_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

Good call, BDD. I had checked continuity but not voltage. It turns out Vcc was quite low in that area of the board; now it isn't and the DS1813 is working again!

_________________
"The key is not to let the hardware sense any fear." - Radical Brad

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

Has somewhat evolved in the world of tech support: "Are you sure it's plugged in?" Being the first thing I would ask when I worked at Dell tech support in the late 90s.

I found asking that tended to get the 'of course it is' response, even if it wasn't... on the other hand, asking them to remove the mains cable (at both ends) and then replace it would often result in an embarrassed 'oh, it's working now...'

Neil

Is it plugged in? Is it turned on? Have you tried turning it off and on again ...

https://www.youtube.com/watch?v=5UT8RkSmN4k

-Gordon

_________________
--
Gordon Henderson.
See my Ruby 6502 and 65816 SBC projects here: https://projects.drogon.net/ruby/

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

Ben Eater only uses a 0.1uF cap I believe. He has said that he needs to redo that reset circuit, but he's only got a million-something views on youtube of his board doing cool things. Ya know, little things The point is that you shouldn't need super specific values, though I do take everything Garth says as gold. Glad it was fixed with BDD's suggestion though.



Small victories are important. They keep you going. Though, having it work the first time is more than just a 'small' victory. Great job!



Good point. That would tell you a LOT more than just a probe or scope.

Good looking board by the way. It's interesting how you actually DON'T have a jumble of wires as usual with breadboards. It's a lot of wires, but not jumbled. They seem organized and honestly because they are so thin you can still see everything going on. You have used the strengths of the breadboard and wire-wrap to create a good design!

Thanks for the update!

Chad

When I put my 6507 board together I too used a 0.1µF capacitor - figuring that if it was good enough for 30 million Atari 2600s it was good enough for me.

And it was, although not 100% reliable and I sometimes needed to push the reset button ... that is until I plugged in a modern 65C22 - then it became 100% reliable from power on.

Weird little things.

-Gordon

_________________
--
Gordon Henderson.
See my Ruby 6502 and 65816 SBC projects here: https://projects.drogon.net/ruby/

Thanks Chad! When I started, I had in mind that the construction method would just be a kind of interesting novelty. But now I really like it, both for the aesthetic reasons you just mentioned, and also for the speed / ease of construction.

In other news, I did some basic testing of my ACIAs. The "Rockwell" ones all seem fine, except that two of them lost pins when I pulled them back out of the breadboard. The "CMD" ones are at least actual 6551s; 4 of them work, 1 of them spews garbage as soon as it is enabled. I don't really trust them because they look so wonky, so I will probably stick with one of the Rockwell ones.

When I say "testing," what I mean is that I was able to set up a 3-wire serial connection with my old Windows XP laptop and echo back what I type from its terminal emulator. I haven't done anything to check the current draw to see whether or not they're actually CMOS. I may do that after lunch!

Edit: I did that. The Rockwell ones seem to draw around 2mA. The "CMD" ones draw 1mA, but don't work at all when their power supply goes through my DMM!

_________________
"The key is not to let the hardware sense any fear." - Radical Brad

I've continued to tinker with this project a little bit - tidy up the wiring and so on.

One of the many containers in my workshop is a flat Sterilite box that I got a Publix a while back. At some point I noticed that my breadboard assembly fits really nicely into it, and I got the crazy idea of turning it into an enclosure for an honest-to-goodness breadboard retrocomputer. So... I did that. As Mary Berry would say, "It does look home-made." The keyword here is "semipermanent." Everything is adhesive mounting foam, gorilla glue, and heavy card, rather than the ABS plastic, acrylic, and steel I have on the back burner for sturdier projects. Still, it's nice to have a power switch, reset button, and serial port that are away from the breadboards and able to take a little more pressure.

_________________
"The key is not to let the hardware sense any fear." - Radical Brad

So the obvious question is: is it bigger than a breadbox?

(those of a certain age and locality might remember that...)

Neil

_________________
"The key is not to let the hardware sense any fear." - Radical Brad

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

The breadbox question came to me(*) rather late in the game, with the little gadget 20Q. It turns out the game Twenty Questions, which I have played (here in the UK), is from the USA, late in the 1800s, and by the time the breadbox question was popularised in comedy (mid 1900s) must have been old hat. Different cultures will of course have converged on different standard tactics in the game.

I did find it surprising that the 20Q gadget's training comes from the web game, which in turn comes from a version circulating on floppy disks. The thing being that the gadget needs a database which is trained on lots of human challenges - the gadget cannot learn, but the web game, and the floppy game, could.

For me, at the time of coming across the gadget, the web was very new indeed, and I would have supposed a different ordering.

As it turns out, the gadget wasn't so American in its outlook that it didn't impress here in the UK. A breadbox here is presumably something like a breadbox there. (Edit: except, I now realise, we say breadbin here.)

I'm still intrigued as to how much data is in the gadget, and how it is made use of (and by what kind of CPU, of course.) I gather it's some kind of machine learning or neural net - it can cope with inconsistent answers from the user. A simple decision tree won't.

Also of interest, some million valid answers can be reached with twenty questions, but the gadget is apparently primed with just ten thousand. Still, as it turns out, enough to be entertaining.

To come very slightly more towards on-topic, there was a Basic guessing game called, I think, ANIMAL which self-modified, learning with play ever more examples of animals and their distinguishing characteristics. While ANIMAL can run on a microcomputer, it might be that 20Q requires just a little too much data. Or maybe not, maybe it's a good challenge to implement it on a 6502, or '816.

(*) Edit: Oops, I've made a huge mistake. In the case of the 20Q device here in the UK (and it might have been localised) the size question wasn't about a breadbox at all - but a loaf of bread.