#lt6502b — Public Fediverse posts

This was supposed to be a test print, but I think I’ll be keeping it. #LT6502b

Next will be the screen front bezel.

WOOT!

Sorted the bugs (intermittent framing issue when sending) and added TX_BUSY status bit.

Here you can see me checking connectivity and then connecting to a simple BBS (this is the console on my laptop).

I do think I'm getting buffer overflows on some servers (m68k one and a couple of others) but this one worked a treat!

So next I need to write a terminal emulator (which I am not looking forward to).

#LT6502b #Retrocomputing #BBS

oOO, Getting closer to a working CPLD/WiFi modem for my 6502 Laptop
got a couple of bugs that are irritating.

Here I'm connecting to a BBS using BASIC to send the ATDT command and then display what comes back from the modem.

#LT6502b
#Retrocomputing

Well a lot of work on #LT6502b today.
Now running at 10MHz (up from 7MHz), Keyboard now working and I can receive data from the WiFi Modem.
Next is making the CPLD transmit to the WiFi Modem.
Then trying to figure out why I can't run at 14MHz...

#Retrocomputing #6502

IT WORKS!!!

With a bit of effort I can now receive data from the UART on my keyboard controller, into the CPLD and read it form BASIC on my 6502 laptop revision b (slim version).

I've set the Keyboard controller to output characters from $20 to $7F, here you can see them being shown.

There is also a "new data received" status register too!

#CPLD #LT6502B

Well the CPLD UART is coming along, another big step... Now when it's done receiving a byte, it transfers it to a register, which can then be read at anytime by the CPU. So that the CPU doesn't read garbage if it tries to read mid transfer (double buffer).

more to do yet as I need two such registers and I also want the register to be cleared once it's been read, but this is BIG progress.

This is all done in WINCUPL, not VHDL/Verilog.

#CPLD #LT6502b

Well, I've not found a suitable UART speed for comms between the 20MHz AVR for keyboard and modem and the CPLD, it's 55336 baud, the AVR has a 0.4% error which gives 55556 baud, which *should* be ok with the CPLD running on a divided down 28.332MHz.
So it spits out 1 ASCII character a second to the CPLD.
Later this week I'll get the UART code into the CPLD and write a bit of basic to read it, hopefully it works.
#Retrocomputing #LT6502b #ILaughAtStandards

Ok, so, I have the next part working of my CPLD UART... it now populates a buffer, with the correct values!!
Green = serial in (LSB first),
Orange is parallel out (LSB at the bottom)

Next step is to transfer that to a register that can be read!

I will probably up my sample clock to 16x, as that seems like the common thing, and 4x may not be 100% reliable. But for simulation, 4x is enough.

#CPLD #LT6502b

Fun with WinSim.

After much battling with dodgy tools from the 90s, I've now got two useful signals. One to start the UART receive and it's counter, and one to stop when the 10th bit is received.

Tomorrow, shift register to capture the bits.

#Retrocomputing #LT6502b #CPLD

I've been trying to get my KeyScan (ATMEGA644P) to output onto the 65c02 databus at the correct time and for the correct duration...
I don't think it's going to work, sometimes I get no data back, when I tweak the timing I somehow get a 16bit value back (impressive from an 8 bit CPU eh?).

So, I may have to put a UART into the CPLD, and "WinSim" doesn't seem to simulate properly. So it'll be try and keep trying until it works, or I run out of space.

Has anyone done a UART in a CPLD?

#LT6502b

Well it completed.
No lines so I think the fix has solved the timing issue.
32 iterations, 1024x600 Mandelbrot on #LT6502b

A couple of hours in and looking good.
This is 1024x600 (almost half way) 32 iterations and so far no odd lines.

I’ll let it run in full at the weekend.

#LT6502b

Well display is looking a lot better after a small change and a mod wire, but it needs a lot more testing before I’m happy it’s fixed.
#LT6502b
#retrocomputing #CPLDfun

Working on the keyboard code for the #LT6502b today.
Some keys are missing as once this is done I have another chip to solder in and I don’t want to melt switches/caps

Graphics commands are now back!
(The dots are single colour pixels, not dust)
Currently these are 8bit, I will probably add 16bit versions of the commands later.
But next I want to solder in the keyboard parts and get that working.
#LT6502b #retrocomputing

Ok wow, that was a LOT of work, but scrolling is now working on my #LT6502b screen.
This is 1024x600 so gives me 128 columns but 37 rows (with a bit left over).
I may use the little bit for something like a status bar of some kind. I’m not sure yet.
Anyway, next is adding back in circle, line, plot, etc.
#retrocomputing

Finally got text on the new display for #LT6502b
Wow it’s complex to get going.
No scrolling working yet and a few other issues to work through.

Ah ha!!! I ordered the display with a font chip, which has an 8x16 font on.
After much work that now works fine, giving me 37 lines per screen by 128 characters.
It does have a Yen/Yuan symbol instead of a dollar, but there is a variable width font with the dollar, so I can in theory swap the Yen/Yuan for a dollar on the fly later in assembly.
The viewing angle and contrast on this panel is also fa better than the previous one, image is a bit washed out still.
#LT6502b

I got text running, but the font is 12x24, and there is no 8x16 sadly :(
So only 25 lines on this screen, but 85 characters wide.
Would be nice to get an 8x16 font on here :(
#LT6502b

#LT6502b screen is now alive, just a solid block fill for now, but it’s working. 16bit colour and 1024x600.
Text will be the next thing to get working.
I will say, wow it’s a complex screen to setup.

Codeberg, any good?
I'm thinking of moving my #LT6502 & #LT6502b projects there as I'm tired of AI rubbish.
Anyone else use them? thoughts?

Well, I can read the registers of the RA8889 10" (1024x600) display on my #LT6502b project :)

So I'll be trying to get it to start in the next few days/week.

#Retrocomputing #6502

Current state of #LT6502b. Main problem for tomorrow is the missing RAM.

Getting there slowly with #LT6502b it's now running BASIC and booting.
There are a couple of issues, most notably the lack of the 2nd RAM chip, most odd.

Second version of #LT6502 underway, so fa so good. Powers up, all voltages are correct.
So onto soldering in cpu and bring up.
#LT6502b #retrocomputing