Showing all 201 replies.
>>
>>
File: 0707001752654_099_03954u.jpg (1.6 MB)
https://en.wikipedia.org/wiki/Sergey_Prokudin-Gorsky
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
File: 03959u.jpg (1.5 MB)
I "stole" them from the Library of Congress Archive.
here https://www.loc.gov/pictures/collection/prok/
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>russian empire
Prokudin-Gorsky - he went to Paris
>90s russia
Pinkhassov - the only "russian" in Magnum, married and went to Paris
>today's russia
Markov - pathetic fucking junkie who stayed in russia, couldn't take it anymore and died because of overdose
Really makes you think
>>
>>
File: 04652u.jpg (1.2 MB)
>>4505563
yes, you can see marked differences in contrast in the 3 frame b/ws
The process is still mind boggling to me, it's b/w photography.
Does not compute
Check this river shot. That's him in picrel.
>>
>>
>>
>>
>>
>>
>>
>>
File: 0707001717066_028_00906a.jpg (1002.8 KB)
>>4505588
The LOC are the ones who rendered these.
My understanding is they replicated the 3 phase negatives
and merged them on a white background as seen above.
I am sure the finer lens quality, image alignment and 100 other
peripherals added to an overall enhancement.
Also they had to take a pic of the pic so to speak, so the camera used
would also need to be considered.
Fyi the 3 frame b/ws have the same clarity
I can't crop this right now , it's just an example
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
File: camera.png (253.4 KB)
>>4505829
more to come ;)
>>4505830
"There is no known replica or illustration of the camera that Prokudin-Gorskii used. It was a view camera of his own design, perhaps similar to a model [left] designed about 1906 by Dr. Adolf Miethe, whom Prokudin-Gorskii had met previously in Germany".
more here on the camera tech https://www.loc.gov/exhibits/empire/making.html
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
File: 04660u.jpg (1.2 MB)
If you discard the epic time frame, the epic locations
and the epic state of the art photog method
a lot of the the shots are true art.
The better photos stand up artistically in any era.
That's the Joy of Photography.
That moment we have captured one of those Greats and feel a little bit part of the Art.
>>
>>
>>
>>
>>
>>4505988
88, great painter. Just thought I'd point that out.
Not only that, but taking a photo back then wasn't exactly just snapshitting away, not to mention also very expensive so of course they paid much more attention to composition.
>>
>>
File: 04658u.jpg (1.2 MB)
>>4505830
>>4505993
I posted a 3 frame b/w here. >>4505619
That is the original un-cropped negative print.
The lenses had to be in a stack with the center lens on center.
No external mirrors.
Look carefully at the images.
Top left corner roofline is off set in a down/up angle in the top and bottom images
Same with the straight line made by the steps. Smaller line on top
huge line on bottom.
Bottom lens pointed slightly up, top lens pointed down.
Mega long glass plate.
>>
File: 04669u.jpg (755.3 KB)
>>4505962
In full disclosure of the 1400 pics 1/3 were strictly
government type stuff with no real subject matter.
Major river junctions, rail and farmland shots.
While cool for the historical aspect not much to see.
I guess it makes the money shots that much better
cause you can see what was capable by this crew.
picrel
>>
File: Screenshot_20241202_103923_Chrome.jpg (245.5 KB)
>>4505996
Didn't see that post before making my own.
These cameras were designed for taking trichromes using one lens and 3 sheets of film, but the 3 lens setup on a single big negative also makes perfect sense.
>>
>>
>>
>>
File: 04666u.jpg (997.7 KB)
oh fyi
This only one of the killer sets I have/errr created.
I have a huge archive of many topics wsf related.
Western Art
War (most conflicts)
Awesome Architecture folder
Planes, Trains and Automobiles
etc, etc
All super Hi Res, cause that's how I pho/t/o roll (ouch) :)
Not being a dick but there are several old dormant threads
on the board and I could post in several threads a day.
Suggest a topic and i could make another thread
>>
>>
>>
File: 1346.jpg (861.2 KB)
>>4506018
You want the autochromes sounds like.
Those are numerous but not in high res.
The color substrates cloud the glass
vastly reducing image size and quality.
2 major sets for 3/4 threads. !880ish-c1930
I have another well worth mentioning done recently.
Great shots in cemeteries and mausoleums.
picrel
>>
>>
File: 20194u.jpg (1011.8 KB)
If anyone is interested in chilling and watching a slideshow of the pics in this thread i made one
here https://www.bitchute.com/video/ugPuZKa43h4x/
>>
Proof you don't need the latest canonikon 2026 edition f1.4 L VCM to get sharp good looking images.
And the guy had to do three simultaneous exposures with red, green and blue filters and combine them during development to get a color imag. The man was basically doing color stacking, hence the color fringing on some images as each frame is never perfectly aligned with the others.
>>
>>4506033
Lenses have been good enough for 100+ years now, and technically speaking wetplate/dryplate has significantly finer grain than digital or modern films. The lenses couldn't really push resolution as far as modern lenses can, but that didn't really matter when the plate was the final image, or you were contact printing glass negatives.
In a sort of ironic way the push towards lenses with corner to corner sharpness and minimal distortion was just as big of a deal back in the early 1900s as it is today. Petzvals and their like had great sharpness in the center, were very fast for the time, but had lots distortion, major sharpness falloff and a curved focal plane.
>>
>>4506036
You're looking at compressed jpgs of three overlaid images, they're not really representative of the best IQ from that time.
4x5 wetplate can do atleast 1000MP, but considering the grain size is on an atomic scale it could theoretically get much better than that if you had a crazy 4x5 lens that could resolve more than that.
Are you pricing drum scanned slide film at 50 dollars a pop?
OP please post a 100% crop from one of the 100MB tiffs.
>>
>>
File: 04665u.jpg (1.4 MB)
>>4506038
this.
old school wet plate smokes almost any thing
all that is on another drive and will search tomorrow.
this was in the Windows7 days and storage was an issue.
I am certain I saved some of them for archive purposes.
10pics x100mb back then was enormous
we will see
>>
>>
Great thread, OP. I too have family photos from that time period in old imperial Russia but they are all black and white and what stuns me is that the images are so much sharper and pleasant to the eye than the later 40s 50s 60s photos in our collection.
>>
>>
>the negatives were recorded on were finally stored in the basement of a Parisian apartment building, and the family was worried about them getting damaged. The United States Library of Congress purchased the material from Prokudin-Gorsky's heirs in 1948 for $3,500–$5,000
>The library counted 1,902 negatives and 710 album prints without corresponding negatives in the collection
>$4,000 in 1948 is equivalent in purchasing power to approximately $54,807.14 in 2026.
Not bad
>>
>>
>>
>>
>>
>>
>>
>>
>>
File: 04663u.jpg (1.3 MB)
>>4506184
>>4506038
ok
Down the rabbit hole of compression.
All the details come back as I pull back the curtain.
The pics posted in the thread are 2x compressed
tiff<jpeg<jpeg
99% of the 2nd gen were to large, Over 5mb.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>4506196
Crop the tiff to a small section then export as a jpg at 100% size. The center of the image will be sharper than towards the edges.
Cropping small enough will allow you to convert to a file under 5mb without compression. That way we can see a closer approximation of what the scanner actually got out of the negatives. A b&w one with just a single channel and a color could be cool also to see how much detail is lost when overlaying the 3 images.
>>
>>
File: Rabbit Whole.png (105.1 KB)
>>4506216
Hey friend
I never really surfed around on the LOC site after I got all the collections I wanted.
This whole thing is a rabbit hole into the abyss.
Way beyond my skill set.
I compiled ready access to exact details from the various processes they used.
Confusingly the cropped ones posted are a different variant
from the color overlay/compositions
ie this pic >>4505565
has at least 2 possible sources
I linked it in the LOC here https://www.loc.gov/pictures/item/2018679177/
Details are here for the numerous variations used picrel
I would like to know how this all sorts out as I came here as some kind of Curator.
>>
>>
>>
>>
>>
>>
File: master-pnp-prok-00200-00253a.jpg (855.8 KB)
>>4506310
Nice. Ty for the info. Here is a 100% crop of one of his images to view the "full" detail acquired from the archival scans. Crop was taken from the middle frame.
Each frame in negative was around 3x3.5 inches in size, and these were scanned at 1010ppi as your info says.
There's some haze on this image, but you can see that the lens and film was, IMO, absolutely "good enough" to take a detailed photograph that would print or project very well. You can tell that there is still detail in the negative that the scanner did not capture, and absolutely zero grain. I bet that a negative that didn't have the slight haze that this one has would reveal even more detail and contrast.
Interesting to note that the upper and lower frames on the glass negative are noticeably softer than the middle one.
>>
>>
>>
>>
File: 0707001752654_084_04420u.jpg (1.0 MB)
>>4506317
Glass plates are the Goat. I built my collection around them.
Even a 4x6 bw print from a glass negative can reveal incredible details when viewed with a magnifying glass.
I have one question, if you could.
this pic >>4506314 and >>4506315
were created differently
14 is RGB 3 panel overlay as we understand
wtf is 15. It is is described as a different render
that part is the question. what did they do??
Also I am convinced he had the lenses stacked on top of each other.
I noticed the contrast variations which leads the question;
in what order are the gels set up? Simple answer probably but idk.
Thanks for your time anon :)
>>
File: 20306u.jpg (1.4 MB)
>>4506321
>>4506346
Jumping in here anons
This picrel is from the Finland/Russo border.
You could place it in any of the 49 States or Canada
and it would look like a local build.
This style is the most conmen in the forested regions.
I believe this timber style construction originated
in Russia and propagated West.
ie they were building these structures in Russia
before they were building them in the US or elsewhere.
>>
>>
>>
>>
>>
>>
File: 20267u.jpg (2.1 MB)
We only skimmed the surface.
Most of the Gold never made it to the rest of the internet.
All the rest is uncharted territory. c.1250 more pics.
I never clicked all 1400. Thought I would leave some surprises.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>4506367
The lenses were definitely stacked. You can tell because the full glass plates are oriented in a top down direction. The aspect ratio is very close to 1:1, so you wouldn't need to rotate the plate holder.
Its difficult to say what they meant by two different renderings. They could have possibly used a different color value for each of the frames before overlaying. My guess is that the church picture had additional post processing that the town landscape did not. Straight RGB vs a more tuned color overlay to produce a balanced looking photo. You can see it in the layer mismatch in the tree leaves.
>>
File: 20218u.jpg (1.1 MB)
>>4506441
Excellent work. You figured it out.
A standard value for the rgb A tuned ratio
that they could use as a default across the cropped samples.
I would accept these cropped pics as original representations of the color palate
in the process above. The tech limits of the time would have prevented
Gorsky from fine tuning his results. Given the man he was, He would have been front and center in the LOC render room doing this exact same thing.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>4505996
>>4506000
>The lenses had to be in a stack with the center lens on center.
if it was taken with three lenses, we'd see way more parallax "CA"
but all I've seen is from stuff moving in-frame in different colors, which also implies multiple shots on a single lens
>Top left corner roofline is off set in a down/up angle in the top and bottom images
the framing offsets in >>4505619 are probably from imperfect alignment when moving the plate between the shots, otherwise the top & bottom images would be offset in opposite directions relative to the center one
so yeah, all evidence points to a setup like Miethe's 1906 >>4505938 on top
>>
>>4511141
Thank for your input and I can't argue with the logic.
A closer comparison to other three frames would be important.
If the declamation in framing observed in the sample I gave is consistently present,
I would say no to the theory.
You would not move the base the same way each time in other locations for other shots.
My description includes your offset angles approach.
Center lens straight
top lens 20 degrees down
bottom lens 20 degrees up.
Odd none of his gear was in any photos. The Mirthe camera seems gigantic and cumbersome. I don't see how you could slide into the next frame regularly
without a noticeable difference in border framing.
The weight of that slide device with the huge glass plate
on the back of that camera is enormous and may be the answer.
>but all I've seen is from stuff moving in-frame in different colors, which also implies multiple shots on a single lens.
Time is also an issue here.
timing for three different exposures, lets calculate the time for three frames versus one.
Then compare the ghosting (color shift) you see in a non cropped example.
That much time between getting all three shots and you have nothing but a blur.
Forget the grass waving in the breeze
As I said you made good points but I have also and am not convinced yet,