CompuServe Thread

I had failed in attempts to increase either my fast RAM or chip RAM beyond a total greater than 3M. The problem seemed to be in my accelerator board. Please see "Stuck at 3M?" for details. But I tried one other experiment. VXL-30 allows mapping of its piggy-backed fast RAM to $800000 as an option (either by jumper or by software), in case another memory board refuses to give up the space at $200000. I had tried that with the AdRAM board (which is designed not present that kind of problem anyway), to no avail (but the computer had been well warmed up by the time I tried that). I decided to see what would happen if I set the RAM high at a really cold boot, with no other memory boards installed. Success! (To be followed soon by failure.) Showconfig showed 2M of RAM at $800000. I went into AutoPilot and composed a forum message. Clicked on "Save." Hard disk spun a bit, but nothing happened on screen. Tried saving the message to Clipboard, then got out of the forum message area by clicking on "Cancel." ("Cancel" worked, though "Save" did not!) Returned to forum message window and copied the message back from Clipboard, but attempts to save it as a forum message met with the same results as before. I Amiga-N'ed back to WorkBench, called up Edit and saved the message from the Clipboard to disk. Back to AutoPilot, and got a Guru: 8000 0025 @ 0000 2EC0. Repeated Ctrl-A-A's got me as many Recoverable Alerts: 0100 0008 @ 000B D700, but once I accidentally hit the Workbench close gadget, and when I selected "cancel" to the requester that came up, I had my Workbench screen back, with no Alert in the way. The title bar told me I had a reasonable amount of chip RAM and "0 other memory"! So I applied a little bit of logic(?): This "0 other memory" is the very same RAM, physically, as that which is usually reported as close to 2M of RAM at boot-up. Only the software, trying to find that memory at $800000 instead of $200000, is different. What reason is there, then, to think there is something wrong with the accelerator board? Couldn't it just as well be my mother board? On the other hand, how could circuitry warming up cause a software problem? Bart Mathias