Sigma or Panasonic macro lens for scanning film

pkr1979 said:

But surely it has to be possible to arrive to a conclusion? Wouldn't it be possible to 'fact check' those claims - I assume they cant be both right?

Click to expand...

My conclusions would be:
a. either lens should produce acceptable results, and,
b. once the goal of acceptable has been achieved, then any additional time spent searching for better-than-acceptable would probably be better spent actually digitizing your negatives.

So my recommendation would to pick one of these two lenses, and get to work. Only after you have digitized several hundred negatives and done the post-processing on them will you have the experience needed to know the strengths and weakness of your process. What I am trying to say is, only after you have copied a lot of negatives will you know what is acceptable for you, and only then will you be able to make an informed decision about what part of your process is the weakest link. I'm guessing it will not be the lens, no matter which one of these two you choose.

But if at some point you do decide your process needs a better lens, then that is the price of your education. I started out with a decent, but far-from-exotic Schneider-Kreuznach Componon-S 100mm enlarging lens on a bellows. Later I upgraded to a Rodenstock APO-Rodagon D 75mm duplicating lens. I can see a small improvement in sharpness and detail, but honestly, the modestly priced enlarger lens was probably good enough for my purposes. But only after I was able to try them both was I able to learn what "good enough" means to me.

_T_ said:

It is possible to fact check claims, but doing so requires you to do the research.

Right now there are some technical, or at least technically inclined, tests of the Sigma lens, but there are none of the Panasonic.

So you could ask different ai agents all day, but you could never possibly get an answer because there is no data to compare.

Click to expand...

A lot of the data we do have. And the one thats missing is software uncorrected distortion in the Panasonic. Which Panasonic says needs to be applied. One can assume then that its not very different from the Sigma. For film scanning purposes there is then a significant chance the Sigma is the best suited.

runswithsizzers said:

My conclusions would be:
a. either lens should produce acceptable results, and,
b. once the goal of acceptable has been achieved, then any additional time spent searching for better-than-acceptable would probably be better spent actually digitizing your negatives.

So my recommendation would to pick one of these two lenses, and get to work. Only after you have digitized several hundred negatives and done the post-processing on them will you have the experience needed to know the strengths and weakness of your process. What I am trying to say is, only after you have copied a lot of negatives will you know what is acceptable for you, and only then will you be able to make an informed decision about what part of your process is the weakest link. I'm guessing it will not be the lens, no matter which one of these two you choose.

But if at some point you do decide your process needs a better lens, then that is the price of your education. I started out with a decent, but far-from-exotic Schneider-Kreuznach Componon-S 100mm enlarging lens on a bellows. Later I upgraded to a Rodenstock APO-Rodagon D 75mm duplicating lens. I can see a small improvement in sharpness and detail, but honestly, the modestly priced enlarger lens was probably good enough for my purposes. But only after I was able to try them both was I able to learn what "good enough" means to me.

Click to expand...

I agree - theres a pretty big chance I would not be able to tell them apart. And I suspect that I would go for the Panasonic for general photography. But as said - for scanning Im probably going for the Sigma (its also cheaper).

koraks said:

The question is how well it does so. IME it depends; sometimes it does a fairly good job, sometimes it 'overlooks' crucial points. It also depends a heck of a lot on how much prior data is available on the question asked, and the fuzziness with which seemingly related, but different instances are going to be included. E.g. when asking about the "Sigma 105mm f/2.8 DG DN Macro Art and the Panasonic Lumix S 100mm f/2.8 Macro", there's a good chance that data associated with other Sigma Art lenses, other brands of 105mm lenses, Panasonic Lumix cameras (instead of lenses) etc. etc. will be included in the linguistic mix. Since AI cannot differentiate between such things, as long as the letters are sufficiently similar, it'll all treat it as part of the same complex. In comparisons like this one, this can become a major problem.

If AI is to be used in a case like this, I'd always try to use a tool that at least allows for checking its references. Something like Perplexity for instance: https://www.perplexity.ai/search/out-of-these-two-which-is-the-n1htxzc.TRm7YgWSg_tZDQ While that does allow for tracking the sources used, note that the answer (as is typical for LLM's) is for the most part tangentially related to the actual question at best, and consists to a large extent of irrelevant filler.

Click to expand...

Well as I said, Ai as it stands is the new Ken Rockwell and a jumping off point for further research, but it can only come up with likes or dislikes if people have aired them on the internet, it can only come up with product specs if they've been published on the internet.

And if your further research comes down one way or another and you've picked the most perfectly designed lens you still don't know if your example is the most perfectly made lens because there can be a vast difference between design and assembly. So all this worrying about Ai is moot given a graph somebody shows of a lenses good and bad points may be from the ultra checked and tested example sent out to reviewers. Hence 'why worry about it', it's whoever you trust to build the lens they've designed that matters. An example. When Leica introduced the R series they borrowed some lens designs from their collaborators Minolta. They were very good designs, there were only so many ways to design a 24mm lens and Minolta's was one of the best. So while Leica copied the design they improved accuracy, they changed some of the glass formula, and they built them to much tighter and consistent tolerances, hence the higher price, and it was a 'better' lens for it. And it's no different in more recent years, at the turn of the century Nikon made some of the best designed lenses ever for camera kits, but built to the cheapest standards. So take an overview, but don't believe that if one graph beats another the lens you buy based on that is going to be exactly the same.