Home
SearchSearch MemberlistMemberlist RegisterRegister ProfileProfile Log in to check your private messagesLog in to check your private messages Log inLog in

Sony JPEG engine does not get along with Adaptall glass...
View previous topic :: View next topic  


PostPosted: Thu Jun 28, 2018 2:53 am    Post subject: Sony JPEG engine does not get along with Adaptall glass... Reply with quote

I'm shooting a number of Adaptall 2 era lenses on my Sony a900, my normal capture mode is cRAW+JPEG, which is a good thing, since the JPEG engine in the a900 hates some of the Tamron lenses...

The 28-50, SP 300, SP 500 mirror and to a lesser extent the 35-70 3.5 all produce washed out *and* overexposed in camera JPEGs.

Interestingly (and rather frustratingly) the RAW files when viewed in RawTherapee with default or neutral profiles look like the JPEGs I'm used to getting with my Minolta AF lenses... However, setting the RT profile to portrait (which is what I shoot on the a900 JPEG) gives very similar results to the in camera JPEGs.

This is all a bit odd, especially since the 28 2.5, SP 35-80, 200 3.5 CF and 135 2.5 CF all produce excellent and correctly exposed in camera JPEGs...

Has anyone else encountered this issue?

I know the solution is simply to shoot RAW only, but I'm still really interested in figuring out why these lenses are causing such issues in what otherwise is an unshakable platform (though, my Super Tak 35 f3.5 also suffers from this issue, so Adaptall glass isn't the only victim).

Any experiences or insights are appreciated!


PostPosted: Thu Jun 28, 2018 3:24 am    Post subject: Reply with quote

It seems this may be caused by metering issue. Are you using a chipped adapter? Have you compare the shutter speed using the 28 2.5@F4 and 28-50@28 F4 at a certain ISO?


PostPosted: Thu Jun 28, 2018 12:39 pm    Post subject: Reply with quote

That's a good suggestion, though My guess is that the exposures will wind up being OK, since exposure compensated images (underexposed to get better looking JPEGs) look underexposed in RT... I'll do some tests to see if exposures are within. a stop of camera controlled lenses.

As far as the adapter goes, the Tamron lenses are shot on un-chipped adapters while my current M42 adapter is chipped, so if the max aperture of the lenses doesn't match the max aperture of the adapter you need to make certain that the camera stays set to max aperture for correct metering.


PostPosted: Thu Jun 28, 2018 3:57 pm    Post subject: Reply with quote

Interesting. I wonder if this same issue applies to the mirrorless Sonys? I shoot raw with my NEX 7 and the only thing I've encountered is the converted images (Raw to jpeg) are often darker than what are displayed in the raw converter software, so I'll have to lighten them up a bit in post. But I've never come across anything you're describing -- although I don't have my camera set to raw+jpg. Maybe if I did, I'd see similar?


PostPosted: Fri Jun 29, 2018 3:21 am    Post subject: Reply with quote

so, this evening I shot with the two worst offenders: 28-50 and SP 300, also brought the Minolta AF 24-50 f4 as a similar lens to the 28-50. The bottom line is that the Tamron lenses are making good RAW files, but the Sony in-camera JPEG engine takes strong issue with some older lenses and over corrects the JPEG files. The culprits are DRO and the "Portrait" creative profile, both of which tone map the image and in the process wash out the affected images by creating low contrast, low color gamut, low sharpness renditions of the image... again, the RAW files look pretty much identical to images shot with more modern AF glass, with correct exposure and good color, contrast and sharpness.

If I want in camera JPEGS that are attractive when using these lenses, I can set the creative style to "neutral" and turn DRO off, this yields very decent images, though the tone mapping function of DRO can be useful in cases where PP shadow recovery would otherwise be required if working from a RAW file.

So at least I now know a work-around and which settings are causing the problem, though the fact that some lenses are affected and others aren't is still a bit odd.