Leica M8 , part 3 (E. Putt)

[stan_belyaev]

I just read the 3rd article on Leica M8 vs Canon 5D. Both cameras were tested

with 75mm lenses. Very disappointing....

[pico_digoliardi]

Putts: <i>The smaller size of the Epson image is due to the smaller sensor size. But the difference in performance is quite visible. Leica left and Epson right.</i><p>

However, on the page the smaller image is on the left. It's a typo. The Leica M8 image is the one on the right. <p>

Shame when a typo like that gets through. More coming.

[pico_digoliardi]

I'm just a little critical of Putts' making a statement at the top of the article about defering tests that consider IR until he gets a hot-filter, and then nonetheless makes comparisons of the B&W characteristics. IR effects B&W rendition considerably. B&W IS COLOR, FAPP. Profound IR lowers contrast, which effects accutance.

 

IMHO, of course. It seems to be a test make prematurely, or at leastg with certain ambivalence.

[vinay_patel]

I just read it too, and as usual I understood about 40% of what I <i>think</i> he was trying to say, and I fear that the answers to two big questions I have might be buried in that other 60% :-( </p>

 

The first question is: in the comparison of the two resolution charts, aside from him having reversed them vs the text, why would an image shot from the same distance with the same lens be more than twice as big from a 1.33-crop camera as a 1.5-crop camera? </p>

 

The second question is: is it me or in the pixel-peeping comparison of the girl's eye, the Leica image looks like total crap? He mentions artifacts in the eyebrow but I see it everywhere in the skin, and what looks like tapioca noise in her iris. It looks like an interpolated and over-sharpened file. Or is it my monitor playing tricks?

[vinay_patel]

OK, nevermind the first question I figured it out, it's because the RD1 has only 6MP vs 10MP for the M8, the image is 60% of each dimension. Duh!

[pico_digoliardi]

And why weren't each given proper white-balance?

[nels]

"It (the IR issue) has been discussed ad nauseam on many internet forums (by the way: the best information can be found in the non-leica user groups)..."

 

Think he's reading PNet?

[roger_smith4]

Do you have a link to his test?

[andrew robertson]

Those moire patterns the Leica generates are pretty cool. They remind me of the wooden

trim you often see on Thai Buddhist temples.

[jtdnyc]

Here's the link:

 

http://imx.nl/photosite/leica/m8_3/m8part3.html

[brodeub]

It's hard to conclude too much from this test. The Canon is mounted with a zoom vs. a prime for the Leica; white balance appears different; exposure appears to be a little different (Canon slightly underexposed); and no mention of apertures (other than "middle"). It's interesting that Mr. Puts claims 9 stops dynamic range for the M8 but it appears that highlights in the model's hair are blown out. It would be interesting to know which rendition the model preferred.

 

Brian.

[marc_bergman1]

This article is getting noticed all over the web. I noticed on Fred Miranda's site, Scott Goosman did some sharpening of the Canon image. He posted the results here.

<p>

<a href="http://www.pbase.com/image/70812328">Comparison Shots</a>

<p>

I think it points out one of the problems testing digital cameras. Post-processing plays such an important in the overall quality of the final image.

[jtdnyc]

At a glance, the sharpened Canon image seems to be dramatically more similar to the Leica image than was the unsharpened Canon image in the Puts article. On closer inspection, however, the sharpened Canon image seems to have slightly less actual detail and even more artifacts.

 

Erwin was right to stress that, with digital, the imaging chain has more links and it can be hard to tease out the relative performance of one link against another. I'm skeptical of web-based visual comparisons. As PopPhoto (or was it Modern?) used to say in the text accompanying their tests, if the words seem to contradict the printed images, believe the words, since much detail is lost in the magazine production process that would nonetheless be visible in acutal prints. Apparently the same is true on line.

[marc_bergman1]

"On closer inspection, however, the sharpened Canon image seems to have slightly less actual detail and even more artifacts."

 

If you click on the "Other sizes- Original" you can see the images in greater detail.

 

I find the skin and eyebrow details to be much more evident in the Canon image. It also seems to have better 3-D depth.

[graham john miles]

I would be curious to know if the M8 can can resolve more angels on the head of a pin than a Canon.

[nels]

"I would be curious to know if the M8 can can resolve more angels on the head of a pin than a Canon."

 

For Macro work, Canon would kick butt.

[simonpg]

Marc said: "I think it points out one of the problems testing digital cameras. Post-processing plays such an important in the overall quality of the final image."

 

I think it also points out the fundamental problem of digital capture - it's reliant upon post processing.

 

Mmmmmm, for US$5000.00 a camera that's images rely heavily upon post processing. My Zeiss Contessa 35 isn't; nor is my Olympus XAII; don't think my M7 is either. Glad my fridge is full of film.

 

Makes me think that its so sad digital-capture has distorted the concept of value for money so much! :)

[Brad_]

<I>I think it also points out the fundamental problem of digital capture - it's reliant upon

post processing.</I><P>

 

As is film capture - unless you're not too fussy and are OK with drugstore prints.

[Jochen_S]

Drugstore prints? - Brad, you're kidding! - They have at least Photoshop auto everything included in those machines by now. - To be fair towards digital, give some inexperienced half colorblind average shooter a enlarger.

[jeroen dommisse]

Now wait a minute - where are the brick wall shots?!

[hankg]

For exhibition prints film can require just as much if not more post processing as digital. If you had a look at a straight print from an Ansel Adams negative and the final exhibition print you'd think that you wre looking at 2 different images. A master printer has a significant imapact on how the image looks and thats all in post production.

[harry_zet]

hogwash, as usual

[jrjacobs]

It's a stupid comparison. When doing the test shots, he writes:

 

"To eliminate as many variables as possible, I used studio light, a tripod and the Apo-Summicron 2/75 at medium apertures. The Canon had the very good 24-105 at focal length 75 and also at medium apertures."

 

OK - so he is testing cameras for quality of sharpness and rendition, but then he uses a PRIME lens on one camera, and a ZOOM lens on the other.

 

Doesn't seem like a way to "eliminate as many variables as possible" to me - instead it makes for an uneven playing field. Zooms are great, but good primes always outdo them when you are making rigorous tests like this.