EP2 incredible sharp!

[Rens]

cameras also have weak AA filters.

Between my own experience with lots of SLR models, various online lens tests, and my experience with the GF1, I'd be willing to put it up against any SLR of a similar pixel count....

......As I said earlier on, if people on here want me to I'll gladly test the GF1 and D700 against each other with the same lens and post the results. A GH1 would be a better test since it can do 3:2 and add horizontal pixels, but I don't have one.

I didn't use the same lens so it's not worth posting results, but I did this test for myself with G1 and D700 before buying a GH1. Had there been a big IQ difference I wouldn't have bought one, but I found none at all - see earlier post.

--
Rens

 

[Fearless_Photog]

Fearless_Photog wrote:

I didn't use the same lens so it's not worth posting results, but I did this test for myself with G1 and D700 before buying a GH1. Had there been a big IQ difference I wouldn't have bought one, but I found none at all - see earlier post.

Same here, I was originally considering an LX3 for a walk around camera. The main reason I bought the GF1 instead was so that I could have a light setup and leave the SLRs at home, but not give up their level of image quality. If there was a significant difference the camera wouldn't be able to serve that purpose.
--
http://www.flickr.com/photos/mcullenphoto/

 

[Jonas B]

After having read through the post above I think some balance is needed. It's not like a 12MP FF camera is sharper by some magic automatic. In this case however a FF with a faster lens would have been good, read better, as the background is terrible and the bokeh isn't that good.
Jonas

(...)

The original photo is sharp and well exposed because the OP used the camera to its strengths. He chose a medium high shutter speed (1/160th) suitable for the focal length, moderate portrait focal length (38mm aka 76mm eq) and stopped down to f5.6 so he had some good depth of field (maybe cause that is what the kit lens was open to at that FL). This lessened the requirement on him to hold the camera overly steady, both for shutter blur and to maintain good front to back depth of field. As a result, he got some of the fullest capability of the camera to record a crisp, sharp image.

If as the previous poster had suggested he was using a faster lens (I don't care about the sensor size it is mostly immaterial) then you are going to lose the depth of field, gain nothing useful in shutter speed and probably not take nearly as "sharp" appearing photo. I am not saying that the point of focus won't be sharp, but you won't get the great depth of focus shown in this photo that makes it look so outstanding.

I understand your comments but some further discussion can't hurt.

OP's image was taken at f/5.6. That is f/11 in FF terms. As a result he got all what you mention above.

When I suggested a faster lens, a larger aperture opening, I did so cause in my opinion the image suffers from too much DOF combined with a not very suitable background. To that the bokeh is disturbing (while bokeh is subjective I think most would agree that the combination of distances, focal length and aperture in this case is less than stellar).

So, I guess we just disagree; you like the f/11-style DOF here and I think it is too much.

I see a huge number of photos posted here and on other forums where someone takes a fast lens, crank it wide open, often in low light where the shutter speed is about 1/4 the focal length, and then wonder why their shots are not sharp. The recent thread about the Canon 50mm 1.8 comes to mind.

That's very true.

Take a lesson from the OP, use the tools that your camera comes with that allow you to take great photos. Sure it is fun to play around with shallow depth of field and "bokeh" but moderately stopped down, reasonable shutter speeds will deliver great photos every time.

I don't know if you are addressing me there or the audience in general?

regards,

Jonas

 

[PeterLeyssens]

Hi,

Crutch of an amateur is maybe slightly over the top, but it does become quite gimmicky when a series of photos has nothing but extremely thin DOF. A bit like HDR when all the photos are too much HDR. The nice thing is that you can indulge in it for a while, then come back with one more tool in your baggage.

This photo is good because the background is quite recognisable. Most great portraits are in context.

Peter.

--
gallery at http://picasaweb.google.com/peterleyssens

 

[Jogger]

This is such a mean remark I must say, regrettably.

Well, there are a whole series of canned, ideal responses above, if that is what you are looking for. I dont like distracting backgrounds in my headshots... weird things happen, like trees popping out the side of people's ears, rooftops out of people's heads, etc.. that why they should be blurred out and why people on here are screaming for a FAST portrait... seriously, there is a person a day on here asking for a FAST portrait. Which this photo would have benefited from.

 

[knickerhawk]

Hi,

Crutch of an amateur is maybe slightly over the top, but it does become quite gimmicky when a series of photos has nothing but extremely thin DOF. A bit like HDR when all the photos are too much HDR. The nice thing is that you can indulge in it for a while, then come back with one more tool in your baggage.

Anything can be overdone, of course, but subject isolation through careful choice of focal length and aperture is one of the most important considerations for serious photography. There's a good reason why fast (and expensive) glass still sells in an age of 100,000 ISO cameras.

This photo is good because the background is quite recognisable. Most great portraits are in context.

It's a matter of taste, but I disagree that some mostly cropped-out, slightly unfocused plants, wall and window sill provide meaningful context. The image is too cropped for there to be any contextual value with a focused background. In this instance, a defocused background would have been more satisfying.

--
My photos: http://www.pbase.com/imageiseverything/root

 

[Jon Rty]

You do know you can't compare MTF numbers between different cameras? If anything, you'll have to compare the performance in relation to the maximum of said camera. By doing so we can see that the ZA 135mm almost reached the maximum of 2350 in the center as well as in the corners, while the 20mm F/1.7 reaches the maximum of 2750 in the middle, but only about 3/4 of that in the edges. Thus, both being 12mp sensors, the A700 with the ZA resolves more. That, I'd imagine, has more to do with the lens as the ZA is one astonishingly good lens, and shouldn't be compared to the pancake. I'm almost certain the m43 with the same lens would show similar performance.

On a different note, the bigger the pixel-pitch, the more forgiving a sensor is to the lens. That's why you do not need top of the line optics to get almost 12mp out of a 12mp full-frame camera. That's not true with a 24mp camera, which is very unforgiving.

NB: Why stop at a S2 when there are Phase One backs available? 65 megapixels or bust!

Here's another one for you. Two 12 megapixel cameras, one has a smaller sensor, one also has a lens with much smaller elements, but which has more resolving power? Is it getting through yet? Notice I also picked a very exceptionally good SLR lens, a lot of ones in the APS-C tests can barely break 2000 LW/PH

http://www.photozone.de/olympus--four-thirds-lens-tests/464-pana_20_17?start=1

http://www.photozone.de/sony-alpha-aps-c-lens-tests/381-zeiss_za_135_18?start=1
--
http://www.flickr.com/photos/mcullenphoto/

 

[Jonas B]

You do know you can't compare MTF numbers between different cameras? If anything, you'll have to compare the performance in relation to the maximum of said camera. By doing so we can see that the ZA 135mm almost reached the maximum of 2350 in the center as well as in the corners, while the 20mm F/1.7 reaches the maximum of 2750 in the middle, but only about 3/4 of that in the edges. (...)

One has to agree though that it isn't bad that a lens can achieve 2750 line widths when used with a sensor not having more but 1500 line widths in theory.

Jonas

 

[Jon Rty]

Well, it's LW/PH, which I think explains it.

You do know you can't compare MTF numbers between different cameras? If anything, you'll have to compare the performance in relation to the maximum of said camera. By doing so we can see that the ZA 135mm almost reached the maximum of 2350 in the center as well as in the corners, while the 20mm F/1.7 reaches the maximum of 2750 in the middle, but only about 3/4 of that in the edges. (...)

One has to agree though that it isn't bad that a lens can achieve 2750 line widths when used with a sensor not having more but 1500 line widths in theory.

Jonas

 

[MikeK5117]

I don't know if you are addressing me there or the audience in general?

regards,

Jonas

No, not you specifically, not having seen your photography I would never make that specific a comment. Just to the community at large. And again, not suggesting experimentation is not a good thing, that is how we get better and experienced with our equipment. Just that let the OP be an example, use what the camera gives you to work with and you will get great photos as well especially when the shot is important.

 

[Fearless_Photog]

You do know you can't compare MTF numbers between different cameras? If anything, you'll have to compare the performance in relation to the maximum of said camera. By doing so we can see that the ZA 135mm almost reached the maximum of 2350 in the center as well as in the corners, while the 20mm F/1.7 reaches the maximum of 2750 in the middle, but only about 3/4 of that in the edges. Thus, both being 12mp sensors, the A700 with the ZA resolves more. That, I'd imagine, has more to do with the lens as the ZA is one astonishingly good lens, and shouldn't be compared to the pancake. I'm almost certain the m43 with the same lens would show similar performance.

You can't compare lenses across systems because the sensors are different. Obviously if we're talking about the results generated by both lenses and sensors of a given system, you can compare them, because the final result is what's really relevant. I didn't post those tests to say one lens is better than the other. I posted them to counter the ridiculous claim that a larger sensor will always have more resolution.

Anyway, if you want a micro four thirds lens that's better across the whole frame, this one will do.

http://www.photozone.de/olympus--four-thirds-lens-tests/490-leica_45_28?start=1

--
http://www.flickr.com/photos/mcullenphoto/

 

[Jonas B]

Well, it's LW/PH, which I think explains it.

Of course, I confess to confusion. Thank you,

Jonas

 

[PeterLeyssens]

Hello knickerhawk,

There's a good reason why fast (and expensive) glass still sells in an age of 100,000 ISO cameras.

Sure. Because it's hard to design and appeals only to a limited audience. Oh, that's two reasons already. With your reason added, that's three.

This photo is good because the background is quite recognisable. Most great portraits are in context.

It's a matter of taste, but I disagree that some mostly cropped-out, slightly unfocused plants, wall and window sill provide meaningful context. The image is too cropped for there to be any contextual value with a focused background.

Placing those 2 sentences right after one another like I did, made it look like they are referring to the same subject. I didn't mean it that way: the second sentence is a general remark that refers to most great portraits that have come down to us from the past.

About this photo, you're right that the plants are very limited as far as a "meaningful" context would go. I think they're okay, but not brilliant. But I'm not sure that just blurring them out would dramatically improve the photo. It would be better to take your other suggestion as an approach: crop less and show more context. Or rearrange the composition completely and put the man in front of something else. That's not always possible, but I've had great results turning around the subject, even until I was slightly behind them, to make a photograph of their face with a much better background.

Peter.

--
gallery at http://picasaweb.google.com/peterleyssens

 

[Jon Rty]

That's what I said.

I'd also say that those tests aren't really proving your point, as according to them a A700 with the ZA 135mm resolves more than a GF1 with the 20mm F/1.7. On the other hand, I'd imagine it's got more to do with the lens, and if you were to use the ZA 135mm on the GF1, I'd say that the difference would be negligible.

As for the Leica, yes it's more even, but it also preforms worse in the center. I'd say the Olympus 50mm F/2 would preform more like the ZA 135mm.

You do know you can't compare MTF numbers between different cameras? If anything, you'll have to compare the performance in relation to the maximum of said camera. By doing so we can see that the ZA 135mm almost reached the maximum of 2350 in the center as well as in the corners, while the 20mm F/1.7 reaches the maximum of 2750 in the middle, but only about 3/4 of that in the edges. Thus, both being 12mp sensors, the A700 with the ZA resolves more. That, I'd imagine, has more to do with the lens as the ZA is one astonishingly good lens, and shouldn't be compared to the pancake. I'm almost certain the m43 with the same lens would show similar performance.

You can't compare lenses across systems because the sensors are different. Obviously if we're talking about the results generated by both lenses and sensors of a given system, you can compare them, because the final result is what's really relevant. I didn't post those tests to say one lens is better than the other. I posted them to counter the ridiculous claim that a larger sensor will always have more resolution.

Anyway, if you want a micro four thirds lens that's better across the whole frame, this one will do.

http://www.photozone.de/olympus--four-thirds-lens-tests/490-leica_45_28?start=1

--
http://www.flickr.com/photos/mcullenphoto/

 

[Fearless_Photog]

That's what I said.

I'd also say that those tests aren't really proving your point, as according to them a A700 with the ZA 135mm resolves more than a GF1 with the 20mm F/1.7. On the other hand, I'd imagine it's got more to do with the lens, and if you were to use the ZA 135mm on the GF1, I'd say that the difference would be negligible.

As for the Leica, yes it's more even, but it also preforms worse in the center. I'd say the Olympus 50mm F/2 would preform more like the ZA 135mm.

What part of that test leads you to believe that it performs worse in the center?
--
http://www.flickr.com/photos/mcullenphoto/

 

[Jon Rty]

The Leica peaks at a smidgen less 2400 in the center, while the 20mm F/1.7 peaks at 2750. To clarify, I meant worse in the center than the pancake.

That's what I said.

I'd also say that those tests aren't really proving your point, as according to them a A700 with the ZA 135mm resolves more than a GF1 with the 20mm F/1.7. On the other hand, I'd imagine it's got more to do with the lens, and if you were to use the ZA 135mm on the GF1, I'd say that the difference would be negligible.

As for the Leica, yes it's more even, but it also preforms worse in the center. I'd say the Olympus 50mm F/2 would preform more like the ZA 135mm.

What part of that test leads you to believe that it performs worse in the center?
--
http://www.flickr.com/photos/mcullenphoto/

 

[Fearless_Photog]

The Leica peaks at a smidgen less 2400 in the center, while the 20mm F/1.7 peaks at 2750. To clarify, I meant worse in the center than the pancake.

Oh ok, I thought you meant relative to the ZA 135, because it beats it pretty much across the board, I had to ask because you said you think the Oly 50mm f/2 would give a similar performance to the Zeiss, the 45 already does, aside from being slower. Speaking of slower, I still prefer the rendering of the 135 STF to the ZA 135. That's one lens I'd love to use on my GF1.
--
http://www.flickr.com/photos/mcullenphoto/

 

[Jon Rty]

I'm sorry, but your again comparing absolute MTF numbers across different cameras, something that isn't comparable. As a example the same ZA 135mm resolves over 3050 LW/PH across the frame on a A850:

http://www.photozone.de/sonyalphaff/500-zeiss_135_18ff?start=1

Thus, you can't compare lenses directly like that. You could compare the performance as percent of the maximum theoretical LW/PH resolved, but even that has it's own problems. For instance, the Panasonic resolves as good as 100% of the maximum, eg. 2750LW/PH, in the middle, and about 75% in the corners.

The Leica peaks at a smidgen less 2400 in the center, while the 20mm F/1.7 peaks at 2750. To clarify, I meant worse in the center than the pancake.

Oh ok, I thought you meant relative to the ZA 135, because it beats it pretty much across the board, I had to ask because you said you think the Oly 50mm f/2 would give a similar performance to the Zeiss, the 45 already does, aside from being slower. Speaking of slower, I still prefer the rendering of the 135 STF to the ZA 135. That's one lens I'd love to use on my GF1.
--
http://www.flickr.com/photos/mcullenphoto/

 

[GregF]

Not sure what that has to do with anything, and isn't entirely true in terms of resolution, at least for camera lenses. Why do some F/4 lenses have better resolution than f/2.8 lenses, for the same maker and focal range? I bet you couldn't tell the difference between most any 12MP camera, and 12MP full frame, in good light. The difference is in DOF control and high ISO.

Yes, Oly P2 is "good"
FF (full frame) is GREAT

A full frame camera with the same megapixel count as an EP2 such as the D700 wouldn't have any advantage over the EP2 when it comes to sharpness alone. In fact the weak AA filter combined with a lens like the 20mm f/1.7 actually gives the micro four thirds cameras an advantage in actual resolving power. Dynamic range is another story.
--
http://www.flickr.com/photos/mcullenphoto/

Ability to receive the light transmission from the lenses has an effect on resolution. This is why an equal design telescope with larger objective will always have more resolution. The Olympus models have great resolution for their spec and sensor size but in varying light there will be times they still can not match the resolution of the 12 megapixel Full Frame cameras.

It's not just about dynamic range and high ISO ability.

 

[IchiroCameraGuy]

Not sure what that has to do with anything, and isn't entirely true in terms of resolution, at least for camera lenses. Why do some F/4 lenses have better resolution than f/2.8 lenses, for the same maker and focal range? I bet you couldn't tell the difference between most any 12MP camera, and 12MP full frame, in good light. The difference is in DOF control and high ISO.

It's not an absolute statement to rule at all times. This is why I said "in varying light there will be times when..."

When the light levels are lower than a brightly lit studio things change. All criteria of image quality is factor in by the image being entirely made up of light and a larger sensor being superior at receiving light.

You could have a rather dim outdoor conditions or indoor conditions and shoot both on low ISO and see the results.

There are other factors but given a good performing current generation for both you will see the full frame model being better when it comes to color sensitivity, tonal range, dynamic range, sound to noise ratio as DXO mark report. It's not anything magic but simply the better efficiency at receiving the light transmission.

All depending on the light and other factors involved you might not ever care about the difference and hardly notice. What I said was not to talk down on micro 4/3 or any smaller sensor but to keep facts in line because a lot of new people are registered to learn or on-looking these forums.

In the work I do for NHL stock photography and awards events all this stuff matters a lot and it is why you won't see anybody using a smaller sensor camera. For family fun, vacation, memories of the kids, it's a different thing all together. I think it is where the mirrorless models shine and the portability is way more important than any image capability differences.

you can message me if it still doesn't make sense.

And I still like the OP's photograph very much as I said, I'm not talking down on his camera.