fredmiranda.com

Moderated by: Fred Miranda

New fredmiranda.com Mobile Site

New Feature: SMS Notification alert

New Feature: Buy & Sell Watchlist

FM Forums | Leica & Alternative Gear | Join Upload & Sell

1 2 3 4

6 end

Archive 2011 · �350 vs �3500 50mm lens test

bluetsunami
Offline
• • •
Upload & Sell: Off

p.5 #1 · �350 vs �3500 50mm lens test

Looking at Steve Huff's site the only comparison I'm seeing is an M9 with a 50mm Summitar vs the NEX with its Zoom lens. I'd actually like to see a comparison between both with a single lens being shared between them (like a 50mm Summicron).

Mar 30, 2011 at 09:26 PM

sebboh
Offline
• • • • • •
Upload & Sell: Off

p.5 #2 · �350 vs �3500 50mm lens test

millsart wrote:
This has not been my experience in regards to the M9. Yes it holds true if we are talking about, lets say a very packed Canon crop body vs a FF body with lower pixel density, but we are talking about a CCD sensor with no AA filter and offset microlenses vs a CMOS sensor with an AA filter.

I think Steve Huff has a comparison with a M9 with 50 Cron against a NEX5, using the same lens, and while the NEX has the edge in pixel density, the per pixel resolving power of the M9 is superior.

When its...Show more →

you're making a few faulty assumptions here. because the NEX has higher pixel density, one would expect it to have lower per pixel sharpness. the relationship between pixel density and per pixel sharpness is an inverse one. given the same technology, which this obviously isn't, larger pixels are always better. the higher pixel density (smaller pixels) of the NEX makes it harder for a lens to outresolve the sensor than it would be on a camera with lower pixel density (e.g. the m9). again, the interesting comparison for lenses is how the lux compares to a lessor lens on the same sensor. i don't think anyone should be surprised by the lux performing better on the m9 than on the NEX.

Mar 30, 2011 at 09:31 PM

Jonas B
Offline
• • • •
Upload & Sell: Off

p.5 #3 · �350 vs �3500 50mm lens test

Maybe we can agree that we are discussing a confused mix of fruits due to me not expressing myself very good? As you understand I have no experience from any digital Leica. You are probably right.

If I limit my comparisons to one camera only, and in this case the NEX or a �4/3 camera, I dare to guess the better lens will show a better result than the relatively poor lens.

regards,

/Jonas

Mar 30, 2011 at 09:32 PM

douglasf13
Offline
• • • • •
Upload & Sell: Off

p.5 #4 · �350 vs �3500 50mm lens test

Frankly, Steve Huff has one of my favorite blogs, but I don't really consider him a trustworthy tester for technical differences. I'd like to see a thorough test between the M9, M8 and NEX-5 in regards to sharpness with tripods and the works. As far as the NEX goes, it has better per pixel sharpness than any dslr I've used, and I see moire all the time in my shots. If it has an aa filter, it is very weak.

Mar 30, 2011 at 10:06 PM

millsart
Offline
• • • • •
Upload & Sell: Off

p.5 #5 · �350 vs �3500 50mm lens test

Figures that I just sold my M9 and shipped it off yesterday now that we've got a debate today between it and the NEX which I wish I would of done some direct comparisons on (though I do not own a 50 Lux ASPH but do have some other high quality optics)

I'll see if I've got some shots from the M9 I can easily duplicate on the NEX using the same lenses at least. Its not going to be the exact same light or framing of course, but I think it will help make the results, which I've seen firsthand repeatedly in editing my files, a little more clear.

What I'm seeing in a nutshell though, is that I'm not getting close to the same amount of detail from NEX that I can get on the M9, with the same lenses. The NEX with its 14meg 1.5x sensor has more pixel density than the ff 18meg M9 sensor, but the overall resolving power on the M9 is clearly better.

Only other argument that could be made is that the NEX then needs an even better lens because its more demanding than the full frame M9 ?? That doesn't really make any sense to me because how much better can glass get than modern Zeiss and Leica lens ??

As such, I find that on the M9, I can tell more of a difference, between two different levels lenses than I can on the NEX

Case in point, on the M9, my Biogon C 35mm is clearly sharper than the VC 35mm f2.5 II lens, HOWEVER, in trying those both on the NEX as well, I couldn't see as big of differences and other than the corners being worse on the VC, the centers looked the same.

There go, I would argue better glass makes a bigger difference on the M9, and not just in terms of the corners.

Perhaps the M9, without its AA sensor is able to resolve more than the NEX sensor at its limit.

All comes back down to though that you will be able to clearly see the difference of a 50 Lux ASPH on the M9 over a lesser lens more than you will on the NEX, and not just in terms of the corners.

Diminishing returns are always in effect, even on the M9, but if you want to really see what your new $4500 50 Lux is capable of, you'll feel you got your moneys worth far more if you put it on a M9 than you will if you mount it on a NEX

Mar 30, 2011 at 10:06 PM

sebboh
Offline
• • • • • •
Upload & Sell: Off

p.5 #6 · �350 vs �3500 50mm lens test

millsart wrote:
All comes back down to though that you will be able to clearly see the difference of a 50 Lux ASPH on the M9 over a lesser lens more than you will on the NEX, and not just in terms of the corners.

^^ this is the part i have trouble believing (not the part about the m9 having better per pixel sharpness). it doesn't make sense from a physics perspective and it goes against all my experience using progressively higher density sensors. in the center of the frame, i've never had a lens look better on higher pixel density sensor than on a lower pixel density sensor. on an old CCD sensor all my lenses look good. on the NEX lots of them start showing subpar sharpness on the pixel level and CA and SA become more apparent. on 4/3 only my greatest lenses look sharp and any aberration is obvious.

Mar 30, 2011 at 10:28 PM

millsart
Offline
• • • • •
Upload & Sell: Off

p.5 #7 · �350 vs �3500 50mm lens test

Again though, you have to realize its not apples to apples. Have any of the cameras you've had experience with not have an AA sensor for example ?

Again, why on the M9 does the difference between my Zeiss Biogon and the CV Skopar 35mm II reveal more than when I put both those lens on the NEX ?

On the NEX no matter how much I sharpen the images, or what method I use to sharpen them, there seems to be a limit of how much fine detail I can render. Things like individual hairs etc.

On the M9 and even my M8 before it, I can pick out finer resolving power if I'm really pixel peeping.

I don't know if I can attribute that all to just the lack of the AA sensor, but there certainly seems to be a limit to what the NEX sensor can resolve. Its good, don't get me wrong, I'm not knocking it, but I'm just not able to see differences between a very good and an excellent lens past a certain point.

sebboh wrote:
^^ this is the part i have trouble believing (not the part about the m9 having better per pixel sharpness). it doesn't make sense from a physics perspective and it goes against all my experience using progressively higher density sensors. in the center of the frame, i've never had a lens look better on higher pixel density sensor than on a lower pixel density sensor. on an old CCD sensor all my lenses look good. on the NEX lots of them start showing subpar sharpness on the pixel level and CA and SA become more apparent. on 4/3 only my...Show more →

Mar 30, 2011 at 11:02 PM

sebboh
Offline
• • • • • •
Upload & Sell: Off

p.5 #8 · �350 vs �3500 50mm lens test

millsart wrote:
Again though, you have to realize its not apples to apples. Have any of the cameras you've had experience with not have an AA sensor for example ?

Again, why on the M9 does the difference between my Zeiss Biogon and the CV Skopar 35mm II reveal more than when I put both those lens on the NEX ?

On the NEX no matter how much I sharpen the images, or what method I use to sharpen them, there seems to be a limit of how much fine detail I can render. Things like individual hairs etc.

On the M9 and even...Show more →

i'm quite sure the level of fine detail you see at the pixel level is much higher with m9 due to the lack of AA filter and different size and type of photosites. let me explain what i'm talking about because it's not clear to me we're talking about the same thing. say you set up a shot with an m9 on a tripod and then remove the m9, replace it with a NEX, put the same lens on the NEX, focus it on the exact same spot and take a shot at the same settings. the shot with the NEX should look like a crop of the picture with the m9. lets say the lens was the zm 50/1.5 shot at f/1.5 and you were extremely lucky and got both shots focused on the same point on something near to the lens. the ghosting due to SA on that lens should be visible in both shots, but should take up more pixels on the NEX due to its higher pixel density (appearing larger on zooming to 100%). the same thing should be true of the loCA on any oof highlights in the shot or any lateral CA. are you saying that you don't think this would be the case or that you know it's not the case?

Mar 30, 2011 at 11:36 PM

millsart
Offline
• • • • •
Upload & Sell: Off

p.5 #9 · �350 vs �3500 50mm lens test

I never did a back to back on a tripod test, but in looking at samples of both, and using the same lens, the M9 image is delivering more detail than the NEX image, same lens and shooting position. I honestly never really tested or looked for CA or SA, as all the lens I've used seem pretty to handle those things pretty well, though I do tend to see more CA on the NEX in some cases, such as with the CV75 f2.5 Lots of CA with the NEX, much less with the M9, though either can produce it if you have the right subject.

Basically as an analogy for the NEX and glass I see it like this, lets say you've got a car. you put some 4 for $99 discount store tires on it and thats the baseline performance. You then move up to a name brand Goodyear tire, and it clearly handles better. You then move up to the higher end Goodyear and it handles better still. However, you then move up to the top of the line, Z-rated tire they make and your not seeing any more gains over the midlevel model. Why ? Because your car is pulling .8g on the skidpad which the midgrade tires can handle and even with the better tires, your overall suspension is the weakspot, not the tires.

On a high end sports car though, you can notice an advantage going up to those z-rated tires because that car can pull .95g's and those midgrade tires weren't able to deliver all the grip the car is capable of.

Its the same with the NEX, going from a 18-15, to a voigtlander, to a Zeiss all shows some noticeable gains, but going from a Zeiss to a 50 Lux isn't going to show anything more.

No matter what lens I put on my NEX, I'm only able to render so much fine detail in things like hair, leaves on a tree etc.

The M9 is able to go further and resolve those individual hairs, leaves etc, if you've got good glass

Its not that the NEX is bad, or that it doesn't render good detail, because it does, but its simply not a M9 which is unique in its resolving ability for fine detail, unlike any camera I've owned, including a higher mp 5D mkII.

Again, I'll work on seeing if I can find some files tomorrow and try to go reshoot some stuff with the NEX and the same lens and post both files, but what your going to see is that the NEX is just not going to ever match the M9, no matter how expensive of lens you put on it, short of getting a Kodak 18meg CCD into the NEX.

Mar 30, 2011 at 11:59 PM

Graham Mitchell
Offline
• • • •
Upload & Sell: Off

p.5 #10 · �350 vs �3500 50mm lens test

Thanks for posting this. It's a healthy reminder to those caught up in the megapixel race that megapixels mean nothing without good optics!

Mar 31, 2011 at 12:09 AM

sebboh
Offline
• • • • • •
Upload & Sell: Off

p.5 #11 · �350 vs �3500 50mm lens test

millsart wrote:
I never did a back to back on a tripod test, but in looking at samples of both, and using the same lens, the M9 image is delivering more detail than the NEX image, same lens and shooting position. I honestly never really tested or looked for CA or SA, as all the lens I've used seem pretty to handle those things pretty well, though I do tend to see more CA on the NEX in some cases, such as with the CV75 f2.5 Lots of CA with the NEX, much less with the M9, though either can produce it...Show more →

yes, we are clearly talking about different things. there is no doubt in my mind that the m9 will outresolve the NEX by a lot. it has more pixels, bigger pixels, and a higher per pixel sharpness. looking at 100% crops or enlargements on print the m9 will always be sharper (unless you put a horrid lens on it and stellar one on the NEX).

that is not what i'm talking about at all though. i'm talking about CA, SA, and wide open softness being more noticeable on the NEX. my contax g 45/2 is an extremely fine lens on the NEX but i certainly notice that it sharpens up from f2 to f/4, has SA wide open, and always has both laCA and loCA. if the 50 lux is as sharp at f/1.4 as the contax at f/2 that would make it interesting proposition for me because i certainly haven't found another f/1.4 lens that could come close to that. if the lux is as sharp at f/2 as the contax is at f/4 (that would be amazing), has less CA, or less SA i would certainly notice that on the NEX (and i believe they would be more obvious on the NEX despite all images looking sharper on the m9).

Mar 31, 2011 at 12:43 AM

douglasf13
Offline
• • • • •
Upload & Sell: Off

p.5 #12 · �350 vs �3500 50mm lens test

M9 should be sharper than NEX because of sensor size and pixel size, but I'm not convinced the M9 has higher per pixel sharpness...nor am I convinced the NEX has an AA filter. Yet another thread about the NEX aa filter, or lack thereof. Moire. link

Mar 31, 2011 at 02:06 AM

denoir
Offline
• • • •
Upload & Sell: Off

p.5 #13 · �350 vs �3500 50mm lens test

douglasf13 wrote:
M9 should be sharper than NEX because of sensor size and pixel size, but I'm not convinced the M9 has higher per pixel sharpness...nor am I convinced the NEX has an AA filter. Yet another thread about the NEX aa filter, or lack thereof. Moire. link

There have been tests done here in the forum before. I don't want to speculate about the NEX AA filter, but from the samples I've seen there seems to be no question that the M9 has better higher per pixel sharpness. The photosites are larger so anything else would have been odd indeed.

http://peltarion.eu/img/m9/zm18-16.jpg

100% crop:
http://peltarion.eu/img/m9/zm18-16_crop.jpg

I suspect though that the NEX may give you the third highest per pixel sharpness of the cameras available today (after the M8 & M9).

Of course, a proper side by side test is the only real way of knowing...

Mar 31, 2011 at 02:19 AM

sebboh
Offline
• • • • • •
Upload & Sell: Off

p.5 #14 · �350 vs �3500 50mm lens test

denoir wrote:

I suspect though that the NEX may give you the third highest per pixel sharpness of the cameras available today (after the M8 & M9).

Of course, a proper side by side test is the only real way of knowing...

yeah, i don't think there's any question in my mind that the m9 has higher per pixel sharpness from the 100% crops i've seen. i can't get anything that looks that good with the NEX even when i know the sensor not the lens is what is limiting my resolution.

Mar 31, 2011 at 02:43 AM

douglasf13
Offline
• • • • •
Upload & Sell: Off

p.5 #15 · �350 vs �3500 50mm lens test

Which raw converter are you using?

Mar 31, 2011 at 02:51 AM

millsart
Offline
• • • • •
Upload & Sell: Off

p.5 #16 · �350 vs �3500 50mm lens test

so I guess the question then becomes can a lens be "too good" for the NEX?

A VC nokton for example isn't going to look as good as a 50 lux on the M9, though it's got probably the second best center sharpness, but you can still see a difference if your really looking for it.

On the NEX though, would there really be any discernible difference one could see between them, or are we at the point where the sensor is the limiting factor?

If so, is there really an advantage still to using a lux asph if it's corner sharpness isn't seen and it's out resolving the sensor in the middle ?

It may be better corrected for CA and SA, but then again the NEX cam sometimes show sensor booming, which would again sort of defeat some advantages to the lux again.

As such, it comes back to if a $450 nokton is going to work just as well as a $4500 lux and what if any marked differences could be seen. On the M9 we know the difference is there and while a steep price, it could make sense, but on the NEX or m4/3.....

Mar 31, 2011 at 12:31 PM

sebboh
Offline
• • • • • •
Upload & Sell: Off

p.5 #17 · �350 vs �3500 50mm lens test

millsart wrote:
so I guess the question then becomes can a lens be "too good" for the NEX?

A VC nokton for example isn't going to look as good as a 50 lux on the M9, though it's got probably the second best center sharpness, but you can still see a difference if your really looking for it.

On the NEX though, would there really be any discernible difference one could see between them, or are we at the point where the sensor is the limiting factor?

yes, you can see the difference on the edges between the two lenses even on aps-c. here is a test of the two lenses (and many others) on an epson R-D1: http://www.luminous-landscape.com/reviews/lenses/fastlensreview.shtml
the epson has the same size sensor as the NEX, but with lower resolution and (it appears) lower per pixel sharpness. yet one can still see the superiority of the lux versus the nokton in the corner crops. interestingly the nokton appears sharper in the center, though that is most likely due to the vagaries of rangefinder focusing. so my answer is no we are not at the point where sensor is the limiting factor and we are getting further from that point as pixel densities increase.

millsart wrote:
If so, is there really an advantage still to using a lux asph if it's corner sharpness isn't seen and it's out resolving the sensor in the middle ?

It may be better corrected for CA and SA, but then again the NEX cam sometimes show sensor booming, which would again sort of defeat some advantages to the lux again.

As such, it comes back to if a $450 nokton is going to work just as well as a $4500 lux and what if any marked differences could be seen. On the M9 we know the difference is there and while a...Show more →

sensor blooming occurs on the m9 as well, though i imagine like CA it is smaller on the m9 from a per pixel perspective. this is just another reason to prefer a better corrected lens on the NEX. sensor blooming appears to be correlated to CA and is seen more often in voigtlander and zeiss m-mount lenses than modern leicas.

this is kind of a silly debate though because other differences are much more visible, e.g bokeh and contrast.

the question is how big are the differences and are you willing to pay a ridiculous sum for them. i don't really see why the price of the camera should play a role in that. if one is looking for the ultimate in image quality, then yes, a bigger improvement will be seen by getting a FF camera to replace the NEX then by getting a better lens for the NEX. however, both the m9 and NEX are unique cameras and there is no step up from the NEX in image quality that doesn't also get rid of many of the reasons for buying a NEX. i'm hopeful that this will change, and when it does, i'll agree that getting a summilux for a NEX doesn't make sense. if that doesn't happen in the next couple years though. . .

Mar 31, 2011 at 02:25 PM

millsart
Offline
• • • • •
Upload & Sell: Off

p.5 #18 · �350 vs �3500 50mm lens test

What I think we need to do is get Luka a NEX or convince him to buy one (which again is a great idea for any M owner because its so small and portable its the perfect travel backup body, and also a great tool if you want to shoot some video, or do something where you might want to be able to use liveview, 7fps etc. Just makes a ton of sense to me)

Then we would need him to perform 3 test

1. With the 50 lux on both bodies, take the same shot, so we can see how the M9 will produce better per pixel sharpness and render more detail than the NEX, with the same lens, even with the NEX having the pixel density advantage. I think this one is pretty clear even without doing it. M9 files just have amazing per pixel sharpness and clarity. Looking at a 100% crop looks like a 50% view of any other camera. The clarity and detail is amazing and I've never seen a 100% crop like that from a NEX.

2. Shoot a shot with his 50 sonnar, which is a pretty good lens stopped down to f8 on the M9 and then the same shot with the 50 lux so we can see how much better the Lux is on the M9

3. Then do that same test, with the Lux and Sonnar on the NEX, and let us see if the differences between those lenses is more apparent on the NEX, or on the M9.

We know already that we can see a pretty clear difference between them on the M9, the big question is how it will work on the NEX

Will we see a more dramatic difference due to the pixel density of the NEX ?

or

Given the lower per pixel resolving power of the NEX, will the difference between the Sonnar and the Lux appear more similar than different due to the NEX having reached a limit of resolving power of the sensor and not the lens.

I think it would be a pretty interesting comparison to see, though of course its largely academic because the cameras are otherwise quite different and a NEX simply is not a M9 regardless of if its IQ is pretty close. Just a different type of camera all together with its own unique strengths.

I certainly wish I still owned both as they compliment each other nicely and are both really fun cameras to shoot, but given some financial reasons, had to just sell the M9.

Mar 31, 2011 at 03:32 PM

sebboh
Offline
• • • • • •
Upload & Sell: Off

p.5 #19 · �350 vs �3500 50mm lens test

millsart wrote:
What I think we need to do is get Luka a NEX or convince him to buy one (which again is a great idea for any M owner because its so small and portable its the perfect travel backup body, and also a great tool if you want to shoot some video, or do something where you might want to be able to use liveview, 7fps etc. Just makes a ton of sense to me)

Then we would need him to perform 3 test

1. With the 50 lux on both bodies, take the same shot, so we can see how the...Show more →

sounds like a good plan to me. i would like to see some wide open comparisons as well, since wide open performance is much more important to me. i suggested as much to luka earlier in the thread just so he could show a direct comparison of lenses (at least in the center) without the confounding factor of different sensors. what do you say luka? i think you can pick up a NEX-3 for roughly the cost of extra leica lens cap or filter these days.

Mar 31, 2011 at 03:42 PM

Jonas B
Offline
• • • •
Upload & Sell: Off

p.5 #20 · �350 vs �3500 50mm lens test

Millsart, I think your experience is interesting. To me it is counter intuitive; I would expect the sharper Summilux perform significantly better on the NEX than say an EF50/1.4 would do (on the same camera).

Like sebboh I would be interested in seeing a NEX: Lux 50 f/1.4 and f/2 compared to a Cron 50/2 and f/2.8, or a Summarit. If the 50Lux is better than the Cron at the corresponding f-stops I would expect some difference to be visible. I haven't used a NEX camera but with the experience from all my other digital cameras it feels safe to say a better lens will give better result.

Why do I write this? I simply got puzzled by your statements earlier. I'm trying to get over it...

And, I'm still interested in a super 50 f/1.4 lens.

Mar 31, 2011 at 03:56 PM

1 2 3 4 5 6 end
FM Forums \| Leica & Alternative Gear \| Join Upload & Sell 1 2 3 4 5 6 end