wip: ripple control stability #1
Conversation
| GAL16V8 | ||
| Counter | ||
|
|
||
| Clock NC NC NC NC NC NC NC R GND |
There was a problem hiding this comment.
It doesn't look like R is used on this chip, is that right?
…tbox/cuddly-robot into fix/ripple-control-stability
| {"installs":{"galette 0.3.0 (git+https://github.com/rustbox/galette.git?rev=2ba00658608c4e4d1fd51d903a7799dcabe2f9cc#2ba00658608c4e4d1fd51d903a7799dcabe2f9cc)":{"version_req":null,"bins":["galette"],"features":[],"all_features":false,"no_default_features":false,"profile":"release","target":"x86_64-unknown-linux-gnu","rustc":"rustc 1.62.1 (e092d0b6b 2022-07-16)\nbinary: rustc\ncommit-hash: e092d0b6b43f2de967af0887873151bb1c0b18d3\ncommit-date: 2022-07-16\nhost: x86_64-unknown-linux-gnu\nrelease: 1.62.1\nLLVM version: 14.0.5\n"}}} |
There was a problem hiding this comment.
You gotta run rustup update my man: rustc from July? And/or, we should drop a .rust-toolchain.toml file in this repo too.
| @@ -3,3 +3,5 @@ | |||
| setup: install [minipro](https://gitlab.com/DavidGriffith/minipro) | |||
|
|
|||
| flash me with `make generate && minipro -p ATF22V10C -w GAL22V10_fifo_buf.jed` | |||
|
|
|||
| https://www.proto-advantage.com/store/product_info.php?products_id=2200264 | |||
There was a problem hiding this comment.
I like the idea of this repo being notes & such about different hardware-y things, maybe we make a NOTES.md for that stuff? And/or, drop a touch of context in here for this?
| Clock NC NC NC NC NC NC NC NC NC Reset GND | ||
| /OE Q0 Q1 Q2 Q3 Q4 Q5 Q6 Q7 RO NC VCC | ||
|
|
||
| AR = Reset |
There was a problem hiding this comment.
whoa, does this asynchronously reset the counter to 0 whenever Reset goes high? That's a 22v10-only feature, I take it?
| ; 0 1 1 | ||
| ; 1 0 x | ||
| ; 1 x 0 | ||
| Q2.R = /Q2*Q1*Q0 + | ||
| Q2 * /Q1 + | ||
| Q2 * /Q0 |
There was a problem hiding this comment.
Love this! In order to check my understanding, I want to build a logical equivalence to what I did in the 6-bit horizontal counter. Here's my attempt: in order for the later bits to "matter," the higher bits have to be 1. So, the Σ Q_N * Q_N-1 .. * /Q_M strategy ends up with at most one 1 going into the sum term, whereas this will OR together a lot of ones when e.g. Q = 0b100. But 0+1+1 is the same result as 0+1+0 in the boolean ring (field?), so they're equivalent functions.
I like this much better!
| @@ -0,0 +1,136 @@ | |||
| GAL22V10 | |||
| SIPO4to8 | |||
There was a problem hiding this comment.
| SIPO4to8 | |
| 10-bit counter with settable high bit |
Not for merging yet, but check this out:
Not bad for a ripple counter on a breadboard!