DPR ISO Invariance: a6300 vs a6400

Started 2 months ago | Discussions thread
OP sankos Senior Member • Posts: 2,237
DPR ISO Invariance processing mistake for a6400
1

PankajDubey wrote:

sankos wrote:

It seems there's something wrong with the DPR ISO Invariance and Exposure Latitude image comparisons for the Sony a6400 camera. When comparing it with the a6300, the widget shows a dramatically worse performance of the newer camera. But when I download equivalent raw files and process them in the same raw converter (DxO PhotoLab 2.3 in this case), the difference is not there (when I equalize things like brightness and WB). Both cameras should perform at an equal level, judging by Bill Claff's shadow improvement comparison.

Did the DPR upload wrong jpeg previews for the a6400?

Good observation.

ISO 400

For the A6400 jpeg- there are two variations -

the white balance is Auto(Manual for A6300) and Brightness 2.53 ( A6300 it is 2.20) so they are not exactly comparable.

Yes, the perceived brightness difference is 1/3 stop between them -- I think this is the result of using DRO on the photos from the a6400; DRO is off on a6300 (as it should be in this test because it affects camera metering, and hence exposure).

Looking at the Exif data of the ISO3200 jpegs reveals the true cause of the difference, though. The ISO invariance jpeg rendering for a6300 (and other cameras) is based on raising the Blacks slider (+100) in ACR; but for some reason DPR changed the Shadows slider in a6400 files instead, that's why the shadows are lifted much more aggressively (flat global contrast) and the edge-detection which is built into the Shadows slider exacerbates the luma noise. It's clearly a processing mistake. The question is, how does one report it to DPR?

Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow