Steve Spencer Offline Upload & Sell: On
|
DavidBM wrote:
Been doing a bit of thinking about the 24-70.
Apart from being able to use my old Canon glass, this lens was one of the reasons I move back to full frame from M43; at last I could have a full frame camera I was willing to take on hiking trips.
well I have loved he A7r, but the 24-70 is not as good as I had hoped.
BUT
If you compare its the absolute performance of the lens+sensor combination at its worst with what you are getting out of the best m43 zooms (which are not that smaller) like the panny 12-35 and olympus 12-40 ON THIER RESPECTIVE SENSORS, the worst case performance is usually a bit better and the best case performance is very much better. To make this comparison you can use perceptual MP from DXO, or even the raw imatest numbers from photo zone (which you cannot compare between systems to compare lens quality by itself, but which you can roughly compare between systems to compare overall lens-sensor system quality - it's just hard to say how much is the lens and how much the sensor)
Now those lenses are in some sense better: they are getting much more of the potential performance out of their 16MP sensors. If we want to award well done you badges to lens designers that is maybe what matters.
But if we want to print images, what matters is total system performance. And I think I'm satisfied that the move was OK from that point of view. When hiking the 24-70/A7r is still giving me better performance (normalised to print size) than the m43 system; and when I'm not hiking and use primes I'm getting amazing quality.
...Show more →
There are a couple of problems with this analysis. First, photozone does not use the same version of ACR with all systems. So, in comparing between systems you can't say how much is the sensor, how much is the lens, and how much is the RAW converter. That makes comparisons between systems invalid. It might be the system creating the differences and it might be the RAW converter. One would only want to compare differences in the systems and not in the RAW converter.
Second even if there wasn't this problem at 24mm and 70mm in the corners the GX1 Oly 12-40 f/2.8 does a little better at least wide open (although the numbers are very close at 24mm but noticeably different at 70mm). These numbers should also be taken with a huge grain of salt. Only one lens and camera from each system is tested, at only one focal distance, with only three points across the frame tested. To say this is representative of the how the systems perform is quite a stretch. It is a very limited test. Also note that the sensors in the newest Olympus cameras are quite a bit better than the sensor in the GX1 (a huge difference in SNR--at base ISO it is a whole stop different--for example could well lead to higher resolution scores on the Oly cameras than on the GX1 that was tested). In fact, some evidence consistent with that reasoning is that the lens gets a 9MP rating at DXO Mark on the OM-D EM5 or EM-1 and only a 7MP rating on the GX1.
The same thing shows up when you look at the acutance profile measurements at DXO. The OMD and 12-40 at 12mm slightly outperforms that A7R and 24-70 at 24mm in the corners, although the A7R combo outperforms the OMD combo in the centre.
So of course with the much higher MP count the A7R is going to get more resolution in the centre than m4/3rds, but in the corners even with a huge MP disadvantage the m4/3rds camera is going to do a bit better at 24mm and 70mm where the Sony/Zeiss lens is weak. If it was me for hiking I would want a really small Olympus cameras which would make a much smaller kit and still have the pretty much the same IQ. For example if you went with the EPM-2 and a Panny 14 f/2.5, 20 f/1.7, and Oly 45 f/1.8, you would have a kit that would be a lot lighter (about a pound lighter), would take up a lot less volume and would come close to rivalling the A7r and Sony/Zeiss zoom.
|