Mercedes-Benz Vision V Concept MPV
seen from China
seen from New Caledonia

seen from United States
seen from Germany
seen from Vietnam

seen from Mexico
seen from China
seen from United States
seen from Hong Kong SAR China
seen from China
seen from United States
seen from Italy
seen from Mexico
seen from United States

seen from United States

seen from Malaysia

seen from United States
seen from Malaysia
seen from United States
seen from United States
Mercedes-Benz Vision V Concept MPV
Presented at the 1979 Paris Motor Show, French coachbuilder Heuliez created an MPV-type leisure vehicle based on a Peugeot 504 pick-up. The Peugeot 504 Loisirs was conceived as a versatile lifestyle vehicle that blended the utility of a truck with the passenger comfort of a station wagon. Although it was submitted to Peugeot for consideration as a production model, it remained a one-off prototype
Linux philosophy: You can do everything you want with your PC. Me: Okay, then I'll make my two favorite boys kiss in my terminal.
Goddamn it, I love libre, free, and open software.
All Purpose MPV by Diftong Studios
gif making help!!
gif makers, I cannot get MPV to work so I uninstalled and redownloaded it but now it won't even open on my Mac...can anyone help???
canmom's notes on fixing the colours
ok so if you've been following along on this blog for the last week or two i've been banging on about colour calibration. and i feel like it would be good to sum up what i've learned in a poast!
quick rundown on colour spaces
So. When you represent colour on a computer, you just have some numbers. Those numbers are passed to the monitor to tell it to turn some tiny lights up and down. The human visual system is capable of seeing a lot of colours, but your monitor can only display some of them. That's determined by its primaries, basically the exact colour* of its red, green and blue lights.
(*if you're wondering, the primaries are specified in terms of something called the CIELAB colour space, which is a model of all the different colours that humans can possibly see, devised by experiments in the early-mid 20th century where the subjects would turn lights at different frequencies up and down until they appeared visually the same. Through this, we mapped out how eyes respond to light, enabling basically everything that follows. Most human eyes tend to respond in pretty close to identical ways - of course, some people are colourblind, which adds an extra complication!)
Now, the problem we face is that every display is different. In particular, different displays have different primaries. The space in between the primaries is the gamut - the set of all colours that a display can represent. You can learn more about this concept on this excellent interactive page by Bartosz Ciechanowski.
The gamut is combined with other things like a white point and a gamma function to map numbers nonlinearly to amounts of light. All these bits of info in combination declare exactly what colour your computer should display for any given triplet of numbers. We call this a colour space.
There are various standard sets of primaries, the most famous being the ITU-R Rec.709 primaries used in sRGB, first defined in 1993, often just called the sRGB primaries - this is a fairly restricted colour space, intended to be an easy target for monitor manufacturers and to achieve some degree of colour consistency on the web (lol).
Since then, a much wider gamut called Rec.2020 has recently been defined for 'HDR' video. This is a very wide gamut, and no existing displays can actually show it in full. Besides that, there are various other colour spaces such as AdobeRGB and P3, which are used in art and design and video editing.
What you see above is something called a 'chromaticity diagram'. the coordinate system is CIE xyY with fixed Y. The curved upper edge to the shape is the line of monochromatic colours (colours created by a single frequency of light); everything other colour must be created by combining multiple frequencies of light. (Note that the colours inside the shape are not the actual colours of those points in CIE XY, they're mapped into sRGB.)
In this case, the red, green and blue dots are the primaries of my display. Since they are outside the green triangle marked sRGB, it qualifies as a 'wide gamut' display which can display more vivid colours.
Sidebar: you might ask why we didn't define the widest possible gamut we could think of at the start of all this. Well, besides consistency, the problem is that you only have so many bits per channel. For a given bit depth (e.g. 8 bits per channel per pixel), you have a finite number of possible colours you can display. Any colours in between get snapped to the nearest rung of the ladder. The upshot is that if you use a higher gamut, you need to increase the bit depth in order to avoid ugly colour banding, which means your images take up more space and take more time to process. But this is why HDR videos in Rec.2020 should always be using at least 10 bits per colour channel.
in order to display consistent colours between different computers, you need a profile of how your monitor displays colour. Yhis is something that has to be measured empirically, because even two monitors of the same model will be slightly different. You get this information by essentially taking a little gadget which has a lens and a sensitive, factory-calibrated colour meter, and holding it against your screen, then making the screen display various colours to measure what light actually comes out of it. This information is packed into a file called an ICC profile.
(Above is the one I got, the Spyder X2. I didn't put a lot of thought into this, and unfortunately it turns out that the Spyder X2 is not yet supported by programs like DisplayCal. The Spyder software did a pretty good job though.)
Wonderfully, if you have two different ICC profiles, and you want to display the same colour in each space, you can do some maths to map one into the other. So, to make sure that a picture created on one computer looks the same on another computer, you need two things: the colour space (ICC profile) of the image and the colour space (ICC profile) of the screen.
Now different operating systems handle colour differently, but basically for all three major operating systems there is somewhere you can set 'here is the icc profile for this screen'. You might think that's the whole battle: calibrate screen, get ICC profile, you're done! Welcome to the world of consistent colour.
Unfortunately we're not done.
the devil in the details
The problem is the way applications tell the operating system about colour is... spotty, inconsistent, unreliable. Applications can either present their colours in a standard space called sRGB, and let the OS handle the rest - or they can bypass that entirely and just send their numbers straight to the monitor without regard for what space it's in.
Then we have some applications that are 'colour managed', meaning you can tell the application about an ICC profile (or some other colour space representation), and it will handle converting colours into that space. This allows applications to deal with wider colour gamuts than sRGB/Rec.709, which is very restricted, without sacrificing consistency between different screens.
So to sum up, we have three types of program:
programs which only speak sRGB and let the OS correct the colours
programs which aren't colour aware and talk straight to the monitor without any correction (usually games)
programs which do colour correction themselves and talk straight to the monitor.
That last category is the fiddly one. It's a domain that typically includes art programs, video editors and web browsers. Some of them will read your ICC profile from the operating system, some have to be explicitly told which one to use.
Historically, most monitors besides the very high end were designed to support sRGB colours and not much more. However, recently it's become easier to get your hands on a wide gamut screen. This is theoretically great because it means we can use more vivid colours, but... as always the devil is in the details. What we want is that sRGB colours stay the same, but we have the option to reach for the wider gamut deliberately.
Conversely, when converting between colour spaces, you have to make a decision of what to do with colours that are 'out of gamut' - colours that one space can represent and another space can't. There's no 'correct' way to do this, but there are four standard approaches, which make different tradeoffs of what is preserved and what is sacrificed. So if you look at an image defined in a wide colour space such as Rec.2020, you need to use one of these to put it into your screen's colour space. This is handled automatically in colour managed applications, but it's good to understand what's going on!
(*You may notice a difference in games even if they're not colour managed. This is because one of the things the calibration does is update the 'gamma table' on your graphics card, which maps from numeric colour values to brightness. Since the human eye is more sensitive to differences between dark colours, this uses a nonlinear function - a power law whose exponent is called gamma. That nonlinear function also differs between screens, and your graphics card can be adjusted to compensate and make sure everyone stays on the standard gamma 2.2. Many games offer you a slider to adjust the gamma, as a stopgap measure to deal with the fact that your computer's screen probably isn't calibrated.)
For what follows, any time you need the ICC profile, Windows users should look in C:\Windows\System32\spool\drivers\color. MacOS and Linux users, see this page for places it might be. Some applications can automatically detect the OS's ICC profile, but if not, that's where you should look.
on the web
Theoretically, on the web, colours are supposed to be specified in sRGB if not specified otherwise. But when you put an image on the web, you can include an ICC profile along with it to say exactly what colours to use. Both Firefox and Chrome are colour-managed browsers, and able to read your ICC profile right from the operating system. So an image with a profile should be handled correctly in both (with certain caveats in Chrome).
However, Firefox by default for some reason doesn't do any correction on any colours that don't have a profile, instead passing them through without correction. This can be fixed by changing a setting in about:config: gfx.color_management.mode. If you set this to 1 instead of the default 2, Firefox will assume colours are in sRGB unless it's told otherwise, and correct them.
Here is a great test page to see if your browser is handling colour correctly.
Chrome has fewer options to configure. by default it's almost correctly colour-managed but not quite. So just set the ICC on your OS and you're as good as it's gonna get. The same applies to Electron apps, such as Discord.
To embed a colour profile in an image, hopefully your art program has the ability to do this when saving, but if not, you can use ImageMagick on the command line (see below). Some websites will strip metadata including ICC profile - Tumblr, fortunately, does not.
For the rest of this post I'm going to talk about how to set up colour management in certain programs I use regularly (Krita, Blender, mpv, and games).