Leica champion finally frustrated with Leica SL3, help!

[jaapv]

Advertisement (gone after registration)

What is over-manipulation, given that the noise reduction in LR is AI now as well. You could get around it by converting in DXO, I suppose; it is not labelled AI AFAIK. 

[ALScott]

Totally with you, it's vague at best and looking at past winners I can't believe none of them have used noise reduction.  I have thought about calling to ask someone but I doubt that will get anywhere.  May call anyway just to see. 

[jaapv]

I think that the distinction to generative AI is lost on them - anyway, as Adobe has AI-driven tools all over the place it is a lost battle anyway.

[Jeff S]

No matter, soon the judges will be AI tools.

[PhotoCruiser]

In Italy during underwater photography contests the photographer has to hand the SD card immediately after the dive and entering the boat to the judge aboard to avoid manipulations. This is already many years like this, long before AI became widely known and implemented, but will obviously not work for online contest.

However, AI will become a problem for competitions where AI could be used to have a advantage and soon AI will be so good that it's virtually impossible to discover the use of it. 

Honestly i am not really sure if Ai is inside everything where Ai is written on it and guess that many brands use actual buzzword AI to raise sales.

Yes, DOX PureRaw does not mention AI, i will download a evaluation copy to check if it's in fact reasonable better than same LrC functions.

Chris
 

 

 

Edited 11 hours ago by PhotoCruiser

[jaapv]

Well, given that cameras use AI for AF and matrix metering and who knows what is in the pipeline, there are other complications. However, insisting on the use of Content Credentials should go a long way to resolving these dilemmas. 

[Stuart Richardson]

Advertisement (gone after registration)

I think the contests are trying to grapple with the full detachment from reality that is now possible with the sophistication and ease of AI tools and generative models. If they want the work to be representative of the real observable universe then they have to draw the line somewhere, and limiting the use of AI post process editing is a better demarcation than most. 
 

Edited 7 hours ago by Stuart Richardson

[jaapv]

We had this discussion in the postprocessing forum and this one will be as inconclusive. 
we must ask ourselves first what reality is in photography. To me, a rendering of the perception of the photographer- which may be radically different from the perception of someone else. It may even need AI tools to get as close as possible. 

We need to look at results, not methods. 

And then, which methods. AI is too broad a concept. Matrix metering, AF subject recognition, how are you going to exclude that? Further on, noise reduction, dust removal, AI selection tools, what about those?  We are allowed to play with colour, convert to monochrome even,  but why exclude recolorizing with a Neural Filter? And so on. 
 

If we attempt to repress, the battle is lost. The only way to make sense of this is transparency about methods related to the result. 

[LD_50]

17 hours ago, ALScott said:

I have been an ardent supporter of all things Leica since making a heavy investment into the system a little over a year ago starting with the Q3.  Moved to SL3 soon thereafter then M11P, then SL3-S and an embarrassing amount of lenses along the way.  This last weekend was the first time I truly felt let down by the system.  I was trying to shoot some deer and the AF would bounce like crazy making me completely miss some shots.  There was a shot with the yearlings below where one came up and nuzzled its mom and the camera completely blew it.  I was focused but had to move a bit to get the full shot I wanted and it bounce focus to nowhere land and I missed it.  It was bouncing focus more than it ever has before.  Where I am it's rare to see this and it completely blew the shot.  Also, on the shots frame to frame the focus is off just a bit, enough that I have focus on one shot but just off on the shot I really want where they are facing the same direction.  Blow it up and you can see.  In LrC all that has been done is crop and auto adjust.  Does anyone have tips on how to make the AF work better?  I was shooting in AF Field mode.  Thanks for the help.  I am so disappointed in how this went I was seriously considering dumping all SL for Sony and I really, really do not want to do that but this was ridiculous. These were shot with Sigma 15-600, same ISO, shutter speed and aperture.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

AF settings matter a lot. How did you have the AF system set up? I find with each system there’s a learning curve on settings and associated technique.

I haven’t had any issues shooting anything like what you posted here. I posted some SL3-S shots in my thread on the move from SL2-S. I included AF settings

[Jean-Michel]

I do not want to derail the thread, still, I would like to add something about "post processing". For one, it is not post processing, it is simply processing: developing film, enlarging, spotting, matting and framing are processes, LR, PD, and whatever are also simply processing, nothing "post" about that.

As to AI, and such: here too, there is nothing particular new. We had darkroom magicians and when they failed to produce the very best result we sent the prints to air-brushing artists. I welcome the new tools in LR or PS that let me remove dust, and even such element as unsightly wires in an image — I once had to spend way too many hours spotting out that kind of stuff from large format negatives.

For a while there was a fad to fully printing an image showing the negative's frame edges — I did that for a short while — that proved absolutely nothing as to the purity of the image.

While I wish they did, none of my cameras have the Content Credential feature. I do think that this is a very important, maybe even essential, feature for documentary imaging; much less so for other works. 

[LD_50]

1 hour ago, Jean-Michel said:

I do not want to derail the thread, still, I would like to add something about "post processing". For one, it is not post processing, it is simply processing: developing film, enlarging, spotting, matting and framing are processes, LR, PD, and whatever are also simply processing, nothing "post" about that.

As to AI, and such: here too, there is nothing particular new. We had darkroom magicians and when they failed to produce the very best result we sent the prints to air-brushing artists. I welcome the new tools in LR or PS that let me remove dust, and even such element as unsightly wires in an image — I once had to spend way too many hours spotting out that kind of stuff from large format negatives.

For a while there was a fad to fully printing an image showing the negative's frame edges — I did that for a short while — that proved absolutely nothing as to the purity of the image.

While I wish they did, none of my cameras have the Content Credential feature. I do think that this is a very important, maybe even essential, feature for documentary imaging; much less so for other works. 

I would assume the “post” in post-processing refers to processing outside the camera, rather than the early digital days of using on-camera processed jpeg files.