New film simulation software

[Stevenson Gawen]

Andrea over on discuss.pixls.us has recently published an experimental film simulation program which is pretty much unique as far as I know in that it actually simulated in software the complete chemical and physical processes in the film development and printing process, unlike all the others I've seen that simply recreate the colour palette of the end result.

He's anxious to point out that it's not finished software yet, but it is usable and I for one am quite impressed. The halation especially is very impressive - there's quite a few sample images in this thread: https://discuss.pixls.us/t/spectral-film-simulations-from-scratch/48209

The software is written in Python, so it's not quite a one click installation, but with some help from a friend and some more from ChatGPT (!) I got it running on my Windows laptop and am loving some of the results I'm seeing. It's slow though.
Here's a recent street shot from me, shot on my D750 with 50mm f1.8D.

 

[Gary R. Smith]

Once upon a time I was semi-active on pixls.us but I can't seem to find my password. I'm confused as to what exactly this s/w is trying to simulate and how or for that matter how it differs from what is done in FilmLab. Any file that you'd use would have to be digital and as far as I know no s/w exists that converts the film chemistry and/or physical properties along with the color.

I never bothered to learn Python but I'd be interested in better understanding what exactly is being done.

I read through the discussion and would have commented but (so far) I can't find my pw.

In the end isn't it always just about the color?

 

[Gary Ayala]

Fuji, (my camera of choice), has film simulations as standard choices built into its X series cameras. I haven't found a need/urge to use them, but I have read they are quite good. I know this has no direct bearing on "New film simulation software" ... so I'm just sayin'...

 

[Gary R. Smith]

Fuji, has film simulations as standard choices built into its X series cameras

I tried hard to own an x100vi the night they came out. I failed.

On a related topic: I wonder why there are two versions of this post and I wonder why Beth's last name has changed.

So many questions!

 

[Gary Ayala]

I tried hard to own an x100vi the night they came out. I failed.

On a related topic: I wonder why there are two versions of this post and I wonder why Beth's last name has changed.

So many questions!

I loved my X100S. It took a great image. When I wanted to go simple ... wanted a camera for no particular purpose ... I'd grab the X100S. No real thinking required (just like my photography).

The multiple posts could be due to zeal ... maybe a stutter ... dunno?

You're right on Beth, she used to be a Carter. Hopefully the change is a positive thing for her. (Like she finally got accepted into the Federal Witness Protection Program. )

 

[Beth Anthony]

i just realized there was two threads on this, too.. my understanding on this is he's trying to give an authentic film look to digital files. looks like he's mapped out the colors and tones of films into gradients rather than just apply a general recipe like most film simulations do. very few of them consider halation.

nothing too spicy on the name change. my state job had a long term assignment and we were told (oddly enough, by the feds) to stop using our real names online. that assignment is over. pete was kind enough to change my name for me in the meantime.

 

[Gary Ayala]

... my state job had a long term assignment and we were told (oddly enough, by the feds) to stop using our real names online. that assignment is over. pete was kind enough to change my name for me in the meantime.

So I was correct

 

[Beth Anthony]

So I was correct

not entirely. i didn't finally get accepted into it, i finally got out. the whole assignment was bs anyway. but they did check on our online activities.

 

[Gary Ayala]

not entirely. i didn't finally get accepted into it, i finally got out. the whole assignment was bs anyway. but they did check on our online activities.

Got it ... you got kicked out after monitoring your online activities ... (I still think your photos are good ...)

 

[Beth Anthony]

Got it ... you got kicked out after monitoring your online activities ... (I still think your photos are good ...)

more like someone else got kicked out. i'll give you a hint, i work in delaware and my office's additional duties lasted from mid 2019 to the jan 20th this year. someone had to come home just about every single weekend and up to 3 trips in a week. we had a few cops who almost ended up divorced over the mandatory ot.

 

[Charlotte Ortelli]

Andrea over on discuss.pixls.us has recently published an experimental film simulation program which is pretty much unique as far as I know in that it actually simulated in software the complete chemical and physical processes in the film development and printing process, unlike all the others I've seen that simply recreate the colour palette of the end result.

He's anxious to point out that it's not finished software yet, but it is usable and I for one am quite impressed. The halation especially is very impressive - there's quite a few sample images in this thread: https://discuss.pixls.us/t/spectral-film-simulations-from-scratch/48209

The software is written in Python, so it's not quite a one click installation, but with some help from a friend and some more from ChatGPT (!) I got it running on my Windows laptop and am loving some of the results I'm seeing. It's slow though.
Here's a recent street shot from me, shot on my D750 with 50mm f1.8D.

View attachment 21765

Quite an interesting write up, most approach simulation through more simplistic color matching through color charts.
Extremely impressive results so far, there are some very smart people in the photography community

 

[Gord Tomlin]

Fuji, (my camera of choice), has film simulations as standard choices built into its X series cameras.

I believe that Fuji's film simulations are only used if you shoot JPEG. If you shoot raw, all you get is a bit of EXIF info saying what film sim was set in camera. It is possible for software to pick up that EXIF info and then "simulate the simulation". The software being discussed on pixls.us works with raw files.

Rant: Some recent Fuji camera models have used the dial on the left side of the top of the body for film simulation selection. I have no beef with someone wanting to used film simulations, but for those of us that shoot raw this is a waste of a scarce resource (three dial locations).

 

[Gary Ayala]

I believe that Fuji's film simulations are only used if you shoot JPEG. If you shoot raw, all you get is a bit of EXIF info saying what film sim was set in camera. It is possible for software to pick up that EXIF info and then "simulate the simulation". The software being discussed on pixls.us works with raw files.

Rant: Some recent Fuji camera models have used the dial on the left side of the top of the body for film simulation selection. I have no beef with someone wanting to used film simulations, but for those of us that shoot raw this is a waste of a scarce resource (three dial locations).

I've owned a number of Fuji X cameras. None of the X cameras I owned had any dials assigned to film simulation. Presently, my newest Fuji is a X-T5. There are two 'Command Dials' on the XT-5, (front and back), which can be assigned to selected tasks (but these command dials are not on the left and do not come preassigned to film simulation). One of the things I appreciate with Fuji X cameras, is ALL 'primary' exposure controls are manual and on the exterior of the camera.

As I shoot RAW, I have no experience with in-camera software film simulation.

 

[Gord Tomlin]

I am not referring to the command dials, but rather the dedicated dials on the top of the body. For instance, on your X-T5, you have (from left to right) ISO, shutter speed and exposure compensation. On my X-T20, I have drive mode, shutter speed and exposure compensation.

Some new models such as the X-M5 have a film simulation dial where your ISO dial is.