Images posted so far

[ulrikft]

Advertisement (gone after registration)

Jamie, and arthury - let it lie. To a certain extent your are both right. Some browsers have some "calibration" - most just assume any image is sRGB and don't bother to read profiles, thus are relatively uncalibrated. "Calibrated" screens can vary depending on whether they are calibrated with Monaco or Eye-One, and which gamma the users chooses, and whether they are LCD or CRT. Calibration is good - but it is not universal so long as we have 20 different monitor factories and 10 different calibration tools, each claiming to be better - and thus different - from the others.

 

Here is the acid test - and the last test I'll do, since my M8s are now awaiting some lucky buyer on my dealer's used shelf.

 

M8.0 vs. M9, same lens, same position, identical processing from .dngs, ISO 160, 50 Summicron, f/5.6, jpeg compression 95% (very high quality) - 2 crops plus a reference full frame, simple Sharpen filter in PS.

 

Note: looks like there is a slight exposure variation - M8 shows a bit more in the shadows but the gray type in the little red and white label at right is also bleached a bit - just we don't get TOO bogged down in DR discussions. The M9, IMHO, meters about 1/3rd EV darker than the M8 (in my broad experience of having worked with 2 samples of each).

 

So, you did a smaller crop of the M9 file, and therefore gimping it's performance in a resolution test? Why do people do this over at this forum?

[Advertisement]

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

[plasticman]

So, you did a smaller crop of the M9 file, and therefore gimping it's performance in a resolution test? Why do people do this over at this forum?

 

For goodness sake, this has been discussed ad nauseam now - get over it.

It's a 100% crop of each sensor - yes it covers a smaller proportion of the total area of the M9 sensor, but believe it or not, the way that each sensor handles color, detail, microcontrast, dynamic range and so on at the pixel level MAY have an influence on the final printed image.

 

People are grown-up enough to see the crop factor for themselves - why do you feel the need to lecture everyone on it? I'm intrigued to see that this argument was never used when discussing detail rendition in comparisons between the M8 and the Canon or Nikon FF cameras, for instance.

 

PS: and incidentally your sentence is wrong - Adan did not do a "smaller crop" of the M9 sensor, the crops are exactly the same size. The crop is only smaller in these terms: it is a smaller proportion of the total sensor area.

[ulrikft]

For goodness sake, this has been discussed ad nauseam now - get over it.

It's a 100% crop of each sensor - yes it covers a smaller proportion of the total area of the M9 sensor, but believe it or not, the way that each sensor handles color, detail, microcontrast, dynamic range and so on at the pixel level MAY have an influence on the final printed image.

 

People are grown-up enough to see the crop factor for themselves - why do you feel the need to lecture everyone on it? I'm intrigued to see that this argument was never used when discussing detail rendition in comparisons between the M8 and the Canon or Nikon FF cameras, for instance.

 

PS: and incidentally your sentence is wrong - Adan did not do a "smaller crop" of the M9 sensor, the crops are exactly the same size. The crop is only smaller in these terms: it is a smaller proportion of the total sensor area.

 

That argument was, is and will be used everyone someone compares a crop sensor/lower MP sensor with a higher one, and gimps the performance of the higher one. The way you compare things, is to gimp the largest/highest resolving sensor. The best methodology is to upres the m8 image to 18MP and crop the same crop. That way you can actually see the performance on a similar scale. That way you do not gimp the larger/higher resolving sensor, the way you, consistantly do.

[bybrett]

Brett - Are these the prints that you had at Reddotcameras last week? (Boat on shingle, Highlands). They were great pictures to show off some of the M9's abilities - good colour , lots of fine detail, butter smooth out-of-focus areas. They certainly helped to convince me that I'd made the right decision to buy. Out of interest - where was the shot taken?

 

Stephen

 

Yes, many thanks it was a day shoot on the Isle of Barra. Nice Scottish light. Cheers

[plasticman]

That argument was, is and will be used everyone someone compares a crop sensor/lower MP sensor with a higher one, and gimps the performance of the higher one. The way you compare things, is to gimp the largest/highest resolving sensor. The best methodology is to upres the m8 image to 18MP and crop the same crop. That way you can actually see the performance on a similar scale. That way you do not gimp the larger/higher resolving sensor, the way you, consistantly do.

 

yawn.

[bybrett]

I expect they are, as they were the same prints that Brett had at Stephens in Manchester on Sunday. They looked superb. It's difficult at this stage to imagine wanting more from such a compact body such as the M9.

 

If the images that have been posted so far haven't looked good, then blame the photographer not the camera.

 

Ha! Yes, I'm, "on tour" Thanks for your comments.

[ulrikft]

Advertisement (gone after registration)

yawn.

 

It is good to see that you obviously realize who's right, and act accordingly

[plasticman]

It is good to see that you obviously realize who's right, and act accordingly

 

Read my first response above then.

The methodologies are testing different things - one is not necessarily more legitimate than the other as you appear to be so certain about. I can see that both tests are interesting but you are so certain that you have the one and only answer that you can't even see that this is also an interesting test for some.

Your oafish and abrasive answer was boring enough for me to respond in kind - I want to see example links from you of where I do what you accuse me of: " That way you do not gimp the larger/higher resolving sensor, the way you, consistantly do." (my italics - your incorrect spelling).

[ulrikft]

Read my first response above then.

The methodologies are testing different things - one is not necessarily more legitimate than the other as you appear to be so certain about. I can see that both tests are interesting but you are so certain that you have the one and only answer that you can't even see that this is also an interesting test for some.

Your oafish and abrasive answer was boring enough for me to respond in kind - I want to see example links from you of where I do what you accuse me of: " That way you do not gimp the larger/higher resolving sensor, the way you, consistantly do." (my italics - your incorrect spelling).

 

 

What do you actually test when you test a higher resolving sensor pixel against pixel with a lower resolving sensor? It is not final image quality, beacause then you would gimp the higher resolving sensor beacause it uses more pixels to create the final image. It is not noise characteristics or similar things, since as proven with the 1ds mk III vs. d700 debates on other forums, even though the 1dsmk3 is far worse when it comes to noise at a pixel vs. pixel level compared to the d700, it is actually just as good when one compare final prints. It is not dynamic range, beacause again, that would be more valid when using upressing and then*cropping. What is it you actually test? Per pixel sharpness? What does that translate to in the real world? What good is that comparison?

[Shootist]

So, you did a smaller crop of the M9 file, and therefore gimping it's performance in a resolution test? Why do people do this over at this forum?

 

No not at all. He cropped both image to the same number of pixels, or close to it (the M9 images is slightly larger). Since the M9 has more pixels and captures a wider and taller image then the M8 with any given focal length lens this is the only way to compare the true resolution of the sensors.

Now if you were to print both full size images at there native size @ 300PPI the M9 image would be bigger, need to be printed on a larger piece of paper or take up more space on the same size of paper, then the M8 image and have more imformation in it, wider and taller, but when you overlayed the 2 images everything would be the same.

 

He has not gimped anything.

[ulrikft]

No not at all. He cropped both image to the same number of pixels, or close to it (the M9 images is slightly larger). Since the M9 has more pixels and captures a wider and taller image then the M8 with any given focal length lens this is the only way to compare the true resolution of the sensors.

Now if you were to print both full size images at there native size @ 300PPI the M9 image would be bigger, need to be printed on a larger piece of paper or take up more space on the same size of paper, then the M8 image and have more imformation in it, wider and taller, but when you overlayed the 2 images everything would be the same.

 

He has not gimped anything.

 

... And is that the reality of things? not really. If you want to compare _sensor_ resolution, and not _pixel_ resolution, upressing is the only right way to do it.

[Shootist]

What do you actually test when you test a higher resolving sensor pixel against pixel with a lower resolving sensor? It is not final image quality, beacause then you would gimp the higher resolving sensor beacause it uses more pixels to create the final image. It is not noise characteristics or similar things, since as proven with the 1ds mk III vs. d700 debates on other forums, even though the 1dsmk3 is far worse when it comes to noise at a pixel vs. pixel level compared to the d700, it is actually just as good when one compare final prints. It is not dynamic range, beacause again, that would be more valid when using upressing and then*cropping. What is it you actually test? Per pixel sharpness? What does that translate to in the real world? What good is that comparison?

 

 

The M9 does not have a higher resolution sensor over the M8. It does have a bigger sensor in size, wider & taller, and the corresponding number of pixels for that bigger size but the actual pixel size and density is exactly the same.

Why can't some understand this.

 

Now if the M9 had a 24MP sensor at the same 24x36 sensor size it would have higher resolution then the M8. The pixels would need to be smaller and more closely packed.

 

All the M9 does is captures a larger image for each given focal length lens.

If you crop out a 3916 x 2634 pixel image from a M9 image you would end up with a image of the exact same size and resolution (number of pixels and same phisical demension size at the same PPI setting) image as a M8.

[Shootist]

... And is that the reality of things? not really. If you want to compare _sensor_ resolution, and not _pixel_ resolution, upressing is the only right way to do it.

 

 

What are you talking about. You can not upres a M8 image to match a M9 image because the M8 captures a smaller area.

 

Yes you can take a shot with the M9 then use the same lens on the M8 and step back to get the same subjects in the image and then yes the M9 will actually hold more information.

But that was not what Andy (adan) was testing.

[ulrikft]

What are you talking about. You can not upres a M8 image to match a M9 image because the M8 captures a smaller area.

 

Yes you can take a shot with the M9 then use the same lens on the M8 and step back to get the same subjects in the image and then yes the M9 will actually hold more information.

But that was not what Andy (adan) was testing.

 

 

What was he testing then? individual pixels? I just don't see what you are testing, towards the real world, with that.

[adli]

Get a life....

[Jamie Roberts]

So, Jamie,

 

Now, you are suspecting that the monitor I have been using to edit my M8 images for sale is the problem, then.

So, the JPG compression we're all seeing in this thread's images are due to monitors?

 

Why don't you start telling us what special monitors, profiles and extra steps are required so that M9 images will look normal or great?

 

And, Jamie, just so you know, I have no issues looking at all other images from Nikon, Canon and other M8 images ... except the M9 images I am seeing here.

In fact, if I may add, images from the other lower end cameras are more tolerant of the lousy monitor I am using. These observations are really getting interesting.

 

 

And, of course, your assumption was that I was silly enough to pay $5K for my M8 and $5K for my D3 so that I can edit my images in lousy uncalibrated monitors to yield successful images. It was, indeed, a pure delight to read all these assumptions you have made.

 

Hmm. Methinks someone is "protesting too much."

 

I didn't say or imply anything you've accused me of here. Frankly, I don't care what you've spent or how. It means zero as to the results you get. And you're the one wondering why the M9 (JPEGs as it turns out!!) files don't look very impressive on the Web.

 

And you said there was lots of good color management on the Web when the fact is (pace, Andy!) there isn't. So, even though something told me I probably shouldn't do this, I simply asked some questions which you've refused to answer.

 

Talk about making assumptions!! LOL!!

 

I haven't said a single word to you about anything you've done or seen, except that it doesn't match my experience so far.

 

But since you made the comment about how great color management is, then it's only fair and reasonable to ask you what you're using, isn't it?

 

So you're way off base here, and I resent the implication that I'm going to blame you for something right off the bat. Ok with me if you're fine with what you have and do... it's a huge YMMV.

 

But I will assume that if you don't know about JPEG compression then you aren't fully capable of getting the best out of your D3, no matter what you spent on it.

 

So you might want to back off a bit and learn something instead of taking swipes at people who might be trying to help you

[adan]

Ulrikft: I've tested the M9/M8 both ways (and taken flack both ways )

 

Personally, I agree with you, somewhat, in that I'm more concerned with the total resolution of the system, which is why I posted comparable fields of view FIRST on other threads: http://www.l-camera-forum.com/leica-forum/leica-m9-forum/98684-m9-versus-m8-1-tests.html

 

But some folks are scared that the M9 sensor is "softer" or has less dynamic range, or more noise, or more noise suppression, or some such - and thought using different lenses might skew the results. So given the second brief opportunity with both cameras in hand, I did this test as well.

 

IMHO the M9 gives up nothing to the M8 at the pixel level - and has more pixels. End of story.

[arthury]

Hmm. Methinks someone is "protesting too much."

 

I didn't say or imply anything you've accused me of here. Frankly, I don't care what you've spent or how. It means zero as to the results you get. And you're the one wondering why the M9 (JPEGs as it turns out!!) files don't look very impressive on the Web.

 

And you said there was lots of good color management on the Web when the fact is (pace, Andy!) there isn't. So, even though something told me I probably shouldn't do this, I simply asked some questions which you've refused to answer.

 

Talk about making assumptions!! LOL!!

 

I haven't said a single word to you about anything you've done or seen, except that it doesn't match my experience so far.

 

But since you made the comment about how great color management is, then it's only fair and reasonable to ask you what you're using, isn't it?

 

So you're way off base here, and I resent the implication that I'm going to blame you for something right off the bat. Ok with me if you're fine with what you have and do... it's a huge YMMV.

 

But I will assume that if you don't know about JPEG compression then you aren't fully capable of getting the best out of your D3, no matter what you spent on it.

 

So you might want to back off a bit and learn something instead of taking swipes at people who might be trying to help you

 

Jamie,

 

There's absolutely no problem with my monitor/profile.

End of story.

[Advertisement]

Hallo,

Look here: Leica M

[Jamie Roberts]

Jamie,

 

There's absolutely no problem with my monitor/profile.

End of story.

 

And yet you won't tell us 1) what it is or 2) how you've profiled it--or even how you're viewing them!

 

Why is this important?

 

Because, simply, if your monitor is correctly profiled for 1,8 gamma (like many are) and you're looking at someone's JPEG profiled for gamma 2.2 (as many are--one's a Windows and one's a Mac "standard"), and you're looking at them on a browser that doesn't adjust this, or doesn't do it properly, then

 

 

• you may see a brighter picture than was intended by the person who made the JPEG
  
• you may see artifacts that others can't see--even if they won't print--more easily
  

 

This is just one variable. Even worse variables come in on what LCD you're using (laptops are essentially unable to display luminance data correctly), which means, again, you're seeing things others aren't, or you're NOT seeing things others are.

 

As I said, if you're happy, then great. But this is the tip of the iceberg in estimating a camera's performance "over the Web."

 

Ignorance can be bliss. But sometimes it's just ignorance

[ulrikft]

Ulrikft: I've tested the M9/M8 both ways (and taken flack both ways )

 

Personally, I agree with you, somewhat, in that I'm more concerned with the total resolution of the system, which is why I posted comparable fields of view FIRST on other threads: http://www.l-camera-forum.com/leica-forum/leica-m9-forum/98684-m9-versus-m8-1-tests.html

 

But some folks are scared that the M9 sensor is "softer" or has less dynamic range, or more noise, or more noise suppression, or some such - and thought using different lenses might skew the results. So given the second brief opportunity with both cameras in hand, I did this test as well.

 

IMHO the M9 gives up nothing to the M8 at the pixel level - and has more pixels. End of story.

 

I absolutely agree. I actually managed to recover my CF-card with some nikon d700 files that i took with a 24mm vs. m9 with a 24mm.. the nikon is a bit off focus/softer at f/2.8 than the leica, but noise-wise, when i uprez the nikon files to match the leica files... _man_ is that m9 actually close to competitive on noise.