Wednesday, July 1, 2020

SDRAM and Sidbox Woahs!

While the Sidbox 5 hardware is some seriously suped up stuff! I am finding quirks and a love hate relationship… so far.. more LOVE than hate, but the hate part, is spawned from having to deal with much higher clock speeds,.. this also means signal induction and ringing on the traces, so i have to worry about those now too.

FOR NOW. the SDRAM we plan to use, 8Meg of, took me a week, but finally got it wired up. after severl attemps

The RAM is working! well the proof of concept works! i had to drop to 8bit data lines just to get things working, but the wire as you can see

looks more rats nest than anything useful… actually suprised this is working! due to its very very bad electrical setup! the wires are meant to be as close to the CPU as possible, but i’ve clocked the SDRAM at 16Mhz for now.

the peek and poke functions work fine, but its an odd memory view everything looks offset!

after turning off Burst Read, and Write and RAM CAS and RAS clocks to 3, things dropped into place, but the slower the ram is being accessed the better chance data is being found.

SID Tunes appear to be running from the RAM and as a test, we got a bit of music playing, pulled the ram out, and the song just broke!
THIS Setup appears to be compatible with another ram too, SO our choices of ram is much better for now!

eventually once I have the PCB done, I can open all 16bit data lines allowing 16bit full access instead of 8, while this CPU is MIGHTY fast.. while the rats nest is happening, everything needs to be slow.

SO One happy result after 5 days of research and hard work!