DIY 8-bit computer, Episode 1D: TMS9918A shenanigans
HTML-код
- Опубликовано: 26 июл 2024
- The goal of this episode is to experiment with a TMS9918A Video Display Processor and to attempt to use it as a display for the 6809 8-bit computer system.
With the TMS9918A providing a display, we're one step closer to the 6809 system being a fully independent computer.
Link to schematics and code for this episode: github.com/daveho/DIY8bit/tre...
The demo circuit shown in the video, which I will be using as the basis for the TMS9918A module, was created by Dr. Volt.
Dr. Volt's TMS9918A project on Github: github.com/michalin/TMS9918_A...
Dr. Volt's youtube channel: @DoctorVolt
I was inspired to try the TMS9918A based on a suggestion from Troy Schrapel:
Troy's 6502 system (which uses the TMS9918A): visrealm.github.io/hbc-56/
Troy's youtube channel: @TroySchrapel
Video timeline:
00:00 Titles
00:18 Introduction
01:59 Overview of the TMS9918A
04:27 Where to start?
05:35 Overview of Dr. Volt's circuit
12:18 A visit with Ingo
12:37 Completed demo circuit
14:36 Demo circuit success
16:23 Interfacing with the 6809
18:55 All your base are belong to us
19:44 Code walkthrough
22:41 Conclusions, what next?
I salute the humility in you @ I am not Ben Eater ( the god of electronics and breadboard projects )
His breadboarding and design skills are legendary.
Great work, Dave. Awesome to see another TMS9918A project. I've heard the lack of color can be caused by the crystal or its load capacitors being slightly off and the color burst isn't picked up by the display. Other displays might be more tolerant. Looking forward to seeing what you do with it. 😎
Thanks, Troy! I have a composite to VGA adapter that should be arriving today, crossing my fingers that will work better. I could also just get an LCD TV from a thrift store. BTW I used your schematic to double-check my decoding of the read/write signals. I'm now wishing I had the glue logic board generate a -R/W signal rather than just R/-W, that would have simplified things.
Nice to see a new video from you!
Yay, a new daveho hacks video!
Nice video! This has been a fun series to follow
Glad you've enjoyed the series!
So happy to see another video in the series. Thank you so much for continuing. I'm learning so much, and drawing a lot of inspiration from you. Thank you! 🙇👍
Thank you for the kind words! This has been a fun project.
Fun!
Yaay! Great to see you back Dave. I was getting concerned that you had abandoned this project. As usual, our projects parallel eachother - I have a V9958 in the drawer to put into my build at some point. In the meantime I am cheating with a combination of USB-to-serial keyboard interface and serial VGA module. At the moment I am banging my head against a wall trying to get an Arduino Mega to drive a WD37C65 floppy disk controller. This is a test/preparation/proof of concept project for using the disk controller in my 6809 build. I've already realised that trying to use 1.44MB disks is likely to be an non-starter - the 2MHz 6809 is unlikely to be able to keep up with the data transfer rate and response times required - so I am planning to use 720KB disks to enable better portability. Handling the FAT12 disk format is the next thing on the agenda when I finally get this damn controller to work! Looking forward to the next video 👍😁
Wow, I would love to be able to give the V9958 a try. I also have storage in the back of my mind. Using a real floppy would be fun, I still have some 3.5" disks kicking around. At my current rate of progress I'll probably get to that around 2030 :-) I'm also thinking somewhat seriously about building a CPU/glue logic card with a 68008: those can run at 8 MHz, and perhaps would be powerful enough for 1.44 MB floppies? I've also been thinking about SCSI, perhaps the classic NCR 5380?
@@davehohacks The 68008 can certainly handle 3.5inch floppies - I had such a drive for my Sinclair QL back in the day. Not sure if it was 720k or 1.44M - probably the former as I don't thing the latter was common at the time. It should be fast enough for 1.44M though. I think SCSI would be a step too far - the protocol is quite complex, although I had a 30MB(!) external hard disk for my Atari ST which used the Atari butchered version of a SCSI interface on a 68k so it is possible. I am using real floppies at the moment (I have a nice pair of Sony MPF920 drives with black fronts) but I am open to the option of using a Gotek if ever I get the damn thing to work! (I found a couple of mistakes in my code this morning and things are working more sensibly so at least I am making progress!) 😁
@@spacedock873 I remember reading about the QL in Byte magazine. I don't think they were ever widely used here in the US, though. I'm sure you're right about SCSI being too ambitious, and I'm not sure where I would even find working SCSI storage devices at this point, although I can dream 🙂
@@davehohacks The QL was not a great machine on the grand scheme of things (it was made by Sinclair after all! 😂) but after years of 6809 computers (Dragon 32/64) I lusted after a 68k machine and the QL was the cheapest available and the only one a poor University student could afford!
@@spacedock873 We definitely used SCSI hard drives (and MFM/RLL drives with SCSI controller boards) on CoCos back in the day, so the 6809 is definitively fast enough to do SCSI, even using a SCSI interface that is little more than a buffer. We also used 720K 3.5" disks with a stock CoCo floppy disk controller. One of the early CoCo disk controllers could be modified for high density operation because the FDC chip in it used an external data separator chip, having been designed for 8" drives. I've never done this myself, but reportedly it works at least with 1.2M 5.25" drives with a special driver under OS-9 LII.
another interesting video! thank you for posting
I was looking for a project like this.... and this checks ALL the boxes... I would like to start by saying Thanks for all the great work. I will be building this whole project and try to catch up to where you are today. I will be placing an order with the PCB folks and was wondering. This seems to be the first RUclips that you speak about the TMS9918A, but in the github there is a full pcb for TMS9918A along with the sound circuits. I am pretty sure I have gone through all the videos and have not seen that one. Is that PCB fully tested? Can I safetly add it to my PCB order?
Once again thanks for this.. I really look forward to start building.
Glad you have enjoyed the videos! A video on the TMS9918+YM2149F PCB is coming at some point. The TMS9918A part of it does appear to work correctly. I haven't tested the YM2149F part yet.
Oh, also, I think the gerbers need to be regenerated. I forgot to include the component designators, and there was one typo on the silkscreen (which I fixed on the PCB design.)
Also the VDP used on the NABU PC…
I've seen some videos about those! Interesting machine.
So I have got right up to the keyboard and all is working. I am working at porting over Enhanced Basic for this computer. Quick question, do you have a development environment setup (debugger, emulator) that you are using? There are lots of CoCo and Tandy dev environments, but most are not configurable(at least not easily) for the ports addresses we have here. I would like to be able to set Break points and step through the program and see registers and stack. I am looking at one that works with Rust, but it seems not to like all code.
If you have one... maybe a quick video on what you are using, if not, I will continue to try and find something that works reliably.
As always.... Thanks for the great work.
Wow, nice, glad you got things working! I do not have any development environment. I've been writing code, assembling it using asm6809, burning it onto an EEPROM, and then testing it on the actual hardware. A software emulator would definitely be helpful, I might work on putting something together this summer.