OneROM

[antus]

If its working with arcade machines from back in the day it must be able to drive and or be tollerant of 5v. look at how they do it, if its a dumper or just works out of the box. the issues are usually 5v tollerance so its not damaged, I think it is that but not confirmed. And if its levels are high enough to switch the cpu reliably, especially in auto motive. Level shifters can shift but what they do the wave form at 2mhz (?) and if its a problem is another question. I'd try and skip the level sift for initial testing if the data sheet says io is tollerant to 5.5v or above. There might also be an io voltage input already there you can feed 5v in to, to switch 5v directly.

[jxx]

Some of the io pins are tolreant of 5V and they're all in use on the eprom pins (FT).
The 2 that are being used (pins 40 & 41, non-FT) which were originally used for bank selection jumpers are only 3.3v max i believe, i'll trawl over the data sheet to confirm
I've decided to knock up a board based on the old max232 circuit and some 74 logic, propblem still remains that the aldl data is inverted.

[antus]

Why does the memcal need to know about ALDL? wouldnt it be USB direct to PC, or via the ECM CPU UART for ALDL?

[jxx]

The latest alpha firmware has incorporated the datastream in, it now does both with virtual usb ports, 1 for programming the eprom and 1 for the datastream over 1 usb cable.
So now apart from the memcal eprom area, need to connect the onerom to aldl as well.

[antus]

if you do that, then it might be better just to use USB direct. I'm not sure the ALDL part gives you anything if the goal is PC communication.

you could use aldl on the logging side to the PCM, and the emulated moats tool on the emulating side in parallel.

[jxx]

Just a convenience really, feature is there, why not try to use it.
And I'm curious to see if it'll work and how much needs to be done to the hardware to make it work.

[vlad01]

100%, I know I'd want to use ALDL to the full extent. I use the RT features a lot, especially when dialing in warm up transients which requires working fast and being able to update changes real time because there are many time sensitive interdependencies when doing that part of the tuning.

[antus]

if you went car over ALDL to PC you'd have the logging, and if you went PC to onerom over usb you'd have realtime (if the firmware and hardware supports it - and it sounds like they use chip ram to do that, since I think the flash requires a large flash / erase cycle to persist the data). I guess that is two cables. But the onerom doesnt have RW like ram does as it is only emulating ram, so there will be more development to do for the ECM CPU to be able to single a write to the onerom, which means doing all of it over ALDL wont work without a hardware RW pin support.

[jxx]

Not saying to do it all over ALDL, just using that to get the live data, yes the same as having 2 cables but using only 1 cable and 2 com ports on the onerom.
I know we can't see the code at how it's done but simply passing thru the ALDL data via 2 pins to a virtual com port can't be to cpu intensive, it's working on the honda ecu.
I think ya boiz garage mentioned somewhere in one of their videos it takes about 8 seconds from changing and uploading values on the honda ecu before it becomes live and running (i could be wrong, i'm not going back over all the vidoes)

Yes this is going to take a bit more work and some circuitry to mod an 808(and possibly any other ECU/PCM using ALDL), it's not needed but for the s&g's and to see if i can get it working i'm up for a challenge.
I did email ya boiz garage/ALPHA asking for access to the source code (i'm sure many here would contribute to it) but have not recieved a reply, it would be a simple change for the code to invert the ALDL signal instead of needing to add extra circuitry to do it, but need to work with what i have.


As i can't find the 808 i modded 14 years ago for datalogging i'll have to mod another one so i've hacked together the diy inteface and added what i hope will work for interfacing the onerom to it
I know it doesn't need all the extra circuitry, for prototyping and the fact i have most of these parts as tht components and much easier to work on to change if needed the prototype is going to be over complicated.
Pretty sure what needs to be done can be compacted down to about 6 smd components and 1 smd 7404 ic, about 1/5th the size of this board.

ALDL_Onerom_interface.png

You can just drop the onerom in with the firmware and it'll function as the moates emu just like before.

And this is where the rabbit hole begins.... onerom isn't totally relying on the cpu to do its magic.
Yes the initial image is stored in flash, then Piers voodoo code takes over once it's loaded into the 2350, using PIO and DMA instead of relying only on the cpu.

Programmed I/O (PIO), which lets you configure tiny state machines and other facilities (shift registers, scratch registers, FIFO buffers, etc) to process simple I/O logic, freeing up the CPU to do other tasks

I guess in a way it's kind of like an fpga in this use case.... only not quite.


Here's how it's doing it explained better than i could attempt to.
https://www.youtube.com/watch?v=Y8RODQZM2HY
https://hackaday.com/2025/11/30/a-deep- ... he-rp2350/

[delcowizzid]

Got mine today thanks mate