The Raspberry Pi is a tiny computer. Tiny Core Linux is a tiny operating system designed to offer the bare minimum you need to get started while taking up as little disk space as possible. Seem like a match made in heaven? The folks behind Tiny Core thought so too… this year they launched a version of their operating system called PiCore which is designed to run on the Raspberry Pi.
This week PiCore 5.3 was released, featuring an updated kernel, recent Raspberry Pi firmware, and other improvements. The disk image is only 24.7MB.
If you’re wondering what you can do with a modern operating system that weighs in at under 25MB, the answer is not all that much… until you start customizing it.
While there’s a graphical user interface and a desktop environment included in the disk image, you don’t get a web browser or even a file manager. You’ll need to install your own.
The idea is to provide you with the bare essentials needed to create your own experience. There’s a terminal emulator and a package manager which you can use to start making your PiCore system your own.
Tiny Core is designed to run entirely in RAM, which means that you don’t install it so much as load the disk image onto an SD card. When you’re running the operating system, everything is loaded into the computer’s memory.
PiCore isn’t the most powerful operating system available for the Raspberry Pi. But it’s certainly one of the smallest modern operating systems to feature a graphical user interface.
“one of the smallest _MODERN_ operating systems to feature a graphical user interface.” Because, you know, MacOS, TOS, Amiga OS, GEM, RISC OS, … all under 25 MB 🙂
Fair enough… Although I d say “one of…” 🙂
Those don’t have any of the really piggy new subsystems that any ‘modern’ OS must have.
1. Unicode and other character set support as one part of…
2. Full Internationalization support.
3. A lot of smaller but expected to be ‘always available’ libraries that just were not thought of when the 1980’s platforms were released. Everything from SSL, certificate chains and a dozen or so other cryptographic systems to GL libraries. Multiple filesystems and the filesystem checkers for them. And so on.
And of course we have to support the sloppy programming that gives us panel clock apps with multi-megabyte RSS.
It’s a modern OS, and I agree, it’s quite impressive to fit all that into 25 MB. But remember, we had graphical OSes running from a 800k floppy disk, using 38 kbytes of RAM. Those didn’t have net access, so they had to fit everything onto that floppy, had to run on minimal amount of memory, so applications had some RAM too, and they all managed to do this on a primitive few MHz CPU. Written in Assembly, optimized to the limits, squeezing out every bit of that hardware, I think that’s impressive as well. We got really sloppy since then.
Perfect. I used to feel always that we are abusing the hardware and processing capabilities. We are not utilizing the full strength. We should have made the full use of them and made the machines fly .. 🙂 I was able to use photo shop to do a lot of image manipulations in my 16MB RAM Windows 95 Pentium MMX 166MHz PC. After 15 years, with 3000 or 4000 times RAM size and several multiples of clock speed, I am not able to do much better ?? I agree a lot had changed and improved. Still !!!
Comments are closed.