Comment by zahlman
6 days ago
> I like the Raspberry Pi RP2040 a lot. It's relatively cheap (around $1 USD) and has tons of on-board RAM - 264 KB in fact! It also has what is called Programmable IO, or PIO.
I wonder how benchmarks would compare between the RP2040 and, say, a Z80.
It would destroy the Z80. It's a 32bit, dual core CPU running at 133MHz. Even single cored it'll thrash a Z80. Heck, I bet you could create a drop-in replacement board for the Z80 using an RP2040.
Note it was possible to use a Z80 to function as a display controller, people used to do it back in the day...
https://archive.org/details/Cheap_Video_Cookbook_Don_Lancast...
The Galaksija computer used it's Z80 to help generate the video signal. I'm not sure how its implementation compares to your link.
https://en.wikipedia.org/wiki/Galaksija_(computer)
https://media.ccc.de/v/29c3-5178-en-the_ultimate_galaksija_t...
The Z80 was a very powerful CPU for the day. Its still used here and there.
2 replies →
Crazy what you can buy nowadays like the Teensy 4.0 with 600MHz base clock
Granted that's $20 not $1
My sweet spot of choice between power and price is the ESP32 S3 (2x core @ 240mhz) at ~$6 per board, but yeah, the power to dollar ratio is crazy these days, across the board. And they are absolutely tiny and sip power if you write the code well.
The key here is the "PIO" which you won't find on a Teensy. It lets you do extreme "bit banging" tricks including generating video. People have even implemented Ethernet on it. I've used it for some custom serial protocols ("Weigand") used by alarm panels.
3 replies →
Yes, I understand that, but I wonder about the multiple (obviously there is more to it than clock speed). I chose the Z80 because of its long-standing reputation.
I know, I 'm just saying the RP would dominate the Z80 in literally every way. You could design a Z80 replacement with the RP as the CPU and it'd blow away everything.
The RP2040 is a Cortex-M0, which is about the smallest core you find on modern systems but still a pipelined 32 bit RISC machine running in the dozens of MHz.
Note though, that the article is really about the PIO device on these SOCs', which isn't part of the main CPU at all. It's sort of a very limited programmable hardware engine for the specific task of doing PCB level interconnect using GPIO and lightly buffered streaming. In some sense it's like a thematic midpoint between an FPGA and a CPU.
It's... honestly it's just really weird. And IMHO has really, really, REALLY limited application. It's for people who would otherwise be tempted to bitbang an I2C or UART, but not for ones who can put hardware on the board themselves, or who have a FPGA handy, or even for people who want to do non-trivial stuff like QSPI displays[1] or whatnot.
Basically PIO smells like a wart to me. I genuinely don't know who wants it. Regular hackers aren't sophisticated enough to use it productively and the snobby nerds have better options.
[1] The linked article appears to be doing a quarter-VGA display in 3-bit/8-color, and is sort of right at the limit of the power of the engine.
what are you blathering on about, sir?
driving complex displays with no spu use: https://dmitry.gr/?r=06.%20Thoughts&proj=09.ComplexPioMachin... (my work)
pretending to be memory stick and sd card at dozens of mhz as a slave to a sync bus (my work)
ethernet: https://github.com/kingyoPiyo/Pico-10BASE-T (not my work)
68k bus slave (my work)
usb host https://github.com/sekigon-gonnoc/Pico-PIO-USB (not my work)
all on a $1 chip
> what are you blathering on about, sir?
Please don't.
I mean, I applaud your work. But let's also be honest (in the "tough love" sense): those are all toys with significant limitations that preclude anyone shipping any of them on an actual device to an actual consumer. I mean, your SOC (maybe a $2 one) surely already includes a SPI master and USB host!
Actual interconnects that solve real market problems have big boring spec books and competing implementations and silicon vendors. The application for PIO is basically limited to "I have to connect to this crazy old junk and no one makes the part I'd otherwise need".
10 replies →
> The linked article appears to be doing a quarter-VGA display in 3-bit/8-color, and is sort of right at the limit of the power of the engine.
The resolution and color depth restrictions were the product of the low data rate of USB FS (~12 Mbps), not inherent limitations of PIO.
> It's... honestly it's just really weird. And IMHO has really, really, REALLY limited application.
I'd agree with "weird". But it's useful weird; it turns out that there are a lot of situations where PIO can avoid the need for an application-specific peripheral, and can provide that function in a more flexible fashion than a fixed-function peripheral could. Dmitry's SDIO device emulator is a great example - almost every other SDIO peripheral on the market is host-only.
> it turns out that there are a lot of situations where PIO can avoid the need for an application-specific peripheral
And I can only repeat: I think that's an aspirational delusion. I'm not aware of anyone shipping a PIO solution to anyone in volume. It's "useful weird" to Hackerspace nerds like us, and that leads to some epistemological skew.
Hardware needs to be boring and reliably supported (by people you can sue!) or else no one will bet a 10k unit PCB run on it. This is anything but.
2 replies →