I posted this on a thread at dpr today. I find the subject so interesting that I repeat it here:
I did a photo shoot yesterday. It was indoor in dim light, no tripod possible, and flash not really usable, since the images should show people in a natural atmosphere with natural lighting.
I used a Nikon D700, half of the time with an 80-200mm f/2.8 and half of the time with the 50mm f/1.8. ISO varied from 800 to 3200. This morning, after reading a thread at dpr about the E-5 review, I asked myself if I could have done the job with a Nikon D7000 or even an Olympus E-5. So I studied the high ISO samples at dpreview, comparing the Olympus, the D7000 and the D3s (The D700 doesn't come up in the comparisons).
The D3s is easily 2 stops better than the E-5, so even with the 35-100 f/2.0, I would be one stop noisier. The comparison with the D7000 is different, however. I would gain one stop because of the faster lens, and comparing ISO800 on the E-5 with ISO1600 from the D7000, the E-5 is better rather than worse, One can of course claim that the 35-200/2.0 is an expensive lens, but so is the Nikkor 70-200.
If we switch to primes, and let's use 50mm eqv. f/1.4 for comparison's sake (That would be Nikkor 50mm f/1.4 for the D3s/700, Nikkor 35mm f/1.4 for the D7000 and Panaleica 25mm f/1.4 for the E-5), the situation is more or less the same. The Olympus loses 1-2 stops on the sensor, but gains 1-3 stops because of IBIS. I would say that the E-5 combo is superior to the D7000 here, and much cheaper.
I don't own an Olympus camera at the moment, but quality of the lenses is so good that I'm tempted by the E-5. The excellent rendering of detail and the complete feature set of the camera makes it a very usable option for all kinds of photography, and absolutely not only for existing users and fanboys.
Unfortunately, the review at dpr fails to see the whole picture here.