Recent searches
Search options
I'm working on a talk on DRAM compression. Some of the earliest consumer products in this area were SoftRAM for Windows 95 (a fake which didn't compress at all!) and Connectix RAM Doubler for classic MacOS (which did actually work).
Did anyone ever try to reverse engineer RAM Doubler (and write about it) – or is there perhaps source code for RAM Doubler available somewhere?
@me_ The main trick behind RAM Doubler on the Mac wasn’t so much the compression but working around the classic MacOS’s braindead memory allocation scheme where programs had fixed memory allocations which locked up memory even if the applications didn’t need it all.
You had to allocate 4 MB RAM to MS Word for that one time you embed a large image, even though 99% of the time it only needed 900 KB of RAM, and now you can’t open something else at the same time using those 3 MB free
@kalleboo @me_ What always made me wonder (not having programmed System 6/7 back in the days): There _were_ applications that handled memory allocation dynamically (e.g. some versions of Graphicconverter), and some text editors where I forgot the names, where it was _not_ necessary to adjust the memory allocation for an program in the Finder manually.
This dynamic memory API/functionality capability seemed to be added to System 7 back then. It always occured to me that RAM Doubler was a cure for programs that didn’t use newer APIs but the “old” pre System7 way of allocating memory fixed through settings in the Finder.
But, this is obsevation is deduction and guessing. Would be great to read here from someone with actual knowledge of the memory manager of System 7.
Perhaps @Cdespinosa can help or reach out?