reasonable, but I guess you don't count ethical credibility on the Internet

 

Edmund--we disagree again! The M8 with the Nocti or the 75 Lux certainly gives the 1ds2 with the 85 1.2L a run for its money. It really does. (The 1ds2 with the 80 Summilux 1.4 though--that clobbers the 85 1.2L, which is a very good lens, too).

[Jamie Roberts]

{snipped}

They're very different cameras but I wouldn't give up either one of them. The Leica does, however, have a very special kind of quality in its files that is unique. If I was forced to choose only one body I that could keep (of all the digital cameras I've owned) it would be the M8 but I can still appreciate the excellences of other cameras, esp. the 5D.

 

Sean, absolutely well said, as always. For longer stuff and very low light I shoot the 5d (with Leica glass--the 80 Lux / 100 APO macro Elmarit and the 180 Elmarit are all superb on the 5d). But the DMR is going to see less work with the M8 now... it's just that good up to 1250.

[Jamie Roberts]

{snipped}I'm pleased with the M8's low light performance.

 

Thank gosh someone exposed the cameras correctly at high ISOs

[stunsworth]

Please forgive my ignorance but I have this question : the lighting condition in the images doesn't seem too much different but the Leica required greater exposures, does sensor size have a effect on physical aperture?/quote]

 

Dennis the 5D is more sensitive than the ISO figures indicate according to what I've read. Set to ISO 1250 the true sensitivity is something like ISO 1600.

[reven]

"Originally Posted by guy_mancuso

Lurking . Here is the problem just try and convince a 5d owner that the Canon 1dsMKII is the better camera "

 

You would need to own both to see which is better.... For thge fact. Both I nice but for only one camera I would always go with the 1DsMk2,

 

As far as M8 vs 5D... come one plz why do we even have to talk about it ? Buy the camera or not.

[tom0511]

Thanks a lot!

 

Very easy to see:

 

The 5D is the queen of available light photography!

 

Sorry Leica.

 

Is noise at 3200 ISO at 100% screen viewing your main parameter for jedging the available light capabilities of a camera?

[Guest umb]

Sean,

They were all jpegs.

 

JC

 

How do you know? In the initial post it says all files were "opened with Capture 1".

 

BTW, if they were really JPEG's it could also be assumed that the 5D would have produced better results too shooting in RAW.

[Peter41951]

And for the bookcase shot above I wasn't even trying hard. One of the things I love about the M8 files is the are so clean, the noise cleans up really well.

 

Rex

I cleaned up that shot with Noise Ninja; then brushed off any unnecessary cleaning. I thought the results very good, though unfortunately the file was too large to post here.

[eronald]

Hah!!

 

The fact that I own and shoot the 5d professionally (and up till very recently, the 1ds2) also makes my suggestion reasonable, but I guess you don't count ethical credibility on the Internet

 

Edmund--we disagree again! The M8 with the Nocti or the 75 Lux certainly gives the 1ds2 with the 85 1.2L a run for its money. It really does. (The 1ds2 with the 80 Summilux 1.4 though--that clobbers the 85 1.2L, which is a very good lens, too).

 

Jamie,

 

I dunno, try taking the filters off the 85/1.2, stop it down to F8 and ***use the sharpening in DPP***, and then tell me again what you think of the 85/1.2 re. sharpness - I really don't think my 75 Summi and the M8 are in the same league as either my 1Ds2 or my 1Ds when the lens gets stopped down (ie. when you are actually looking at detail). The 5D has never impressed me for detail somehow, even though it's known to be the reigning price/weight/quality ratio champion. Although it's clear that the M8 beats it re weight/quality.

 

Edmund

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[tom0511]

Now this is starting to get really weird. I own Leica too, what EXACTLY does this say?

 

BTW, if you want to shoot in low light conditions, you need fast wide angel lenses which Leica doesn't have anymore, 2.8 is as fast as it gets with Leica-M8 lenses below 37mm [35mm Film equiv].

 

28/2.0 asph

vc 28/1.9

 

both 32mm equivalent