Buttocks aside, it's a nice picture - maybe more elegant then Bliss. But it's a lot less bright and happy. I'm glad they went with Bliss in the end, for whatever reason.
“Microsoft paid photographer Charles O’Rear a confidential amount for the Windows XP wallpaper “Bliss,” but it is widely reported to have been in the “low six figures,” meaning over $100,000.”
Charles should have asked for MS stock instead.
“In 2005, Facebook offered David Choe about $60,000 to paint murals at its office. Instead of cash, he chose Facebook stock. When Meta Platforms went public in 2012, his shares were estimated to be worth around $200 million.”
Does anyone have any idea what RZ67 lens was used? People have found the exact hill and from this one might be able to figure out from what perspective it was captured, thus which focal length it was shot at. I haven't found anyone who has done this, only vague, unconvincing speculation. Maybe confounders like fact that it is now covered by a vineyard and erosion long since changed the shape of the hillsides makes this impossible.
I work in the refurb department of an e-waste recycling company. I take pictures of monitors and TVs showing Bliss, and I test printers with it. It has bright spots, dark spots, it's colorful, and has plenty of fine detail, making it a decent test picture. Bonus points for being familiar to most people.
I always thought it was a synthetic image. I expect many others did too.
On some level, they chose the real image that appeared to be a synthetic image.
I stumbled onto MSN for a story about Bears in a sanctuary who had overhead ropes (horizontally laid), and what a difference it had made.
Actually, I don't remember the story that well. What I do remember is that MSN story used a GEN AI image. Fake bears, fake rope. There, of course, are real photos available.
But MS want that automation, dont want to pay writers, or editors, and don't want to pay royalties or seek permission for photographs.
Microsoft Design also released 4K renders of nostalgic wallpapers (including bliss) a few years ago. I can't find the original link but here's the reddit post with the pictures.
Back when Deep Learning wasn't just LLMs and diffusion models (approximately 5y ago), for my senior project at uni I did a image animation. In goes an image, out comes a short gif. It was trained via (reverse) optical flow.
I used this image for a demo how clouds move and the audience+professors all went WOOOW and that is now a core memory of mine
I’m a bit confused about the claim that the image was altered.
Sometimes skies look like that and grass looks like that and (the right) film is more than capable of capturing that with the appropriate saturation. Especially Velvia. Velvia is probably even cranking up the saturation, to levels you would not see like that with the naked eye.
Look at that first Tuscany image. The colors are a near perfect match. With the others the colors - especially the greens – can also be a lot more muted, however that seems to be down to darker greens as a starting point and also the light/weather (less saturation when it’s overcast and there is no direct light).
On close examination of the wallpaper (to a level of detail not visible on early 2000s screens) also shows all the hallmarks of a real photograph with remarkably little retouching.
On the left and especially the right you can see ugly clutter behind the hills which is only not distracting if you don’t examine the photo to closely. Anyone who photographs landscapes knows the issue of hard to hide clutter that nevertheless from my perspective also grounds the photograph in the real world.
Also clearly visible on the hills: tracks/paths through the hills. This is also something hard to avoid in landscape photography, though you try to minimize it with perspective. The same applies as to the clutter: my view is that this grounds the photograph as an actual photo.
Third hallmark of photography: the foreground grass is all out of focus! This is often hard to avoid. Techniques like focus stacking now exist, but as a single photograph that is often a trade off you have to make if your landscape shows both things close by and far away.
So, yeah, looks 100% like a real photograph and shows what a look Velvia is, mostly.
Microsoft bought all the right and even the original physical film (that I guess they would scan to get the best image possible). So I guess then Microsoft would be on it too.
It was shot on Velvia slide film. Knowing that emulsion you either expose it just right and it looks gorgeous or you over/underexpose and the details are gone and can’t be brought back.
People don’t realize that there’s no such thing as an “unedited photo” because either you’re making decisions in the darkroom or the software/firmware is making decisions in the camera.
I didn't know where it was taken, until I accidentally found myself there! https://www.simoncross.com/p/that-day-i-accidentally-visited...
For the curious, here is Red Moon Desert which, according to TFA, was almost used instead of Bliss but was rejected for looking too much like buttocks: https://windowswallpaper.miraheze.org/wiki/Red_moon_desert
Buttocks aside, it's a nice picture - maybe more elegant then Bliss. But it's a lot less bright and happy. I'm glad they went with Bliss in the end, for whatever reason.
Red Moon Dessert would've clashed with the clown colours XP had going on
It’d’ve been a great background for a red mars inspired alternate theme.
The name didn't help much either, I suppose.
How much did Bill Gates pay for Bliss?
ChatGPT:
“Microsoft paid photographer Charles O’Rear a confidential amount for the Windows XP wallpaper “Bliss,” but it is widely reported to have been in the “low six figures,” meaning over $100,000.”
Charles should have asked for MS stock instead.
“In 2005, Facebook offered David Choe about $60,000 to paint murals at its office. Instead of cash, he chose Facebook stock. When Meta Platforms went public in 2012, his shares were estimated to be worth around $200 million.”
https://stopcitingai.com/
MS was public at that time, he could have bought some directly.
Does anyone have any idea what RZ67 lens was used? People have found the exact hill and from this one might be able to figure out from what perspective it was captured, thus which focal length it was shot at. I haven't found anyone who has done this, only vague, unconvincing speculation. Maybe confounders like fact that it is now covered by a vineyard and erosion long since changed the shape of the hillsides makes this impossible.
I work in the refurb department of an e-waste recycling company. I take pictures of monitors and TVs showing Bliss, and I test printers with it. It has bright spots, dark spots, it's colorful, and has plenty of fine detail, making it a decent test picture. Bonus points for being familiar to most people.
Ironically, I only run Linux at work.
I always thought it was a synthetic image. I expect many others did too.
On some level, they chose the real image that appeared to be a synthetic image.
I stumbled onto MSN for a story about Bears in a sanctuary who had overhead ropes (horizontally laid), and what a difference it had made.
Actually, I don't remember the story that well. What I do remember is that MSN story used a GEN AI image. Fake bears, fake rope. There, of course, are real photos available.
But MS want that automation, dont want to pay writers, or editors, and don't want to pay royalties or seek permission for photographs.
Is this OK for your kids?
Similarly, the Windows wallpaper with the metallic bubbles is the Selfridges shop in the Bullring, Birmingham, UK:
https://en.wikipedia.org/wiki/Selfridges_Birmingham
High-res version for your modern desktop:
https://archive.org/details/bliss-600dpi
The actual *.tif is available which is even higher quality
https://archive.org/download/theoriginalfilesofsomewindowswa... (47mb)
https://archive.org/details/theoriginalfilesofsomewindowswal...
Microsoft Design also released 4K renders of nostalgic wallpapers (including bliss) a few years ago. I can't find the original link but here's the reddit post with the pictures.
https://www.reddit.com/r/windows/comments/ogpni5/microsoft_n...
Can you spot the bird?
Back when Deep Learning wasn't just LLMs and diffusion models (approximately 5y ago), for my senior project at uni I did a image animation. In goes an image, out comes a short gif. It was trained via (reverse) optical flow.
I used this image for a demo how clouds move and the audience+professors all went WOOOW and that is now a core memory of mine
I would love to see the actual negative and the other shots he took at that location.
Isn't there debate in the community that this photo was actually altered and that he has been lying about it?
I’m a bit confused about the claim that the image was altered.
Sometimes skies look like that and grass looks like that and (the right) film is more than capable of capturing that with the appropriate saturation. Especially Velvia. Velvia is probably even cranking up the saturation, to levels you would not see like that with the naked eye.
Here is a landscape photographer showing their own favorite Velvia photographs: https://www.macfilos.com/2022/12/02/vivid-velvia-ten-fujifil...
Look at that first Tuscany image. The colors are a near perfect match. With the others the colors - especially the greens – can also be a lot more muted, however that seems to be down to darker greens as a starting point and also the light/weather (less saturation when it’s overcast and there is no direct light).
On close examination of the wallpaper (to a level of detail not visible on early 2000s screens) also shows all the hallmarks of a real photograph with remarkably little retouching.
On the left and especially the right you can see ugly clutter behind the hills which is only not distracting if you don’t examine the photo to closely. Anyone who photographs landscapes knows the issue of hard to hide clutter that nevertheless from my perspective also grounds the photograph in the real world.
Also clearly visible on the hills: tracks/paths through the hills. This is also something hard to avoid in landscape photography, though you try to minimize it with perspective. The same applies as to the clutter: my view is that this grounds the photograph as an actual photo.
Third hallmark of photography: the foreground grass is all out of focus! This is often hard to avoid. Techniques like focus stacking now exist, but as a single photograph that is often a trade off you have to make if your landscape shows both things close by and far away.
So, yeah, looks 100% like a real photograph and shows what a look Velvia is, mostly.
It looks a little popped even for Velvia. He may not have enhanced it, but what are the odds no one at Microsoft did?
I always thought it was selected because it references the curve and most of the colours of the old Windows logo.
https://www.cleanpng.com/png-windows-7-microsoft-clip-art-wi...
Microsoft bought all the right and even the original physical film (that I guess they would scan to get the best image possible). So I guess then Microsoft would be on it too.
It was shot on Velvia slide film. Knowing that emulsion you either expose it just right and it looks gorgeous or you over/underexpose and the details are gone and can’t be brought back.
People don’t realize that there’s no such thing as an “unedited photo” because either you’re making decisions in the darkroom or the software/firmware is making decisions in the camera.
And photographic film emulsions are already biased one way or another, because a photograph is a memory, and we don't remember colors accurately:
https://youtu.be/GfhdKyyGU74?is=WNUqPA5-ILDzBVji
Very good job, if true.