Olympus E-PM2: Small Camera, Big Surprise

dpreviewreader wrote:

cgarrard wrote:

dpreviewreader wrote:

cgarrard wrote:

dpreviewreader wrote:

The EPM2 is off by no more than 1/3 stop. DPR's test of EM5's ISO rating showed it was off by 1/3 stop and I don't expect the EPM2 to be any different as it uses the same TruePic VI processor. It is certainly not off by 1 stop.

A good argument to present to DXOMark I'd presume.

Instead of presenting anything to DxOMark I present the link below to you.

http://www.dpreview.com/forums/post/4093368

There have been countless threads on this topic. What DxOmark measures as ISO is not what camera manufacturers measure.

Carl
http://www.photographic-central.blogspot.com (Gear reviews)
http://www.carlgarrard.blogspot.com/ (Best work compilation)
Also formerly AlphaMountWorld.com (Now off the web)

I'm not defending DXOMark, nor applauding them.

Nobody said you are. You seem to have missed the point that what DxOMark calls ISO is not what the camera manufacturers call ISO hence you are comparing apples and oranges and coming to the wrong conclusion. EPM2's ISO rating is really what it claims to be for all practical purposes.

No I'm definitely not missing the point.

Since DXOMark uses similar testing methods for all cameras, the fact that the ISO value scores lower than the stated value, for this particular camera and brand (compared to other cameras and brands), should mean something however.

All I'm concerned with is that DXOMark have consistent testing methods for all cameras, which, they do. It gives me an indication of what is going on under the shell.

Nuff said on this topic though :).
Carl

http://www.photographic-central.blogspot.com (Gear reviews)
http://www.carlgarrard.blogspot.com/ (Best work compilation)
Also formerly AlphaMountWorld.com (Now off the web)

cgarrard wrote:

No I'm definitely not missing the point.

Since DXOMark uses similar testing methods for all cameras, the fact that the ISO value scores lower than the stated value, for this particular camera and brand (compared to other cameras and brands), should mean something however.

All I'm concerned with is that DXOMark have consistent testing methods for all cameras, which, they do. It gives me an indication of what is going on under the shell.

Nuff said on this topic though :).

Never enough said

The difference is most likely due to DxO testing say a middle grey exposure and the camera makers use thousands of real images taken in the field under all conditions and then set/declare their ISO to better protect highlights.

Never an issue, nobody is telling lies, just use the camera and adjust to personal taste.

Regards....... Guy

amalric wrote:

Bhima78 wrote:

Blowing up the image by 2x magnification, and interpolating the pixels to get a 16MP image. Not a great idea, but if you need the extra reach it could be worth it.

Shabby in your mind. That is what people do with m4/3 when selling to Stock agencies, and it works quite well.

No need for zooms, with their own limitations.

Even though I have an EP3 with it's lowly 12mp sensor (which works well in good light nonetheless) I find the 2X magnification pretty impressive on it. I've used/tried some fractal interpolation (or whatever it does) in Photoshop, and the Oly seems to do pretty well in comparison. Considering it doesn't have Gigabytes of memory to work with. Yeah sometimes at 100% some/a few artifacts are there, but for the most part people aren't looking at photos with a magnifying glass in the "real" world. It's great to me, to walk around with a small camera, a small prime lens...and yet have the capability to use double the focal length in a pinch. This photo was taken with it, at quite an extreme distance and think it attests to both the 2x zoom and the quality of the cheap 40-150 'R' lens. Regarding the lens, I really don't think there's any more "bang for the buck" in the world of cameras, although I shoot more with primes...

I wouldn't be afraid to print this "fairly" large, since with larger prints most people aren't viewing them from inches away, except on DPR.