The Other Easy Option
Jun. 8th, 2025 08:45 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
krpalmerOne motivation in the mix that got me installing Linux on another of my old computers was to take one more crack at getting a particular emulator running. To begin with, I had compiled “Virtual T” on macOS. As I poked away at it, though, I found its emulation of the floppy disk drive that could be interfaced with the TRS-80 Model 100 didn’t quite work. This wasn’t the only way to get programs and files into and out of the emulated portable. However, after I’d found the disk drive did work with the Windows version of the emulator (running via Wine, but an earlier version lacking a few features), curiosity had me trying to see just what the situation was with Linux.
I managed to follow some online instructions to the point of compiling Virtual T on the “virtualized” system I’d been trying out Ubuntu MATE with. Its disk drive worked. I then got the emulator working on the old iMac that had wound up running only Linux Mint, regardless of this recapitulating an already-grasped peculiarity of a portable computer emulated on a desktop machine. One thing I’d done differently that second time, though, was to install a certain software toolkit called FLTK vital for Virtual T using “Linuxbrew,” which I thought would give me a better handle on keeping it updated going forward or even removing it in some future case. When FLTK upgraded, though, the emulator stopped working.
Around this time I did have my actual Model 100 repaired and upgraded with a hardware add-on the latest revision of Virtual T was supposed to support but which I hadn’t been able to get working there. Again, however, I suppose I did want to at least try and take on the problem. After considering the way I’d tried to apply the compilation instructions I’d used before to that different installation of FLTK I could see how the link to the toolkit had broken. My first thought was to just see if the compiler knew where to look for it when installed via “Linuxbrew” without the too-specific prompt I’d used before.
I folded a certain number of code revisions from a specific “fork” someone had prepared for macOS into the master files, then ran into a compiler error I’d remembered from before that required a certain small “include” revision made in one particular file to be added to another one. (I just hadn’t remembered what file to correct.) At that point, though, the “linking” stage started returning errors; it didn’t seem able to find what it was demanding. As I wondered if the necessary components had been installed alongside FLTK by “Linuxbrew,” I did a bit of searching, and happened on a version of Virtual T I’d already known about, said to have been targeted for a sort of more modern and miniature Model 100-like terminal. The instructions provided there (involving a more fundamental Linux installer) caught my eye, and I tried them just to see what would happen. More or less to my surprise, it worked.
With that taken care of I did start wondering about the code that had been provided with the instructions and whether it would only compile on the unusual hardware it was targeted for. As it turned out, it did compile on my own computer. Although it didn’t offer the most modern options of the latest master files, I thought to time a very basic BASIC program against my actual Model 100 and realised that, in being a little more poky, it seemed to better match the speed of the real machine. All of that switching back and forth, though, seems to have competed over a single preference file and knocked out a modern option I’d intended to exercise. It took me a little while to more or less sort that out.
I managed to follow some online instructions to the point of compiling Virtual T on the “virtualized” system I’d been trying out Ubuntu MATE with. Its disk drive worked. I then got the emulator working on the old iMac that had wound up running only Linux Mint, regardless of this recapitulating an already-grasped peculiarity of a portable computer emulated on a desktop machine. One thing I’d done differently that second time, though, was to install a certain software toolkit called FLTK vital for Virtual T using “Linuxbrew,” which I thought would give me a better handle on keeping it updated going forward or even removing it in some future case. When FLTK upgraded, though, the emulator stopped working.
Around this time I did have my actual Model 100 repaired and upgraded with a hardware add-on the latest revision of Virtual T was supposed to support but which I hadn’t been able to get working there. Again, however, I suppose I did want to at least try and take on the problem. After considering the way I’d tried to apply the compilation instructions I’d used before to that different installation of FLTK I could see how the link to the toolkit had broken. My first thought was to just see if the compiler knew where to look for it when installed via “Linuxbrew” without the too-specific prompt I’d used before.
I folded a certain number of code revisions from a specific “fork” someone had prepared for macOS into the master files, then ran into a compiler error I’d remembered from before that required a certain small “include” revision made in one particular file to be added to another one. (I just hadn’t remembered what file to correct.) At that point, though, the “linking” stage started returning errors; it didn’t seem able to find what it was demanding. As I wondered if the necessary components had been installed alongside FLTK by “Linuxbrew,” I did a bit of searching, and happened on a version of Virtual T I’d already known about, said to have been targeted for a sort of more modern and miniature Model 100-like terminal. The instructions provided there (involving a more fundamental Linux installer) caught my eye, and I tried them just to see what would happen. More or less to my surprise, it worked.
With that taken care of I did start wondering about the code that had been provided with the instructions and whether it would only compile on the unusual hardware it was targeted for. As it turned out, it did compile on my own computer. Although it didn’t offer the most modern options of the latest master files, I thought to time a very basic BASIC program against my actual Model 100 and realised that, in being a little more poky, it seemed to better match the speed of the real machine. All of that switching back and forth, though, seems to have competed over a single preference file and knocked out a modern option I’d intended to exercise. It took me a little while to more or less sort that out.