3 ms·

Ah, this brought a lot of memories back. I remember meeting the sysop of "Alternate Reality" because I needed an ISA IDE interface board, and he had one "somewh

Ah, this brought a lot of memories back. I remember meeting the sysop of "Alternate Reality" because I needed an ISA IDE interface board, and he had one "somewhere" he said. I was a teen, 15 or 16, and when I went to his house I met a slovenly man in his 20's or 30's ("old" to then me) and he had the goods, and I paid him. Later on I got into a 'l33t' BBS through a guy at work and I got the first 28.8k modem of anybody I knew, from him. In retrospect, I'm pretty sure it fell off a truck at some point before I got it. And then there was the time I started my own BBS. The local paper (1995ish) published a technology column every Monday, and I wrote in about my BBS. I got one call from that, from somebody who figured out that they'd mixed up the last two numbers. sigh. I never BBS'd after that, but the early days of Prodigy, AOL, and BBS's set me up for much of who I am and what I still do to this day.

sokoloff 2y ago
I spent a lot of time in the mid-80s on local BBS’s and my first paid work was helping a local sysop upgrade his software to make 1200 baud modems worth it. The BBS itself worked out of the box with 1200baud modems, but calculating the XModem block checksum in Atari BASIC was so slow that 1200 baud downloads were nowhere near 4x as fast as 300. So I ported the (very simple) algorithm to 6502 assembly and worked out the USR($) calling conventions from BASIC*, which gave the expected increase, slightly improved 300 baud transfers, and got me $20 to spend. * You had to make the 6502 code relocatable, encode it into a BASIC string, and then call USR() on that string. At the time, I could only do all of those tasks mechanically without really understanding why I had to do what I had to do. More on USR: https://www.page6.org/archive/issue_11/page_24.htm https://www.page6.org/archive/issue_11/page_24.htm

sokoloff 2y ago

I spent a lot of time in the mid-80s on local BBS’s and my first paid work was helping a local sysop upgrade his software to make 1200 baud modems worth it. The BBS itself worked out of the box with 1200baud modems, but calculating the XModem block checksum in Atari BASIC was so slow that 1200 baud downloads were nowhere near 4x as fast as 300. So I ported the (very simple) algorithm to 6502 assembly and worked out the USR($) calling conventions from BASIC*, which gave the expected increase, slightly improved 300 baud transfers, and got me $20 to spend. * You had to make the 6502 code relocatable, encode it into a BASIC string, and then call USR() on that string. At the time, I could only do all of those tasks mechanically without really understanding why I had to do what I had to do. More on USR: https://www.page6.org/archive/issue_11/page_24.htm https://www.page6.org/archive/issue_11/page_24.htm