Some people use vintage computers now an then!
Many vintage chips like v9958 can be found on ebay. There is one seller who makes bundles with cpu, ram, sound, and video chips for a reasonable price.

Some of the v9958 functions auto increment the vram address in order to speed up block transfers by removing the need to specify the address each time. Very handy and something you really want to have working for fast games.

Check out the CreatiVision console for an example of an early 6502/ti9918 system (precursor to the v99x8 chips).

While it's true problems can arise in some circumstances, I'd say your warning is needless broad, BDD -- and potentially anti-productive. Use of R-M-W instructions on hardware registers can be a real boon, as I can attest from experience. Garth is another who "goes there" routinely.

Yes, that's one solution to Steve's problem, but it's a tradeoff because it abandons the original size/performance goal. Giving up is a viable option, but it isn't the only option! As long as we're willing to use our gray matter (sorry, Steve!) I expect the problem is one that can be gainfully solved.

_________________
In 1988 my 65C02 got six new registers and 44 new full-speed instructions!
https://laughtonelectronics.com/Arcana/ ... mmary.html

LOL, and has been solved in this case (see CreatiVision schematic). Writing and reading use different addresses, which I didn't understand at first, but now see why.

http://www.madrigaldesign.it/creativemu/datasheets.php

thanks, I will take a look at your page. I hope to post some more progress here soon now that I've got almost all the parts I needed for progress.

eddie

First prototype of Video Adapter PCB. Will see how many mistakes when it comes back from JLPCB.

Sorry I didn't see this I could have perhaps saved you some time. 65C02 + 9958 was demonstrated last year and is the heart of a system called the Steckschwein.

https://steckschwein.de

You can get a schematic of how it was interfaced here:

https://steckschwein.de/hardware/v9958- ... ousel-1193

You could reach out to the designer if you need tips on software and/or timing issues.

thanks, I've come across this web site recently and I've compared my design with their.

Quick update: PCB is back from JLCPCB. I've assembled the first and will begin testing.

Making progress on the V9958 video adapter. Got a picture, but it not clear at all. Will need to debug the video circuit to see what's going on.

got a better picture after some adjustments on the video encoder board. There's still issue with the sync at the top of the picture.

Making more progress with "Hello World" in text mode. But looks like there timing issue with some strange chars appearing.

80 Columns mode now showing a picture. the screen corruption in the previous post was due to the long ribbon cable connecting the 6502 SBC to the Video Adapter. It now uses a much shorter cable and it appears to fix the issue.



The next task is to get a more stable picture. The picture will sometimes get out of sync and the pixels is not too stable on the screen. My current thinking is that the Voltage is not good on the video signal driving circuit. The whole computer is powered by a USB connection (~4.94V on the input side and drawing 465mA). However the voltage on the video output circuit has dropped to 4.51V. Will need to workout how to fix this.

Use a separate power supply if you can, or double up on 2 USB outlets from your PC. USB(1,2) is not designed for high power and if the PC was doing it properly would only allow 100mA before shutting down the power to the USB socket. (there is a power negotiation protocol - rarely used on most PCs) Clearly your PC, like 99.5% of all PCs and Laptops haven't been fitted with current limiting devices, but it may have a 500mA polyfuse (self resetting thermal fuse) or similar.

If doubling up, it might be worthwhile taking the lid off your PC to make sure there is a separate fuse for each outlet - sometimes there is just one for a bank of 4 USB sockets. Polyfuses are usually green, or easy to tell apart from the usual SMT resistors.

I ran into exactly this issue trying to power the backlight a 20x4 LCD board - it dropped the USB supply voltage down to a level that made the entire board unstable. I fitted a separate 5v PSU for the display and all was fine again.

Cheers,

-Gordon

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

500mA is considered the outer limit for power draw from USB. There are much better power sources available.

BTW, the picture is a big improvement over what you had before.

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

Agreed on the picture quality getting better than previous. Seems noise/ringing on the ribbon cable and possibly noise on the power to be a cause as well. Even though older USB 2.0 spec had a 500ma power limit, several motherboard manufacturers have implemented high power USB sockets for some connectors.

What's more interesting is the newer USB specs and the higher power limits as well as the higher transfer speeds. I found this article recently from CUI... fairly interesting and informative:

https://www.cuidevices.com/blog/usb-typ ... -clarified

On a different topic... USB cables. I've received some real cheap ones where the wiring is so thin a gauge, that it barely handles a switch transient without the voltage taking a dive. So, as another possible culprit, poorly made USB cables can also result in poor power to an end-user device.

_________________
Regards, KM
https://github.com/floobydust