NEX-7 vs. M9 – Part Deux

[marcmccalmont]

Michael
Thanks for the comparison. I find 2 things interesting 1. your willingness to redo the comparison, my personality flaw (stubbornness) would probably have prevented me from not "digging my heals in" and 2. not how good the M9 is (we all know it is a great camera) but how good a relatively affordable NEX7 is. Oh and 3. how civil Lulu members tend to be.
Marc

[pindman]

THANK YOU MICHAEL!!!!!!!!

Paul

[AlanG]

Yes, that is very good. I suspect it answers most questions about if the entire image of the Nex 7 can compare with what people are used to getting from an M9.  And, judging from the two Nex 7 images - before and after downsampling to 18MP, perhaps there is not much difference between 18MP and 24MP APS size sensors using that lens.

The one thing I don't quite understand is his remark about not wanting to scale up an image for concern it would generate pixels. Fair enough if he doesn't want to do that, but... In the first test using a 50mm lens on both cameras, it seems to me that he must have scaled up the tight M9 crops from 7.8MP to 18MP equivalent or they would not have matched the crop from the 18MP downsized Nex 7 shot.  

An APS-C size crop from the M9 is only 7.8MP to start with, not 18MP.  So if it was shown at 100% it would be much smaller than the same crop from a Nex 7 at 100%  And if he had scaled down the Nex 7 image to 7.8MP it would have thrown away the extra detail from the higher density sensor and not made much sense.

[JohnBrew]

Thank you, Michael, for going to all that extra trouble to essentially show what you had already shown! I can't see anything for either camp to complain about. Now back to our regular programming...

[Anadrol]

Thank you very much Michael, thanks to you I'd rather buy a koala fur coat now, rather than a M9 !

I find the difference between the two cameras less striking now though.

[Ben Rubinstein]

To be honest the tiny difference seen between the two could possibly be attributed just to the difference in lenses, I'd say this test proves that the differences are not worth quibbling about in the real world. Very valuable. I do rather want that Nex!

[bjanes]

The one thing I don't quite understand is his remark about not wanting to scale up an image for concern it would generate pixels. Fair enough if he doesn't want to do that, but... In the first test using a 50mm lens on both cameras, it seems to me that he must have scaled up the tight M9 crops from 7.8MP to 18MP equivalent or they would not have matched the crop from the 18MP downsized Nex 7 shot.  

An APS-C size crop from the M9 is only 7.8MP to start with, not 18MP.  So if it was shown at 100% it would be much smaller than the same crop from a Nex 7 at 100%  And if he had scaled down the Nex 7 image to 7.8MP it would have thrown away the extra detail from the higher density sensor and not made much sense.

Alan,

Your comments mirror exactly those of mine, and Michael has really never addressed the criticism directly, but he did repeat the test with better methodology. He stated that the results were similar to his previous tests, but the images he shows indicate that the Leica comes out better when its full resolution is utilized.

Whenever one re-sizes an image, there is an inevitable loss of quality that can be compensated for to some extent with proper sharpening. However, re-sizing is an inevitable step in digital photography. One can view on screen at 100%, but when printing, resizing to the resolution of the printer must take place. Up-sizing "invents" pixels, but that is exactly what Jeff Schewe recommends when printing to Epson printers with a native resolution of 720 ppi and when the native resolution of the image is less than 720, so up-sizing must not be all bad. Improper downsizing can create aliasing as Bart has pointed out in various posts and Eric Kaffer has reiterated. At least, up-sizing does not result in aliasing.

Regards,

Bill

[JohnNewman]

THANK YOU MICHAEL!!!!!!!!

Paul

+1 from a new member.  I don't expect to get my Nex 7 until Jan/Feb but the rolling review (and now this addendum) shows me that if I put good glass on it in due course, I will quality images comparable (if not exactly identical) to an M9 which is an object of desire way outside of my pension.  She Who Must Be Obeyed said today she is quite happy for me to have the NEX but I would imagine she would draw the line at anything with one of those little red stickers on  Wishing to live a few more years, I wouldn't dare ask her anyway!

So, many thanks again for the reviews.

John

[B-Ark]

It was generous of Michael to revisit this, based upon the hew and cry from the great unwashed  ;-)

Its still not the comparison that I would like to see, but that's neither here nor there.
The very fact that 'NEX-7' and 'M9' appear in a single sentence that does not include the word 'sucks' tells me to start saving up for an NEX-7.

[BradSmith]

NEX-7 vs M9

Let's see.....
roughly equivalent size
orders of magnitude higher lens availability for NEX
real world equivalent image quality between them
NEX costs 17% of the Leica's price

Boy, that would be a tough choice to make!!!!

Brad

[EricV]

Very nice comparison, Michael.  You nailed it this time.  Thanks!

[AlanG]

Alan,

Whenever one re-sizes an image, there is an inevitable loss of quality that can be compensated for to some extent with proper sharpening. However, re-sizing is an inevitable step in digital photography. One can view on screen at 100%, but when printing, resizing to the resolution of the printer must take place. Up-sizing "invents" pixels, but that is exactly what Jeff Schewe recommends when printing to Epson printers with a native resolution of 720 ppi and when the native resolution of the image is less than 720, so up-sizing must not be all bad. Improper downsizing can create aliasing as Bart has pointed out in various posts and Eric Kaffer has reiterated. At least, up-sizing does not result in aliasing.

Regards,

Bill

Yes, in actual use images are almost always scaled up or down.

I scaled up his M9 crop to match the 100% Sony 24MP crop and the two images looked the same to me. Likewise the scaled down to 18MP Sony crop pretty much matches the M9 100% crop. Now these are using jpegs so that might limit it or maybe 18MP APS density is about the resolution limit of that lens. (24MP is not much greater resolution than 18MP anyway.)  I don't think you could get much closer in a general photographic test and if there are differences they probably would be lost in typical use.

[K.C.]

NEX-7 vs M9

Let's see.....
roughly equivalent size
orders of magnitude higher lens availability for NEX
real world equivalent image quality between them
NEX costs 17% of the Leica's price

Boy, that would be a tough choice to make!!!!

You're not allowing for brand loyalty that's well beyond reason.

I sold a large Leica system years ago and have never looked back. I do shoot with SONY and ZA glass though and love every minute of it. A NEX is a definite consideration. A Leica, never again.

[bradleygibson]

Michael,


Thank you for the review, and the revisit.  You've definitely piqued my interest in this combination, as well as that of many others, I'm sure.

The moire you have called out in the review is likely a result of the demosaicing algorithm used to process the raw files.  AA filters wouldn't take care of this because the problem isn't actually in the raw file but occurs when developing the raw file.

If you're curious, you could try running the files through other programs (RawDeveloper or DCRaw) to watch the problem get worse or better depending on the algorithm you select.

Do you or does anyone else know whether the color issues with Leica wide angles (eg. 21mm 'lux) and the NEX-7 can be fixed with the lens (/sensor) cast correction tools in C1?

[AlanG]

Do you or does anyone else know whether the color issues with Leica wide angles (eg. 21mm 'lux) and the NEX-7 can be fixed with the lens (/sensor) cast correction tools in C1?

C1 has a Lens Cast Correction system (LCC) This requires you to shoot a test image with a piece of white plastic over the lens and use the information from that image to offset the lens' color cast.  If you follow the link below and scroll down on the page you'll see details about it.  I think if you shoot at high ISOs or are trying to boost an underexposed shot, you might risk some additional noise in the corners.

http://help.phaseone.com/en/CO6/Editing-photos/Lens-Correction.aspx

[bradleygibson]

Thanks, Alan, C1's LCC feature is the one I am asking about.  Does anyone know if this will correct the issues observed with the Leica 21mm and the NEX-7?

[Wayne Fox]

The one thing I don't quite understand is his remark about not wanting to scale up an image for concern it would generate pixels. Fair enough if he doesn't want to do that, but... In the first test using a 50mm lens on both cameras, it seems to me that he must have scaled up the tight M9 crops from 7.8MP to 18MP equivalent or they would not have matched the crop from the 18MP downsized Nex 7 shot.  

I think a lot of todo was made of something that was probably nothing ... the images that were displayed in the first series from the M9 were probably just small crops from 100%.  While he "cropped" them to match field of view, I didn't see anywhere in the article that indicates he resampled them to anything, so the article images where just cropped further.  So the "cropping" really didn't have any affect on the image quality.

Probably the important point which he addressed very well in this article was using a different focal length lens to match FoV from the two cameras.  There's certainly some logic to this and that's what most do when trying to compare resolution of two different sensor sizes.

To me the resolution quality differences are a "quibble" (a word I learned from Michael) so the real discussion is about FF vs APC (and the associated DoF issues), price point, and of course usability.

We've had NEX 7's on order from day one. I won't be able to keep them in stock for a while, the waiting list is growing. I'm planning on selling my M9 and lenses, but will wait to try some of them on the NEX 7 first. May keep a couple that I really like, now that I can focus things quickly.

[AlanG]

... the images that were displayed in the first series from the M9 were probably just small crops from 100%.  While he "cropped" them to match field of view, I didn't see anywhere in the article that indicates he resampled them to anything, so the article images where just cropped further.  So the "cropping" really didn't have any affect on the image quality.

Yes, he did not mention that he re-sampled. But he must have re-sampled because 100% crops from the two cameras shot with the same lens would not match in size on the screen. (Just to be clear to anyone, I'm talking about the first test not Part Deux.)  Nothing personal, but when you think about it maybe you'll see it is pretty obvious.

[Wayne Fox]

Yes, he did not mention that he re-sampled. But he must have re-sampled because 100% crops from the two cameras shot with the same lens would not match in size on the screen. (Just to be clear to anyone, I'm talking about the first test not Part Deux.)  Nothing personal, but when you think about it maybe you'll see it is pretty obvious.

Nothing personal, but I think everyone else is "over thinking" it.

I did a screen grab of the full image from the Leica, and uprezzed that to the size of a Leica file ... @5200 pixels wide.  While very blurry, you can see that the size of the relevant information exactly matches the detail shots offered in the aricle (the second image is a screen grab of one of those).  The detail shots are at 100%, so the fact that he "cropped" the image doesn't seem to matter, because what he offers as detail shots are not resampled.

[ErikKaffehr]

Hi,

The second test contains two actual pixel crops. I have downloaded both and tried to uprezz the Leica image to NEX resolution. I do actually find the NEX contains more detail, but I also agree that the difference is about 1.0034 imperial quibbles. Sharpening matters a lot. Both images show significant color Moiré which is very similar.

If you can live with the crop factor, the NEX-7 looks like a good alternative to the M9, using Leica lenses.

Best regards
Erik

I think a lot of todo was made of something that was probably nothing ... the images that were displayed in the first series from the M9 were probably just small crops from 100%.  While he "cropped" them to match field of view, I didn't see anywhere in the article that indicates he resampled them to anything, so the article images where just cropped further.  So the "cropping" really didn't have any affect on the image quality.

Probably the important point which he addressed very well in this article was using a different focal length lens to match FoV from the two cameras.  There's certainly some logic to this and that's what most do when trying to compare resolution of two different sensor sizes.

To me the resolution quality differences are a "quibble" (a word I learned from Michael) so the real discussion is about FF vs APC (and the associated DoF issues), price point, and of course usability.

We've had NEX 7's on order from day one. I won't be able to keep them in stock for a while, the waiting list is growing. I'm planning on selling my M9 and lenses, but will wait to try some of them on the NEX 7 first. May keep a couple that I really like, now that I can focus things quickly.