Yeah component bag still needs to be filled 😀

Also featuring LED - so you can see if card is properly activated and mcu has unlocked the latch.

https://docs.google.com/spreadsheets/d/1Fevrq … B4MEPqeVpX3gECc

If anyone still has interested add you to list, sorry guys SMD only.. its decided, only 5 dht kits is not enough 😀
Don`t worry, there will be more 😀

How can I say no to ANOTHER Tandy soundcard by you. 😀

If it can output Tandy sound to the PC speaker, I would like one.

Works indeed 😀

Internal Passtrough -> PC-Speaker to Tandy -> Tandy to Soundblaster Vibra 16C

Can I chime in?
I've made myself my own variant of card on SN76498 for the purpose of FMonster debugging.
It is not based on Lotech or TNDY design, but is compatible with both from the used ports perspective.
I've managed to get most of the games launch and run successfully with TSR (thanks, pdw) and/or patches.

But I can not get VGM players to work.
VGMPlay.exe from Scali just hangs after accepting correct port, even '1' doesn't help.
Also there is one VGMPlayTSR.exe in SNDTSR folder (that I don't remember where comes from) that accepts SOME VGM files but spits out errors and crashes eventually. And it produces sound only if I load both lotech.exe TSR and than SNDTSR.EXE from that SNDTSR folder. But it's just random garbage.
Is it even possible to play some VGMs for SN76498?

From my experiments, Pin 9 ("NC") is a great antenna and capacitor between Pin 7 and Pin 9 helps a lot to reduce noise and whine. I've found that 1 nF capacitor is almost does not affect high frequencies, but everything lower than 470 pF works just fine. But don't go below 68 pF - lower capacitance starts to has lower effect on noise reducing.
Don't put a capacitor from Pin 9 to Pin 8 (GND)! It will just lower the noise/oscillation frequency from ~520 kHz into the hearing range. 100 nF to ground will produce horrible whine at ~15 kHz. And don't short Pin 9 to GND (as datasheet for SN76496 suggests), because there is DC level on that pin.
But a capacitor between Pins 7 and 9 if not enough to eliminate noise at the output of SN76489. RC damping circuit (33 Ohms - 100 nF) from Pin 7 to Pin 8 is a must.
Together those measures brought me less than -60 dB noise floor at the output.

And one more thing: my logic circuit for eliminating tones on power-up before running "sound off" software works as designed. It shuts off sound on power-up or after system reset, but allows sound to pass through after two accesses to the SN76489.

The Second VGMPlay uses a TSR Handler, so it can not work unless you install the TSR for it.

But something is weird there, why it shows Using LPT ?
Maybe you have a older version ? i attached mine.

Thanks for sharing your Information on SN76489,

i use a small attiny13 mcu for now and cut the sn76489 from bus.
the attiny13 can be programmed to detect the ce pulses and enable the latch and disables it on reset.
i also tried with a flipflop before, but it did not work out for me, the sound enabled on 2 machines during bios check because it did access the #2 dma controller on boot.
especially on a Epson POS PC who did multiple accesses.

You can also try out TNDY TSR, which does not require EM386 or QEMM386 at all, note that its not combatible with certain drivers, smartdrv for example.

https://github.com/JKnipperts/TNDY

i don`t know if latest code is on the page so i attach it here too.

Forgot to mention:
http://www.oplx.com/code/

sbvgm also supports tandy, and there are drivers for AIL Soundsystem aka Miles.

matze79 wrote on 2020-01-12, 10:25:

But something is weird there, why it shows Using LPT ?
Maybe you have a older version ? i attached mine.

Thanks, I've tested your version and it doesn't work as well.
It also spits out "Invalid" messages and produces garbage sound. Probably it doesn't like VGM files that I downloaded for SMS...

matze79 wrote on 2020-01-12, 10:25:

the sound enabled on 2 machines during bios check because it did access the #2 dma controller on boot.
especially on a Epson POS PC who did multiple accesses.

So far I've tested my board only on one motherboard, I have to test it on more boards.

matze79 wrote on 2020-01-12, 10:51:

Forgot to mention:
http://www.oplx.com/code/
sbvgm also supports tandy, and there are drivers for AIL Soundsystem aka Miles.

Thank you very much for that!
That program works fine with my board and I've got recordings on Sega Master System music on my board: https://www.youtube.com/watch?v=5B3-YxNLJCE

Fagear wrote on 2020-01-12, 17:47:

So far I've tested my board only on one motherboard, I have to test it on more boards.

this is only the case if you run card at 0c0h and maybe 0E0h on AT+ Systems, all other ports are fine.
some bios seem to access second dma controller on boot, which could potentially unlock the card.
i just introduced a delay long enough to allow system to run POST, afterwards the card can be unlocked.

i spent few minutes to rewrite the sndoff utility so it will properly unlock the sn76489 on tndy isa card,
result is attached.

matze79 wrote on 2020-01-12, 19:46:

this is only the case if you run card at 0c0h and maybe 0E0h on AT+ Systems, all other ports are fine.

Oh, so, basically, it is not a problem for me. Because I use my card on default 0x2C0 port with TSR for redirecting 0x0C0 or newer software that can utilize 0x2C0 directly.
I think that "0x0C0" address option on the card should be advertized as only "non-AT compatible for TRUE software experience (without TSRs or patches)". Basically usage of 0x0C0 on an AT system = improper usage.

Yeah the Experience of using 0C0 / 0E0 is really different, on some AT Systems its not even accessible, for example Intel Batman Chipset.

Its very interesting i run the Card with VIA VT82C580 Chipset with Pentium MMX Overdrive and have no noticeable side effects of running it on 0C0h.
Where my 486 is freaking out with the Card on 0C0h its not even be able to boot Windows 95.
Its really not a good idea to run the Card on 0C0h if not needed on a AT System, but one could put a 3 pol switch on the card to select io address before boot.
So you don`t have to open the case for flipping jumpers around.
The TSR will never achieve full compatibility.

matze79 wrote on 2020-01-12, 20:11:

Yeah the Experience of using 0C0 / 0E0 is really different, on some AT Systems its not even accessible, for example Intel Batman Chipset.

Have you tested 0E0 on a 430TX chipset platform? Very curious about that.

If not, 0B0 should be worth a try too if 0E0 does not work on AT+ systems.

It does work, but not everywhere 😀

i have no 430TX currently installed.

On Sub Pentium it should work fine.

I'll try this out on my 486 that has an OPTi chipset. Still need to figure out why King's Quest II has graphical glitches with the card, even after installing the resistor and capacitor.

i will try KQ2 again, but had no time yet.

Also a whole Load of parts have arrived, so DIY Builders should be prepared.
I still need to figure out howto pack the SMD parts the best Way.

Still miss few parts, and there is also need do to a technical drawing of the slot cover.

Thanks again to dreamblaster! 😀

matze79 that's how VGMPLAY works on my card/machine: https://youtu.be/-4YIZy-07sI?t=31m23s

Almost all parts are here,

Slotcovers are printed from PETG.
Unfortunately, the PLA breaks very easily and when it gets older it loses its elasticity and becomes brittle.

Now with PETG it is much better.
It is more stable, more elastic and you have to be less careful.

Very stable:
https://youtu.be/o2-BVyt_Eoo

i have a whole load of parts and was too lazy to sort them out, i will begin now 😀

can`t await to finish it!

@fagear did you try a different machine too ?
VGMPlayer worked always fine for me 😒

SN76489 Testing Tool.. now works on XT.
LPT and ISA Card supported.
Thanks Dragonsphere!

EDIT:
Hercules Monochrom not working.

Found out the King's Quest II game has graphical glitches and no patches to be found.