VOGONS


Reply 3000 of 3121, by terryfi

User metadata
Rank Newbie
Rank
Newbie

Finally done, so much trouble with solder paste, broken amp chip, weird software setup, but I am happy with the end result.
I am really surprised by sound quality of this, the clarity beats classic GUS, AWE32 and all software trackers I've tried so far.
A recorded tune is attached.

Attachments

  • Filename
    argus_test.flac
    File size
    3.39 MiB
    Downloads
    19 downloads
    File license
    Public domain
  • argus_done.jpg
    Filename
    argus_done.jpg
    File size
    535.53 KiB
    Views
    970 views
    File license
    Public domain

Reply 3001 of 3121, by FreddyV

User metadata
Rank Oldbie
Rank
Oldbie
Kahenraz wrote on 2022-06-21, 22:18:

Does the GUS need DMA to play music? Will it work in a board without functional DMA?

The GUS can use DMA to load samples in memory.
It is not needed to play samples as they are in the GUS Memory.

Reply 3002 of 3121, by shock__

User metadata
Rank Oldbie
Rank
Oldbie
terryfi wrote on 2022-06-22, 04:08:

Finally done, so much trouble with solder paste, broken amp chip, weird software setup, but I am happy with the end result.
I am really surprised by sound quality of this, the clarity beats classic GUS, AWE32 and all software trackers I've tried so far.
A recorded tune is attached.

Feels great to have finished such a project on your own doesn't it? Being able to solder (smaller) SMD also is a great skill to have 😀
Glad you're happy with the outcome of this project.

Current Project: new GUS PnP compatible soundcard

[Z?]

Reply 3003 of 3121, by Shponglefan

User metadata
Rank Oldbie
Rank
Oldbie
terryfi wrote on 2022-06-22, 04:08:

Finally done, so much trouble with solder paste, broken amp chip, weird software setup, but I am happy with the end result.
I am really surprised by sound quality of this, the clarity beats classic GUS, AWE32 and all software trackers I've tried so far.
A recorded tune is attached.

Great work!

Just curious, but what did you use for a bracket?

My YouTube channel (retro game music)

Reply 3004 of 3121, by terryfi

User metadata
Rank Newbie
Rank
Newbie
Shponglefan wrote on 2022-06-22, 12:35:
terryfi wrote on 2022-06-22, 04:08:

Finally done, so much trouble with solder paste, broken amp chip, weird software setup, but I am happy with the end result.
I am really surprised by sound quality of this, the clarity beats classic GUS, AWE32 and all software trackers I've tried so far.
A recorded tune is attached.

Great work!

Just curious, but what did you use for a bracket?

Used one from Sound Blaster 16 Vibra CT4170, it is a perfect fit.
https://www.ebay.com/itm/204001663988?hash=it … BAAAOSwJOlisRtT

Reply 3006 of 3121, by terryfi

User metadata
Rank Newbie
Rank
Newbie

I am having issues dumping iw rom, the content are corrupted repeating part of header, most time repeating corruption is different effect and once in a while they look similar. Anyone?
Is there anyway to test iw rom functionality, e.g. play midi with rom patch, without dumping?

Attachments

  • iwhex.PNG
    Filename
    iwhex.PNG
    File size
    42.23 KiB
    Views
    808 views
    File license
    Public domain

Reply 3007 of 3121, by 640K!enough

User metadata
Rank Oldbie
Rank
Oldbie
terryfi wrote on 2022-06-23, 15:24:

Is there anyway to test iw rom functionality, e.g. play midi with rom patch, without dumping?

To make trouble-shooting easier, this assumes that you are in a pure DOS environment.

We will be able to get a better idea of what is going on if you post your IW.INI and the full output from IWINIT (type IWINIT -v9 > iwinit.txt, then post the files; it is important that 'v' in "v9" be lowercase). It would also be helpful, if you want us to be able to help, if you would post a clear, well-lit picture of the ROM area, including the configuration resistors and latches (naturally, without the SIMM in the way).

The most telling indicator of whether the ROM will be usable is the presence of a line beginning with "Found ROM chip:" in the output from IWINIT. If you don't see such a line, then nothing you try will get it to use that instrument bank, and it's back to the soldering iron. If you're using the dumping tool as per shock__'s instructions, and that is the result that you're getting, then the most likely culprits are that the ROM IC itself is not soldered down correctly/firmly, or the same is true of the latches. If you're getting that repeating pattern, the data and control lines would appear to be functional, and a problem with the connection to the address bus seems the likely problem.

Make sure you have the right parts for the latches, and start by re-flowing just those (non-conductive flux is your friend), then try again. If you're still not having any luck, try the ROM chip, while also making a particular effort not to over-heat anything.

If, by some chance, you do see a "Found ROM chip:" line, and it is correct, then you can try using PLAY.EXE to play any MIDI file. There, too, consider posting a sample of the audio output and any messages it displays.

Trying to validate ROM functionality by playing a file is not exhaustive proof that the ROM is fully accessible and valid. The method shock__ described is a good way to know for sure. Once you get the answer you are looking for the first time, try it a few more times to ensure that it is repeatable.

Reply 3008 of 3121, by terryfi

User metadata
Rank Newbie
Rank
Newbie
terryfi wrote on 2022-06-23, 15:24:

I am having issues dumping iw rom, the content are corrupted repeating part of header, most time repeating corruption is different effect and once in a while they look similar. Anyone?
Is there anyway to test iw rom functionality, e.g. play midi with rom patch, without dumping?

One of the latch chips was dead cold, found cold solder joint on vcc of that, tried again same issue, re-soldered all pins of both latches and IW chip issue fixed.
Maybe I should have wasted a lot of wick removing the solder from all pretinned pads, I think it would be agood idea at least for the ICs; with that pretinning it is very hard to have all pins flush to board.

BTW apparently I did not press submit for this reply, meanwhile 640k explained the issue, thanks 640k issue was exactly the connections.

Reply 3009 of 3121, by Stevogamer

User metadata
Rank Newbie
Rank
Newbie
terryfi wrote on 2022-06-22, 04:08:

Finally done, so much trouble with solder paste, broken amp chip, weird software setup, but I am happy with the end result.
I am really surprised by sound quality of this, the clarity beats classic GUS, AWE32 and all software trackers I've tried so far.
A recorded tune is attached.

What RAM did you use the btw?

Kinda confused about what to buy for the SIMM. 😅 I grew up with PC133 SDRAM and beyond, so SIMMs are new to me.

I want to either have an 8MB or 16MB stick, although I realize this is probably going to be overkill.

Reply 3010 of 3121, by terryfi

User metadata
Rank Newbie
Rank
Newbie
Stevogamer wrote on 2022-06-26, 23:37:
What RAM did you use the btw? […]
Show full quote
terryfi wrote on 2022-06-22, 04:08:

Finally done, so much trouble with solder paste, broken amp chip, weird software setup, but I am happy with the end result.
I am really surprised by sound quality of this, the clarity beats classic GUS, AWE32 and all software trackers I've tried so far.
A recorded tune is attached.

What RAM did you use the btw?

Kinda confused about what to buy for the SIMM. 😅 I grew up with PC133 SDRAM and beyond, so SIMMs are new to me.

I want to either have an 8MB or 16MB stick, although I realize this is probably going to be overkill.

RAM you see if 4 MB FPM with parity. In this thread people mentioned that you can use FPM, EDO parity non parity almost any 72 pin RAM up to 16 MB.

Terry

Reply 3011 of 3121, by 640K!enough

User metadata
Rank Oldbie
Rank
Oldbie
Stevogamer wrote on 2022-06-26, 23:37:

Kinda confused about what to buy for the SIMM. 😅 I grew up with PC133 SDRAM and beyond, so SIMMs are new to me.

It is important to make sure that the module you buy is meant for use at 5V. Some of the memory you find will be for old Apple or IBM hardware, and may be designed for 3.3V operation only; these will not work. Almost any 72-pin SIMM containing 5V FPM or EDO DRAM specified at 70 ns or faster (lower number) should work. Modules supporting parity checking offer no advantage, but if they are cheap or all you can find, they should work without trouble.

It is also important to repeat that the memory test built into the Gravis SETUP application is useless; just because the test "passed", does not mean that your memory is compatible, usable or not defective. If the test detects a problem, there definitely is one; if it doesn't, but you're still getting distorted playback, the memory could still be the cause.

Reply 3012 of 3121, by Stevogamer

User metadata
Rank Newbie
Rank
Newbie
640K!enough wrote on 2022-06-27, 03:56:
Stevogamer wrote on 2022-06-26, 23:37:

Kinda confused about what to buy for the SIMM. 😅 I grew up with PC133 SDRAM and beyond, so SIMMs are new to me.

It is important to make sure that the module you buy is meant for use at 5V. Some of the memory you find will be for old Apple or IBM hardware, and may be designed for 3.3V operation only; these will not work. Almost any 72-pin SIMM containing 5V FPM or EDO DRAM specified at 70 ns or faster (lower number) should work. Modules supporting parity checking offer no advantage, but if they are cheap or all you can find, they should work without trouble.

It is also important to repeat that the memory test built into the Gravis SETUP application is useless; just because the test "passed", does not mean that your memory is compatible, usable or not defective. If the test detects a problem, there definitely is one; if it doesn't, but you're still getting distorted playback, the memory could still be the cause.

Thank you for that explanation 😀

Would it matter what memory configuration in terms of number of chips on the stick? Or do I just need to meet the 70ns, single rank, 72 pin, and 5V requirements?

Reply 3013 of 3121, by 640K!enough

User metadata
Rank Oldbie
Rank
Oldbie
Stevogamer wrote on 2022-06-27, 04:09:

Would it matter what memory configuration in terms of number of chips on the stick? Or do I just need to meet the 70ns, single rank, 72 pin, and 5V requirements?

The arrangement of ICs on the module generally doesn't matter, as long as the other firm requirements are met. Some people have found particular modules that didn't work, but that is a small minority of cases.

It may also be a good idea to match the module to the socket. That is to say, if you use a socket with tin-plated contacts, pick a module to match, and similarly for gold-plated-contact sockets. This isn't an absolute requirement, but may be better for long-term reliability.

Reply 3014 of 3121, by Stevogamer

User metadata
Rank Newbie
Rank
Newbie
640K!enough wrote on 2022-06-27, 04:21:
Stevogamer wrote on 2022-06-27, 04:09:

Would it matter what memory configuration in terms of number of chips on the stick? Or do I just need to meet the 70ns, single rank, 72 pin, and 5V requirements?

The arrangement of ICs on the module generally doesn't matter, as long as the other firm requirements are met. Some people have found particular modules that didn't work, but that is a small minority of cases.

It may also be a good idea to match the module to the socket. That is to say, if you use a socket with tin-plated contacts, pick a module to match, and similarly for gold-plated-contact sockets. This isn't an absolute requirement, but may be better for long-term reliability.

That is good to know. I never thought about the difference in metal for contacts on electrical components. I'll keep that in mind. 😀 I want this card to last me a very long time.

Reply 3015 of 3121, by digger

User metadata
Rank Oldbie
Rank
Oldbie
FreddyV wrote on 2022-06-22, 07:29:
Kahenraz wrote on 2022-06-21, 22:18:

Does the GUS need DMA to play music? Will it work in a board without functional DMA?

The GUS can use DMA to load samples in memory.
It is not needed to play samples as they are in the GUS Memory.

Were there many games that actually leveraged this properly on the GUS, though?

I get the impression that most game developers implemented sound playback in a way that would work with Sound Blaster and compatible cards, and therefore didn't bother to take advantage of the inherent architectural advantages of the GUS, even if the games supported that card. It seems to me that this was definitely the case for games that relied on abstraction offered by DOS sound drivers (Miles, DIGPAK, etc).

Heck, even many DOS games that used tracker music would mix the channels in software, even when playing back on a GUS, which is such a shame. 😕

Reply 3016 of 3121, by FreddyV

User metadata
Rank Oldbie
Rank
Oldbie
digger wrote on 2022-06-27, 09:03:

Were there many games that actually leveraged this properly on the GUS, though?

I get the impression that most game developers implemented sound playback in a way that would work with Sound Blaster and compatible cards, and therefore didn't bother to take advantage of the inherent architectural advantages of the GUS, even if the games supported that card. It seems to me that this was definitely the case for games that relied on abstraction offered by DOS sound drivers (Miles, DIGPAK, etc).

Heck, even many DOS games that used tracker music would mix the channels in software, even when playing back on a GUS, which is such a shame. 😕

The GUS can behave like a sound blaster, play a sample from DMA and generate IRQ at the end of the buffer, It was surely used that way by many application.
For the Demos and trackers, send the samples in memory with I/O is more simple, so the DMA is not used a lot.

Reply 3017 of 3121, by aitotat

User metadata
Rank Member
Rank
Member

I'm almost done soldering mine but I have a question about x1 and x2. Is x2 mic in and x1 line in? So same order as real GUS PnP? I have color connectors so I want to be sure before testing it.

Reply 3018 of 3121, by terryfi

User metadata
Rank Newbie
Rank
Newbie
aitotat wrote on 2022-06-29, 09:33:

I'm almost done soldering mine but I have a question about x1 and x2. Is x2 mic in and x1 line in? So same order as real GUS PnP? I have color connectors so I want to be sure before testing it.

Based on schematic, from top to bottom on latest board, LINE, MIC and OUT.