VOGONS


First post, by Dimitri

User metadata
Rank Newbie
Rank
Newbie

arrived today 5 prototype boards
VAXccTB.jpg

why "advanced"?
*) has 32 words (32x8) FIFO IC (on back side)
*) has precision IC DAC (over FIFO IC on top)
*) has dual power +/-12VDC for filter and amplifier ICs
didn't tested yet 😁

Reply 1 of 16, by doaks80

User metadata
Rank Member
Rank
Member

What would it take to get full dss and covox compatibility?

k6-3+ 400 / s3 virge DX+voodoo1 / awe32(32mb)
via c3 866 / s3 savage4+voodoo2 sli / audigy1+awe64(8mb)
athlon xp 3200+ / voodoo5 5500 / diamond mx300
pentium4 3400 / geforce fx5950U / audigy2 ZS
core2duo E8500 / radeon HD5850 / x-fi titanium

Reply 2 of 16, by Dimitri

User metadata
Rank Newbie
Rank
Newbie

I think nothing for DSS, signals should be the same in LPT port. About covox - the diff is in FIFO stack, I dunno how it does with covox sw. The board is even not alpha, it's pre-alpha prototype, didn't tested yet. If anyone has a wish to test - welcome, I have DIY IC only (RCL you should find yourself) kits.

Reply 3 of 16, by Scali

User metadata
Rank l33t
Rank
l33t

A Covox has no FIFO at all, the 8 datalines from the LPT are passed through the resistor ladder DAC immediately.
So if you want Covox compatibility, you'd need some way to bypass the FIFO. A simple switch on the PCB perhaps?

http://scalibq.wordpress.com/just-keeping-it- … ro-programming/

Reply 5 of 16, by Scali

User metadata
Rank l33t
Rank
l33t
Dimitri wrote:

may be next time, but the glue is to has better sound, which offers DSS.

Better in what way? DSS is limited to 7 kHz audio. Covox can go as fast as your CPU allows. Playing 44.1 kHz is possible on a fast 386 or 486.
The main advantage the DSS has is that its FIFO makes it less CPU-intensive for low-end systems. But even a 8088 at 4.77 MHz can easily get better than 7 kHz audio from a Covox.

Now, if your DSS would not have a fixed 7 kHz clock, but you could get it to 44.1 kHz at least, then you may have the best of both worlds. But you wouldn't have any software that supports it, unlike the Covox, where there's quite a few games, demos and trackers that support 44.1 kHz.

http://scalibq.wordpress.com/just-keeping-it- … ro-programming/

Reply 7 of 16, by Scali

User metadata
Rank l33t
Rank
l33t
Dimitri wrote:

oh, common! Covox and quality sound in 44KHz is impossible.

Say what?
See here: https://scalibq.wordpress.com/2017/11/28/the-covox-years/

Yea, it's mono, and it's 8-bit, but within those parameters it delivers excellent fidelity. 44.1 kHz 8-bit mono sound is not that bad really. Sounds better than an SB2.0 really.

http://scalibq.wordpress.com/just-keeping-it- … ro-programming/

Reply 9 of 16, by dionb

User metadata
Rank l33t
Rank
l33t

Agreed. Some clones do do it justice - I have an Aztech Sound Galaxy NX Pro, which not only does SB(Pro)2.0 but offers Covox support as an aside. It's a horribly noisy (or rather: unshielded) card in general, but as it has a real OPL3 it's a good side-by-side comparison between Covox and SB, and I was surprised by how good the Covox sounded. It's limited, both in terms of being mono and its extremely primitive and timing-sensitive interface, but within those limits it was far better than I had expected.

Reply 11 of 16, by digger

User metadata
Rank Member
Rank
Member

I love DIY projects like this. 😀 Great job so far.

Some things I would very much like to see in a DSS-on-steroids:

* 100% downwards compatibility with the actual Disney Sound Source (duh)
* A Covox Speech Thing (and possibly also stereo-on-one) compatibility mode by bypassing the FIFO buffer, as suggested by others here already
* A mode that allows it to coexist and work concurrently with an OPL2LPT or OPL3LPT on a single parallel port
* An advanced mode that allows the FIFO buffer to be configured for arbitrary bitrates other than just 7kHz, including the more common 11kHz used in most DOS games that supported sound cards
* An even more flexibly programmable FIFO buffer that could offer configurable larger buffer sizes and/or stereo support
* Optional ADPCM decompression support in hardware
* The leveraging of the IRQ of the parallel port to provide some level of hardware assistance to a software-based Sound Blaster emulator tailored for this "DSS Plus" device

These are probably crazy ideas, but hey, a geek can dream, right? 😉

Reply 12 of 16, by matze79

User metadata
Rank Oldbie
Rank
Oldbie

shock_ also has made a combo stereo/mono/dss Plug which is switchable.

I`m now waiting for PCBs to arrive from him. Dunno how long it will take 😀

Reversing the Disney Sound Source

->
* A mode that allows it to coexist and work concurrently with an OPL2LPT or OPL3LPT on a single parallel port

i don`t think this will be possible without breaking DSS combatiblity.

Eg using OPL2 and Digital on Wolfenstein..

https://dosreloaded.de - The German Retro DOS PC Community
https://www.retroianer.de - under constructing since ever

Co2 - for a endless Summer

Reply 15 of 16, by shock__

User metadata
Rank Oldbie
Rank
Oldbie
matze79 wrote:

I`m now waiting for PCBs to arrive from him. Dunno how long it will take 😀

Waiting for the PCBs myself ... according to the PCB manufacturer they should have arrived already, which they didn't. Let's hope they didn't get lost.

While I do like high quality covox stuff, I fail to see the point of this project. The only case where one should care about DSS is backward-compatibility which you're breaking by going with a FIFO that has a larger depth unless you're filling/dropping the second half or "stuffing" every other word (original has 16x8bit) - in that case it's a cool project as the 74HCT40105 I'm using are only 4bits wide and probably soon to be discontinued (if they aren't already). Going high quality when the software expects ~7khz output also seems a bit overengineered 😉
On the other hand - best of luck. Reverse engineering the DSS was one of my first serious projects 😀

Current Project: new GUS PnP compatible soundcard

[Z?]

Reply 16 of 16, by Scali

User metadata
Rank l33t
Rank
l33t
shock__ wrote:

While I do like high quality covox stuff, I fail to see the point of this project. The only case where one should care about DSS is backward-compatibility which you're breaking by going with a FIFO that has a larger depth unless you're filling/dropping the second half or "stuffing" every other word (original has 16x8bit) - in that case it's a cool project as the 74HCT40105 I'm using are only 4bits wide and probably soon to be discontinued (if they aren't already).

Yes, I think first and foremost the device has to have a 100% DSS-compatible mode.
It would be an interesting feature if it can also switch to Covox-compatible mode, and even moreso if there's an advanced mode with a deeper FIFO and higher playback rates than the original DSS.
It would be interesting for people developing new games and demos. But the primary focus should be on compatibility with existing games and demos, because that's what most people would be using it for.

http://scalibq.wordpress.com/just-keeping-it- … ro-programming/