E-M5 or GH3? You tell me

texinwien wrote:

Since I've provided new images , I won't spend a whole lot of time answering your questions, but I'll hit them shortly.

This assessment still stands. I looked at 100% magnification at both files at all stages throughout processing in RT, and the E-M5 file displayed finer detail at every step, both in fine fuzz of the model's sweater, and in the leaves of the synthetic plants. Nothing has changed about that, in my view.

What does seem to have changed, slightly, are the color casts. they seem closer to one another now.

Fixed - I checked for a newer version of RT for my OS before preparing the first images, but somehow overlooked the one you mentioned. I've installed it and re-processed these.

Also, I have provided links (in both posts) to what should be the original JPEG files, hosted at PhotoBucket. Of course, the original RAW files are also available at imaging-resource.com for anyone who'd like to have a go at those

I used (both times) the default suggested by RT. The first time, that was 'amaze'. The second time, the newer version of RT suggested 'hphd'. I used the suggested default each time.

Dunno, but I didn't do any cropping on the new images, nor any resizing, nor even stripping EXIF data from the files, so they should be as RAW as possible, hopefully.

That is correct. Same order this second go-around, although the colors appear more even to me this time. Haven't measured them, however.

Possible - as I mentioned above, I watched closely this time around, and my impression remains the same. The E-M5 file shows, to my eyes, a visibly and significantly finer resolution. I can't say with any certainty that this is due to either camera, of course. It could be down to error by IR or something strange happening during processing, but I do see a difference in resolution.

I have linked to what should be the original files in both posts. If I could enable downloading of originals on a per-file basis in my DPReview gallery, I would, but I don't want to enable it on a global basis. I'll be happy to upload the originals elsewhere if photobucket is also doing any kind of optimization to the files. I'd like to make sure the original files are available for inspection.

My judgments weren't made on that basis, but I do understand your point, and agree with it. That's why I provided links to the original files in both posts.

The files were about 16MB, so they fit into my DPReview gallery.

Understood - am willing to spend some time (but not an eternity) improving the method of presentation in order to make the comparison as valid as I can, within reason.

Let me know what you think...