Jump to content

Well, it might interest some...


chris_tribble

Recommended Posts

Advertisement (gone after registration)

DxO have given us the chance to do serious pixel peeking on our favourite cameras (Welcome to dxomark.com (beta), a free resource dedicated to RAW-based camera image quality). Good to see that they include the M8 in the database - interesting to see how it comes out against the 5D (my normal comparator for this camera). It looks as if objectively the Leica doesn't do too badly against a camera that's been a bench-mark for low noise performance... Especially good IMHO given the contrast in pixel pitch between the two sensors.

 

Sure, there'll be plenty of cup-half-empty takes on this but I find it particularly encouraging when I compare the charts for signal to noise, tonal range and colour range..

 

Enjoy...

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!

Link to post
Share on other sites

  • Replies 132
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Posted Images

One thing I did not understand is that they give the charts for a 8-bit camera for the M8, when we know that it is a non-linear 14 to 0 bits camera after decompression, whereas the header as shown by Chris shows 14 bits again. That suggests that the data are based on in-camera jpeg conversion (not RAW based as they suggest). Not the best place to be for an M8, imo.

Link to post
Share on other sites

Thank you Chris, very very interesting indeed :D I compared the M8 with the D700 & the D3, which is quite an interesting comparison too, with the D3 being 1000 US less expensive (but way bigger) and the D700 being half as expensive as the M8, and not much bigger... :rolleyes: besides the obvious noise advantage for the last FF D-series, I found it interesting how they best the M8 on every parameter - including DR & color depth. Also, I found it interesting to see that the D700 has - very slightly - better high ISO performance than the D3 (supposed using the same sensor and electronics).

 

Thanks for the link, makes me happy about my decision to stay with film M and digital Nikons... :D

Link to post
Share on other sites

"DXO camera rankings" might be a better title, as I am surprised how few people have posted on this topic which is on every other site I have read today.

 

Yeah, a 2+ year old sensor on the M8 holds up pretty well. As usual, Leica is not a cost effective winner. Shows that a Canon G10 is the best of the small sensor cameras, but you do given up alot on this scale with the choice of a small sensor camera. I look forward to them doing this for medium format.

Link to post
Share on other sites

This sounds great news to me at first glance.

 

DxO have a fair degree of credibility, so I hope that they try to retain this and build on this commendable initiative making available open source data, presented in what ought to be an unbiased manner. Thus it should provide a far more informed choice to any potential camera change. Issues such as that you indicate Jaap need eliminating quickly.

 

How many 5D owners are currently wondering how much better the Mk II will be, and what this will bring to their pictures? The M8 update program is guaranteed to fare badly with their criteria!

 

The photographer on the street now has access to decent data before the photo-mag journalists add their spin, thereby ensuring that the camera manufacturers repeat their advertising within the publication.

 

In addition, I hope that this attenuates a lot of the noise/misinformed opinion proffered in web-forums. Having decided type of shooting the camera is needed for, the mix of properties can be determined, then the candidates can be seen plotted against each other.

 

I compared the M8 with the D700 and it did quite badly, but stood up very well to the D300 which was my previous DSLR. Of course, portability/volume/non-threatening are not parameters in the current DxO Mark assessment criteria.

 

As an aside, DxO optics v5.3 is a wonderful product which I have just bought for the Nikon. It is a real shame that they don't support the M8 and different lenses.

Link to post
Share on other sites

I have both the M8 and the Canon 5D. I find the biggest performance difference to be in the lenses. There is absolutely no comparison between Leica M lenses and Canon lenses - even the L prime lenses by Canon can't begin to match Leica M lenses. The only Canon lens that comes even close is the 85/1.2, newest version. I still prefer the Leica 75/1.4, though. The Canon 5D does much better when you put Leica R lenses on it with an adapter.

 

Tina

Tina Manley- powered by SmugMug

Link to post
Share on other sites

Advertisement (gone after registration)

I shoot with both the 5D and M8 but none of these comparisons ever explain why the M8 can't be beaten when it comes to B&W files? I'm not a techy but I can see the difference. The M8's ability for B&W photography is in my view the best that money can buy.

 

Cheers,

Wilfredo

Link to post
Share on other sites

I shoot with both the 5D and M8 but none of these comparisons ever explain why the M8 can't be beaten when it comes to B&W files? I'm not a techy but I can see the difference. The M8's ability for B&W photography is in my view the best that money can buy.

 

Cheers,

Wilfredo

 

 

I find the M8 much better for B&W, too, Wilfredo. I think the lack of the internal filter makes a big difference - that plus the much better lenses. How do you convert?

 

Tina

Tina Manley- powered by SmugMug

Link to post
Share on other sites

I have both the M8 and the Canon 5D. I find the biggest performance difference to be in the lenses. There is absolutely no comparison between Leica M lenses and Canon lenses - even the L prime lenses by Canon can't begin to match Leica M lenses. The only Canon lens that comes even close is the 85/1.2, newest version. I still prefer the Leica 75/1.4, though. The Canon 5D does much better when you put Leica R lenses on it with an adapter.

 

Tina

Tina Manley- powered by SmugMug

 

I'm also a Canon user, along with the M8, and I completely agree with this assessment. The Canon 50mm f/1.2 doesn't even match up with the Noctilux, in spite of the bad rap it often gets.

Link to post
Share on other sites

This is actually the comparison that I found most personally interesting. I am not surprised at all by the result.

 

106046292.jpg

 

I purchased the Canon EOS XSi (450D) earlier this year for its smallness, versatility, and compatibility with my Canon lenses. I did not expect much from it but have been tremendously impressed by its image quality, often equalling or surpassing the M8. It's really been a joy to use and it's certainly one of the best bargains in changeable lens cameras today. At 10% the price of an M8 (and 8.5% the price of an M8.2), it certainly represents a stark reality-check.

Link to post
Share on other sites

Definitely interesting. However, I see some interesting technical discussions over in dpreview (see for example the thread starting with 50D 40D D300 D90 RAW comparison: Canon EOS 50D - 10D Forum: Digital Photography Review) on how objective these tests really are.

- It is mentioned that the test assumes that RAW is unprocessed, but some cameras process RAW in-camera.

- It is also mentioned that even the high ISO images are from well-lit (studio) images, and do not measure results from real-life low-light situations.

- It is also mentioned that print results may differ from these measured results but there seems to be no ranking based on actual prints.

- Cameras have different blackpoints, which influences shadow performance

 

I am sorry that I cannot contribute to these discussions on a technical level, but it would be interesting if more knowledgeable people could contribute to this discussion here, and whether or how it would affect the measured performance of the M8.

 

It is always interesting to discuss how objective "objective" tests are...

Link to post
Share on other sites

I think Michael Reichman's comments here: DXOMark Review are worth considering.

 

What DXO is doing is very interesting and they clearly are trying to be very thorough in their methods. That said, there's no one perfect method for testing a camera and their results are, I think, best considered as one piece of a puzzle. I'm finding the new site quite interesting, though, and can see various aspects in which they're trying to be pragmatic and not just measuring specs.

 

Though cameras do process RAW files to some degree, the "RAW" output they provide is as pure as we can get from a given camera (ie. we can't access the data prior to the tweaking). I do think its very important that DXO is working from RAW files. Many other review sites tend to rely on JPEGs which, of course, is problematic for many serious photographers.

 

Various quick thoughts...

 

I test for ISO noise when the subject is lit by tungsten. That tends to stress the blue channel which, in turn, usually reveals chrominance noise differences among cameras. Also, low light photography is often done under tungsten. I'll have to look again to see what temperature light they're using for the tests.

 

Lens choice can affect observed dynamic range and so that's always a potential confounding variable.

 

When I compare cameras of different MP resolutions, I often include one section where the files from the higher res. camera are down-sampled to match the pixel dimensions of the lower res. camera. That tends to give a more "apples to apples" approach when one thinks about files used for printing.

 

DXO discusses doing something similar and standardizing on an 8 MP file but I'll need to look further to see exactly what they're doing there.

 

I note that they're not reporting camera resolution in these tests and that may well be because they're very aware of lens res. as a confounding variable in that type of testing. If so, kudos.

 

Cheers,

 

Sean

Link to post
Share on other sites

I have both the M8 and the Canon 5D. I find the biggest performance difference to be in the lenses. There is absolutely no comparison between Leica M lenses and Canon lenses - even the L prime lenses by Canon can't begin to match Leica M lenses.

"Absolutely no comparison" and "can't begin to match" are very strong statements. I'm not sure how you're judging the lenses, but if corner performance is a key factor in your comparison, then it's not an entirely fair comparison. The M8 never shows a lens's corner performance, whereas the 5D always does, and this is the worst part of almost any lens. To make a fair comparison, one would have to compare full frame vs. full frame, or 1.3 crop vs. 1.3 crop. Back when Photodo.com still tested lenses (in the film days), they found comparable performance between certain Leica M lenses and certain Canon L primes (such as the 35/1.4 and 85/1.2).

Link to post
Share on other sites

one interesting conclusion was in the background pages about signal to noise ratio over time where they concluded that in the past 5 years, if you normalize for resolution, the actual gains in noise were about 2/3rds of a stop average, with outliers like the D3/700.

 

Not so much really.

 

I think RAW processing has come further if you look at C1, RAW converter, Lr, ACR 5.x etc, recently I did a "test" quick a dirty between the 1dsMkIII at 1600iso converted in the current capture one, alongside the M8. I downsampled the canon to the Leica res which should give it an advantage. Then I made a print full size at 300ppi and compared, and you know what, the difference was negligible, imo.

 

The equalizer I believe is C1 in that case, the other part was that my canon 17-40L (35mm) at 5.6 was no match for my cron 28 at 5.6. You had to crank up the sharpness so much on the 1dsMKiii file that whatever low noise you started with became more apparent as you tried to regain sharpness compared to the already sharp Leica file. So the only advantage of the MKiii is if you need to go big. (guess I need to buy a prime...)

 

DXO is trying to be objective and simply present 'data' but out of context, that data is not too helpful I believe. Now if they can demonstrate that their RAW conversion software performs better than other converters on a given camera/file, then they might have something.

Link to post
Share on other sites

DXO is trying to be objective and simply present 'data' but out of context, that data is not too helpful I believe. Now if they can demonstrate that their RAW conversion software performs better than other converters on a given camera/file, then they might have something.

 

As I recall, the promos for DXO version 5 did demonstrate this.

 

Here is a comparison I made about a year ago with C1 version 3.X, DPP 3.2 and DXO 4.X

 

http://goldsteinphoto.com/Posts/conversion.jpg

 

I found that DXO did better with the angular lines around the windows.

 

I have C1 4.5, DPP 3.5, and ACDSee Pro 2.5. I'll be getting CS4. So if I upgrade to DXO 5.3, I'll compare them all. (If I find the time.)

 

I read everything on the DXO site about how they test and compare cameras and find it very interesting about how thorough they are.

Link to post
Share on other sites

I have both the M8 and the Canon 5D. I find the biggest performance difference to be in the lenses. There is absolutely no comparison between Leica M lenses and Canon lenses - even the L prime lenses by Canon can't begin to match Leica M lenses. The only Canon lens that comes even close is the 85/1.2, newest version. I still prefer the Leica 75/1.4, though. The Canon 5D does much better when you put Leica R lenses on it with an adapter.

 

Tina

Tina Manley- powered by SmugMug

 

Tina, my thoughts (and experience) exactly. I used R glass on a 1dsIII this summer and only then got the best of both worlds. Previously I had sold a 5D because I had not clocked the fact that R glass with adapter was the way to go. Canon glass is drab.

 

I just bought a P45+ back and Phase III body with Phase 80mm 2.8 and the glass is, as usual, the constraining factor. Mine is duff, and while awaiting replacement I am using a Mamiya equivalent loaner, which is highly competent but has no soul or character.

 

I want Leica glass on Mamiya bayonet, and it looks like I might get it!

Link to post
Share on other sites

"Absolutely no comparison" and "can't begin to match" are very strong statements. I'm not sure how you're judging the lenses, but if corner performance is a key factor in your comparison, then it's not an entirely fair comparison. The M8 never shows a lens's corner performance, whereas the 5D always does, and this is the worst part of almost any lens. To make a fair comparison, one would have to compare full frame vs. full frame, or 1.3 crop vs. 1.3 crop. Back when Photodo.com still tested lenses (in the film days), they found comparable performance between certain Leica M lenses and certain Canon L primes (such as the 35/1.4 and 85/1.2).

 

Well, I did say that the 85/1.2 is the best Canon lens. I don't have the 35/1.4. The 24/1.4 is a big disappointment. I'm comparing results. When I use the Leica R lenses on the 5D (that's full-frame to full-frame), there is "absolutely no comparison" and the Canon 24 "can't begin to match" the Leica R19. As far as the M8 vs the 5D, I'm still comparing results. My photos submitted to stock agencies using the M8 have never failed. The 5D photos failed often when I used the Canon lenses. Stock agencies examine the photos at 100% so I do, too. Corner to corner. I have stopped submitting anything from the Canons and only use the M8 now. That could be because rangefinders suit my style of shooting or it could be some subliminal influence because I love my M8s. Whatever. My kids are now using my 5D and 1DMII and I don't miss them.

 

Tina

Tina Manley- powered by SmugMug

Link to post
Share on other sites

The equalizer I believe is C1 in that case, the other part was that my canon 17-40L (35mm) at 5.6 was no match for my cron 28 at 5.6. You had to crank up the sharpness so much on the 1dsMKiii file that whatever low noise you started with became more apparent as you tried to regain sharpness compared to the already sharp Leica file. So the only advantage of the MKiii is if you need to go big. (guess I need to buy a prime...)

 

.

 

Robert, the only zoom I've used with happy results on the 1DSIII has been the 24-105 stopped down somewhat. But if you really want to see that the sensor and body can do, get a Leica R 50 Cron on it. It'll make you very happy.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...