I have a Megadap adapter to be able to use Sony FE lenses but PhotoLab7 seems to get confused by this so for the Sony 85mm lens on the Z-mount body, it keeps wanting to import modules for a number of F-mount/Z-mount 85mm lenses instead.
I’ve had a search through the forums as I thought this must have come up before but I’m not finding much so I’d appreciate any advice from other users with Sony lenses on Nikon bodies.
Even though the lens may have a profile when attached to a body from the same manufacturer, PhotoLab doesn’t recognize lenses when they are attached to a different manufacturer’s body via an adapter. There are no profiles for the huge number of possible combinations.
I realise there’s no profile for it which is fine, the issue is it keeps prompting me to install the wrong profiles. I’d rather it just ignores these lenses since there’s nothing it can do anyway.
To enjoy the benefits of lens sharpening feature (which is the best sharpening in PhotoLab or anywhere else) requires a lens. If any of the non-FE lenses are similar in kind/distortion, it may be worth installing the profile just to have the full set of PhotoLab features.
Of course, if there’s no near match, having obviously wrong distortion numbers in there would only create problems.
There’s the secondary benefit of not being prompted to install a matching lens any more.
To enjoy the benefits of lens sharpening feature (which is the best sharpening in PhotoLab or anywhere else) requires a lens. If any of the non-FE lenses are similar in kind/distortion, it may be worth installing the profile just to have the full set of PhotoLab features.
Of course, if there’s no near match, having obviously wrong distortion numbers in there would only create problems.
There’s the secondary benefit of not being prompted to install a matching lens any more.
I can’t see that there’s going to be a lens similar enough that it wouldn’t cause issues but also it’s only wanting to import the modules, it’s not offering the chance to apply those modules to any of the images.
The adapter sets the lens make as ETZ21 (the model of the adapter) and the lens model as FE, this is the exif for an FE lens on a Nikon camera and a native lens:
Lens Info - 85.00 85.00 1.80 1.80
Lens Make - ETZ21
Lens Model - FE 85mm F1.8
Lens Info - 14.00 30.00 4.00 4.00
Lens Make - NIKON
Lens Model - NIKKOR Z 14-30mm f/4 S
The lens sharpening feature is a sharpening technique and not as closely attached to the specific glass as you imagine. Distortion is lens specific, but lens sharpening less so. Of course, extremely divergent glass would not work well. I expect the issue for lens sharpening is the resolution of the glass. Try to find a lens which is a close match in resolution (modern very sharp glass, moderately sharp glass from the 1990’s forward, or older lenses which are just not as sharp, with exception for certain Leica lenses).
Stenis
(Sten-Åke Sändh (Sony, Win 11, PL 6, CO 16, PM Plus 6, XnView))
9
@uncoy
From what I have read Lens Correction adds sharpening according to the optical properties of the lenses and since these can vary a lot especially at the edges I guess that can be a problem.
These properties of Lens Corretion, are they somehow tied to the camerabody codes in the image files too? I mean can that explain why a Nikon body is not “compatible” with E-mount lenses so you can see these Sony lenses in the lens !-box when we get lens warnings there. or don´t “see” them as an option when using them with an adapter on Nikon bodies? At least we can see even Sigma-lenses for E-mount in parallell with Sony E-lenses sometimes. Is that the case because they are all made for E-mount in that case?
So if they also use that E-mount-code it might be an issue in Photolab that prevents using a body that expect info compatible with Nikon Z-lenses. Isn´t Photolab a little picky and unfamous about vendor codes in the files for the camera bodies? So why shouldn´t it bee with the lenses since in that case it is far more important with the right lens codes than the ones for tha camera bodies. I fully accept these lens code limitations but I really have hard to understand why they lock out images without a supported camera body code in the files. They are of far less practical significance than the lens codes I think.
Alec, it is not just about distorsion it is about periferal unsharpness too.
It’s a good question and I’d assume any corrections applied would be at least very similar to when the lenses were used on a Sony body. However there seems to be very low demand for this combination based on topics on the adapter on here and the fact it’s not an official combination (I doubt Sony want Nikon users mounting their lenses) I can understand why the combination isn’t supported. Just to be clear nor am I expecting it to be, I was just looking for a way around the software always trying to load the wrong modules for the lenses.
Stenis
(Sten-Åke Sändh (Sony, Win 11, PL 6, CO 16, PM Plus 6, XnView))
11
No it is not about Sony, I think. It is about DXO and how they have designed Photolab entirely. Sony have probably no impact on that at all. It is hard also to critisize DXO for designing controls that ensures we use the right lens profiles to the right lenses and nothing else. They have after all probably done all they could to adapt these profiles to the lenses they were made for. I have nothing negative to say about that at all. Making profiles for lenses from other systems on adapters is probably not their main concern.
Lens Correction is a big improvement compared to old “Unsharp mask” and the reason to that is all these excellent lens profiles.
@Stenis I’ve done some hands on testing of this myself. If the alternate lens is not wildly different in resolution, Lens Sharpening works quite well with different profiles.
What lens sharpening mostly does is improve microcontrast which is mostly dependent on optical resolution. I.e. a lens with coarser glass particles needs different lens sharpening than one with very fine glass particles.
Using extreme settings on lens sharpening may have something to do with distortion, but generally distortion is handled by a dedicated PhotoLab
palette called eponymously Distortion.
Stenis
(Sten-Åke Sändh (Sony, Win 11, PL 6, CO 16, PM Plus 6, XnView))
13
That is actually bad news because Microcontrast is often pretty problematic since it often induces a lot of “uncleanliness” is slies e.t.c.
@Stenis it’s not as bad as it sounds. I mostly use the Fine Contrast slider which works very well. Lens Sharpening does something similar but more calibrated to the resolution of the lens (coarseness of the glass, nature of the grain). Nature of the grain can be very lens specific but is usually similar at similar coarseness.
Lens Sharpening is an absolutely spectacular sharpening tool. I’d rather enjoy a lite version based on a similar lens than nothing at all. If the lens is a bad match, delete it and try another one.
One can get the resolution of recent lenses from DxOMark.com. Count on older lenses to be around 12 to 15 MP resolution. I.e. to take a fairly old lens from DxOMark.com. Without lens EXIF data, there’s no offer for a lens at all, so one has to insert EXIF data to receive offers of matching lenses.
Similar substitution can be extremely useful for someone using a third-party lens in a different mount through an adapter.
Anyways, I consider PhotoLab to be a speciality tool that is good for supported gear and not interesting for anyone using unsupported gear - unless one wants to tweak exif tags in order to make PL use a module that arranges some other lens’ shortcomings, which can introduce some additional errors.
If only DxO could allow users to freely select modules! We’d not have to edit tags but could flip through modules and see what we like best. As of now, we get best processing for the least amount of supported gear. What is the product of infinity multiplied by zero? A limited user base?
Lesson: Don’t use PL unless you (only) use supported gear.
4 Likes
Stenis
(Sten-Åke Sändh (Sony, Win 11, PL 6, CO 16, PM Plus 6, XnView))
16
This statement Alec is doubtful and it is very easy to get confirmed using DXO Lens at DXO Mark.
I happen to have one of Sonys sharpest lenses, the 90mm/2,8 macro. If that lens is put on a Sony A7r IV body with 66 MP I get a far higher score than if I put it on a 24 MP NEX 7 or for that matter on any Sony FF model with 24 MP.
Second on the NEX 7 with 24 MP which gives modest 16 perceptual MP.
Same lens and totally different score depending on your camera’s pixel count.
A comparison like this always use to make Fuji users very irritated when they for many years only could get access to camera bodies with 24 MP and I always wondered why it took Fuji so long to break the chains of Sony when it came to sensors. Was this just a strategy from Sony to make people buy the more expensive A7-cameras instead?
Many, many years DXO didn´t just ignore Fuji-cameras in their testing because of the XTRANS-problems and many, many years there were no Camera and lens profiles for Fuji in Optics Pro and Photolab. So many years we had no clue really how their cameras and lenses really performed because they never got tested. Good that have changed but still in version 8 I think there are limitations when it comes to Deep Prime XD2s (if it hasn´t been fixed by now).
It is very interesting that all these testing activities of DXO has got a spinoff in the business best lens profiles in any RAW-converter. That is really a comparative advantage of DXO that we all here can be glad for.
It depends also on various AA/UV/IR filters, color arrays, and perhaps readout noise and crosstalk. For example A7RIV has no low-pass filter (AA), while NEX-7 probably has a quite strong one (please correct me if I’m wrong, but that would also “explain” huge difference in quantum efficiency between the cameras). Hence worse than expected result for NEX-7.
BTW, A7RIV is FF 61 Mpx (not 66 as mentioned), while NEX-7 has 24Mpx APS-C sensor (that would make 54Mpx on FF, so pixel sizes are quite similar). The mentioned lens also has some strange characteristics, being sharpest in the middle and FF corners, a bit softer otherwise (but still deadly sharp).
Stenis
(Sten-Åke Sändh (Sony, Win 11, PL 6, CO 16, PM Plus 6, XnView))
18
New example a Canon-example from an age closer to NEX 7 than A7r IV is, both with FF-sensors: 5DsR (51 MP) and 5D MK II (21 MP). Lens - one of Canons sharpest.
You start to talk about low pass filter, read out noise e.t.c but it seems that the pixel count is the real factor to consider here. It is domination the results.
If we start to divide 51 pixels with 21 we get 2,4.
If we then divide the perceptual megapixel values 42 with 18 we get 2,33.
Not a very big difference there is it?
They correlate pretty well, don´t they.
So, there is absolutely crucial to the perceptual MP to have a camera with as high pixel count as possible.
If I have a FF model like Sony A7 III (24 MP), it will definitely suffer from it`s low MP values compared to Sony A7r IV or V (61 MP).
The same goes for any other camera brands that sells FF-models.
… and as I said Fuji-huggers have many times down played these fact when they were lockes in at 24 MP because of their dependence of Sony sensors and a Sony totally uninterested in increasing the MP-rate in APS-C sensors since they handled this very much as a positioning issue.
APS-C was not seen as a professional segment at all.
Entusiasts and professionals were supposed to buy FF and APS was for the mass market.
Sure, you are right, but I was only trying to explain DxOmark strange results for NEX7 and A7r4, with NEX7 resolution much lower than expected.
BTW, I don’t look at DxO scores for the lenses. Their MP score seems to be undefined, although it clearly relies on pixel count. The 'field map’s provide some insight for sharpness, vignetting, CA. But when making decision on buying some lens, I still rely on downloadable RAWs of images which match my needs. I’ve never bought a lens that wasn’t at least for 6 months on the market. For example, there’s no good measure for “bokeh quality”, but if you shoot everything sharp, it would be irrelevant anyway (e.g. Voigtlander APO). Similar for “ghosts”, coma, etc. Autofocus performance is another example. DxO camera measurement results are more interesting, imho.
Saying definitely is way too much, I think. IMHO, my 45mpx (Nikon Z8) camera photos are only circa 10-20% “better” on average than my old 12mpx (Nikon D700) photos, when judged on 4K monitor. But I concentrate mostly on the “mood” and “meaning”, and don’t care too much about the details, unless they influence the overall perception or the “message” reception. Your case may be different.
For fashion photography, neither FF was seen as “professional”
There are so many breath-taking 4/3 photos around, though (especially nature).
Some “sillyphones” aren’t that “silly” either.
Depends on the purpose, as usual.