Notices by Studio 8502 :verified: (mos_8502@soc.studio8502.ca), page 4

A very basic stackable PCB design. 99.06mm square, grounded M3 mounting holes at the corners, three 2x20 pin stackable headers for a 120-pin bus pinout, leaving the "back" side free for IDC connectors or whatever else is needed.

I figure 120 pins is enough for up to a 32-bit bus. I am increasingly liking the idea of a stackable 3.3V generic retro bus that is as inexpensive and simple to use as the #RC2014 (which I love and respect).

For lack of anything better, I've been calling it Bifrost.

This is all being kept as intentionally simple as possible. One of the reasons, I think, for the failure to launch of ÆSIR and Turaco is that they are too complex, too much going on in a single board. Makes them a bitch to route and debug, hard to get excited about.

This is also meant to be generic, processor-neutral. It could work with anything from a 6502 up to a 32-bit MIPS or RISC-V type thing.

Okay. So, since it's been on my mind, let's talk about stackable PCBs.

This is a PC104 stack. Each of these boards is the equivalent of either a PC motherboard or an expansion card. It's basically like if the backplane of a computer design were split up among the cards. It's very clever.

I'm picturing something like this, but for a 3.3V logic level implementation of a CPU architecture neutral #RC2014 type system.

@thomasfuchs Atari's shell design is better, in terms of label size, but they squandered it on a plain Jane label. Pity.

@RenewedRebecca @thomasfuchs Based on observation, LLM true believers seem to actually *be* stochastic parrots. This is not to say the same about most people, to be clear.

@RenewedRebecca @thomasfuchs Like all things in our reality, sapience is likely a spectrum, and some but by no means all or even most humans will be on the lower end of the bell curve.

The truth is, I’m afraid, that GNUstep is the very last bit of “modern” computing that holds even a little interest for me. Nothing new excites me anymore. It’s all the same shit, a little faster or a little cheaper or running a shinier, slower OS. I want to do what I can to make GNUstep better, and I will do, but the only passion I have left for computers is for ancient hardware and new stuff inspired by it.

I have reached a point where I’m happier typing away in Turbo Pascal 3 on a CP/M machine than I am using a system literally thousands of times more powerful. I’d rather play with a C64 than a PS5, and I wouldn’t trade my ZX Spectrum Next for a fully maxed out Mac Pro.

RISC-V could be interesting, if I could buy a bare CPU I could run at 3.3V and build a system around without going mental. But honestly, who needs yet another Raspberry Pi but with less software?

You outlived Kissinger. Stick around. I want you to outlive Trump, Putin, and Musk as well.

"Measuring purely by confirmed kills, the worst mass murderer ever executed by the United States was the white supremacist terrorist Timothy McVeigh. On April 19, 1995, McVeigh detonated a massive bomb […] killing 168 people, including 19 children. […] McVeigh, who in his own psychotic way thought he was saving America, never remotely killed on the scale of Kissinger, the most revered American grand strategist of the second half of the 20th century."

https://www.rollingstone.com/politics/politics-news/henry-kissinger-war-criminal-dead-1234804748/

In case someone needs it at some point, I made this table of load capacitance values for timing crystals.

You ever find yourself, quite out of nowhere, realizing that you have been very dumb about something technical for a very long time, and immediately want to either drink until you black out or leap into fixing it?

Had a dream last night that I ported V7 Unix to 65816. No idea why.

@thomasfuchs More the 15KHz output. A card that was otherwise bog standard CGA but output a signal a VGA screen could digest would be ideal.

@thomasfuchs I desperately want someone smarter than me to make a new CGA card without an expensive FPGA.

If you were teaching someone 6502 assembly programming, which would you focus on first: program and data structure, or learning the mnemonics?

My wife is revising our continually-growing holiday music playlist, and this is her latest candidate she has sent to me.

My reply: "What the shitbiscuit?"

https://www.youtube.com/watch?si=9W0KRmrPbYnqw8qU&v=BLy2kUAXa6Y&feature=youtu.be

I absolutely believe, based on knowledge of the subject matter, that *real*, sentient, artificial intelligence is not only possible, but will eventually be created.

I also do not believe for a second that current technology is anywhere *close* to the level needed to do the job, and even if it were, our understanding of cognition is far too low to actually pull it off except by accident.

This is not to say that what popular media calls "AI" is inherently bad or useless, only that it isn't what it pretends to be. It's a tool, and it has uses, but those uses are limited, as with all tools. In the proper hands, done in accordance with proper ethics, it could be a boon to creatives, but as it stands it is mostly a tool for capitalist exploitation.