wrote:I'm not sure if you answered this question before, but do you consider this project open-source, or more proprietary than that (i.e. you'd like to tightly control the firmware, etc.)?
The firmware is open source, GPL licensed. Even if I didn't want to, I wouldn't have much choice since most of it was stolen from SoftMPU/DOSBox 😁
The hardware, I don't know yet. Don't want Chinese clones out there, haha
wrote:I wouldn't have much choice since most of it was stolen from SoftMPU/DOSBox 😁
Ahh... makes sense 😀
wrote:Ahh... makes sense 😀
In case you hadn't found it yet... https://github.com/ab0tj/HardMPU
Quick teaser. Here's what I got in the mail last evening. 😀
(edit) Small delay encountered, turns out I don't have several of the ICs in PDIP package (I have them as SOIC only) so it will be a few days until I can finish assembling it. I will solder the passives and IC sockets in the meantime.
I/O, I/O,
It's off to disk I go,
With a bit and a byte
And a read and a write,
I/O, I/O
Hey, I made something too. Anyway, 75% of a thing is better than nothing, right? 😀
I/O, I/O,
It's off to disk I go,
With a bit and a byte
And a read and a write,
I/O, I/O
Card almost complete. I managed to scrounge most of the ICs except for a LS240. I don't have one, period. Waiting for the ones that should be in the mail already. Well, that and a 20MHz crystal. I only have 16MHz and 24MHz ones - but there's a small lot of the 20MHz in the mail heading for me too. (I wonder if it still worked with a 16MHz crystal?)
Notice the DB9F mounted below the PCB, due to the PCB error mentioned earlier in this thread. And the ghetto fuse holder improvised from some scrap wire and nickel strips, because I don't have fancy-schmancy fuses like ab0tj. 😀
I/O, I/O,
It's off to disk I go,
With a bit and a byte
And a read and a write,
I/O, I/O
wrote:Card almost complete. I managed to scrounge most of the ICs except for a LS240. I don't have one, period. Waiting for the ones that should be in the mail already. Well, that and a 20MHz crystal. I only have 16MHz and 24MHz ones - but there's a small lot of the 20MHz in the mail heading for me too. (I wonder if it still worked with a 16MHz crystal?)
Notice the DB9F mounted below the PCB, due to the PCB error mentioned earlier in this thread. And the ghetto fuse holder improvised from some scrap wire and nickel strips, because I don't have fancy-schmancy fuses like ab0tj. 😀[/img]
Nice! 16MHz should be quick enough but you'd have to adjust some settings for baud rate, etc, in the firmware. Or overclock it and use the 24MHz 😎
I guess the 240 was a bit of a curve ball, I switched to that when redrawing the schematic to save on chip count.
Well I'll wait for the 20MHz crystal to get here, should be about the same time the 240 arrives.
Also I removed the optocoupler socket and soldered it directly. It's because the 6N136 chips I have are the short-leaded type with only stubs of the thin part of the leads, and curved horizontally so it can be soldered in surface-mount style. I tried straightening out the stubs, but they're too short and the chip kept popping out of the socket. Eventually I soldered it onto the surface of the mounting pads. 😀
I/O, I/O,
It's off to disk I go,
With a bit and a byte
And a read and a write,
I/O, I/O
Awesome to see this becoming a finished piece of hardware, good job!
I think these are cute together. Just finished the bottom one. Or almost finished, I'm missing the output jack. Off to Radio Shack tomorrow, I think I saw there something that may fit.
I/O, I/O,
It's off to disk I go,
With a bit and a byte
And a read and a write,
I/O, I/O
wrote:Just finished the bottom one.
Adlib clone?
wrote:wrote:Just finished the bottom one.
Adlib clone?
I/O, I/O,
It's off to disk I go,
With a bit and a byte
And a read and a write,
I/O, I/O
wrote:I think these are cute together. Just finished the bottom one.
Fun fact, I used the KiCad files from Sergey's floppy/serial controller as a starting point to get the measurements close. So I think that makes those two cards cousins!
wrote:So I think that makes those two cards cousins!
Yup, hillbilly cousins. The top one with the "underground" connector and improvised fuse holder. the bottom one with the potentiometer precariously hanging by a few re-purposed cut resistor terminals and missing the last part - the output jack.
(I hit ebay and found the exact potentiometer and jack I need, so I'll be waiting for those to test. Story of my life, always waiting on ebay deliveries)
I/O, I/O,
It's off to disk I go,
With a bit and a byte
And a read and a write,
I/O, I/O
wrote:Yup, hillbilly cousins. The top one with the "underground" connector and improvised fuse holder. the bottom one with the potentiometer precariously hanging by a few re-purposed cut resistor terminals
Lol; I laughed at this.
I received the 20MHz crystals today. Unfortunately the LS240 chips didn't get here, and according to tracking they may not be here until Monday or Tuesday, which delays testing.
I/O, I/O,
It's off to disk I go,
With a bit and a byte
And a read and a write,
I/O, I/O
Some quick (bad) news. I got the LS240 (they're actually HCT240, compatible), installed it, fired a test PC with some games on it. The test PC is a 386-40 (the CPU is a TI486DLC) with 16MB RAM and a microdrive for HDD. I only tested with Secret of the Monkey Island (EGA version plus MT32 patch).
I tested it first with keropi's Music Quest clone and my MT32(old style). Works fine with the default settings I/O=330, IRQ2.
The HardMPU board that I built however doesn't work. The MT32 doesn't receive a signal. I used the same settings: IRQ2 (jumper on the first pair in the IRQ block) and I/O=330 (jumper on last pair in the I/O block). I checked that the MIDI_out signal goes to the correct pin on the output connector (pin6 on the DB9F) as is the case with my under-the-pcb-mounted connector. I used thus the same breakout cable as for the Music Quest clone.
Troubleshooting tomorrow, it's late now. I'm bummed.
ab0tj: is the Atmega chip supposed to stay cold? The one on my PCB does, even after 10 minutes powered up and supposedly playing music. The Zilog CPU on the MQ card gets pretty warm. But then it's a completely different beast.
I/O, I/O,
It's off to disk I go,
With a bit and a byte
And a read and a write,
I/O, I/O
I was gonna write to check the pinout of the breakout cable but you did that already.... Run GateWay and see what it reports, if it cannot detect an intelligent mode mpu it will complain on start
Yes, Gateway is complaining of not finding a MPU on IRQ2.
Now about those clock fuses again. There's a bit of ambiguity in the software of my programmer (TL866, aka MiniPro) regarding how the fuses are to be programmed. For external clock you want the 4 fuse bits to be "0000" However the fuse section in the software has 4 checkboxes for said fuses with the description for them "CKSEL0=0" through "CKSEL3=0" I assumed that this means if the checkbox is checked, the fuse will be zero - so I checked them all when burning it. But now I wonder if I should leave them unchecked. It's even more ambiguous when I look at the datasheet for the 1284; the table in section 10.2 that details clock selection fuses says "for all fuses, 1 means unprogrammed while 0 means programmed" which would reinforce my initial interpretation (all 4 checkboxes checked).
I/O, I/O,
It's off to disk I go,
With a bit and a byte
And a read and a write,
I/O, I/O
you need to start looking on the ICs then and make sure the PIC is programmed correctly
