VOGONS


CVX4 : high quality covox adapter

Topic actions

Reply 80 of 484, by Scali

User metadata
Rank l33t
Rank
l33t
keropi wrote:

are you seriously comparing an LPTDAC device with Paula? c'mon man, this is not a dedicated DAC we are talking about, this is as simple as it gets - don't expect it to replace your soundblaster 🤣 🤣 🤣

Well, there's no reason why a DAC connected to an LPT can't sound as good a DAC on an SB, in theory.
And judging MobyGamer's capture, the Covox has a very high-quality DAC, as far as 8-bit DACs go. I always expect to hear some quantization noise with 8-bit samples, especially on more quiet passages, but the Covox sounds very VERY good there.

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

Reply 81 of 484, by keropi

User metadata
Rank l33t++
Rank
l33t++

yes sure, if you make a DAC that costs 30eur in parts then yes *maybe* it will sound as good... let's be realistic here 🤣

🎵 🎧 PCMIDI MPU , OrpheusII , Action Rewind , Megacard and 🎶GoldLib soundcard website

Reply 82 of 484, by Scali

User metadata
Rank l33t
Rank
l33t
keropi wrote:

yes sure, if you make a DAC that costs 30eur in parts then yes *maybe* it will sound as good... let's be realistic here 🤣

Dunno, the Covox seems to do it, and not at a super-high cost.
Besides, the SB is actually pretty low quality. It's nowhere near the quality of a Pro Audio Spectrum, a GUS or such.
Heck, if you hear the quality of 8-bit samples on the C64, which is done by exporting a weird glitch of the SID chip, which was never designed to play samples in the first place... It doesn't have to be wall THAT critical:
https://youtu.be/bDuwh2HOkKA

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

Reply 83 of 484, by lissajous

User metadata
Rank Newbie
Rank
Newbie
Scali wrote:

If you play a single 8-bit sample on an Covox, then it could/should be comparable with a single Amiga channel playing an 8-bit sample. But my impression is that it is.

Why not do that just to test? Play a single high quality 8-bit 44.1 or 48 kHz sample through the hardware just to see what it sounds like. It's actually surprising how close you can get to perceived CD quality within those constraints. If oversampling and noise shaping are also performed then you can comfortably exceed measured CD quality.

Along with a latch powered by a low noise supply (acting as a buffer in order to isolate the DAC from the potentially noisy PC) you could also insert a sample and hold circuit after the R2R ladder to deglitch the inevitably glitchy sample transitions that you get with a R2R design.

Reply 84 of 484, by Scali

User metadata
Rank l33t
Rank
l33t
lissajous wrote:

Why not do that just to test? Play a single high quality 8-bit 44.1 or 48 kHz sample through the hardware just to see what it sounds like. It's actually surprising how close you can get to perceived CD quality within those constraints. If oversampling and noise shaping are also performed then you can comfortably exceed measured CD quality.

Not sure why we should take the Amiga as reference for an 8-bit DAC? While the Amiga certainly sounds okay, it is a custom chip design by Commodore from 1985, and I'm not sure where they stand exactly in terms of quality. I wouldn't be surprised if they just used a simple-but-effective implementation, and there are far more high-end 8-bit DACs available on the market.
Aside from that, the Amiga can only play back at ~28.5 kHz max with DMA. Faster replaying would need to be bit-banged by the CPU, in which case your temporal jitter is as good as the routine you use. So you are introducing extra variables into the equation there...

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

Reply 85 of 484, by keropi

User metadata
Rank l33t++
Rank
l33t++

I hope mobygamer opens his Speech Thing soon so we can have a look and find out why it sounds better so it gets copied to the CVX2 , the discussion here is getting derailed on how to make a LPT Soundblaster/Gravis/Paula/whatever equivalent...

🎵 🎧 PCMIDI MPU , OrpheusII , Action Rewind , Megacard and 🎶GoldLib soundcard website

Reply 87 of 484, by keropi

User metadata
Rank l33t++
Rank
l33t++

make that 3 - dreamblaster is also interested for obvious reasons 🤣

🎵 🎧 PCMIDI MPU , OrpheusII , Action Rewind , Megacard and 🎶GoldLib soundcard website

Reply 88 of 484, by dreamblaster

User metadata
Rank Oldbie
Rank
Oldbie

Yeah, I am interested to know what's inside that Speech Thing.
But don't break it for that, if you can carefully open it, would be great.

Visit http://www.serdashop.com for retro sound cards, video converters, ...
DreamBlaster X2, S2, S2P, HDD Clicker, ... many projects !
New X2GS SE & X16GS sound card : https://www.serdashop.com/X2GS-SE ,
Thanks for your support !

Reply 90 of 484, by QBiN

User metadata
Rank Oldbie
Rank
Oldbie

I have one. Obviously not the one the recording was taken from... but hope it helps all the same. The clamshell on the adapter itself is easy to take off safely with a small flathead screwdriver (very low risk of damage unless you're a careless brute).

I would want to assume that it's just a simple resistor pack. However, since the passive pack has the covox seal on it, it is possible they've sealed up some key bits in that pack.

Unfortunately, it would seem impossible to disassemble any further without destroying it.

Covox1.jpg
Covox2.jpg

Reply 91 of 484, by keropi

User metadata
Rank l33t++
Rank
l33t++

Thanks for the pics QBiN !!!

That covox-branded resistor pack is worrisome... maybe it combines resistors and caps 😐

🎵 🎧 PCMIDI MPU , OrpheusII , Action Rewind , Megacard and 🎶GoldLib soundcard website

Reply 92 of 484, by Scali

User metadata
Rank l33t
Rank
l33t

Thanks for that, QBiN!
I've never seen the insides of a Covox before. Apparently they put everything into a single package, so indeed, there's nothing to see there.
I guess MobyGamer wouldn't have to open his Covoxes, because they'll likely look the same on the inside (even if there may be variations inside the pack).
I suppose they did it this way because they knew it was going to be trivial to clone the device. This way they at least obfuscate their particular implementation.
I guess that means that we'll just have to do research on how to create the best possible 8-bit DAC, and re-invent the Covox wheel ourselves.
The only thing you could do perhaps, is to measure the resistance from each data pin to the output jack (since it's probably little more than an R2R ladder inside). That can be done without even taking it apart. Not sure if that data will lead to any new insights though.

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

Reply 94 of 484, by keropi

User metadata
Rank l33t++
Rank
l33t++

maybe there is still a point to open all 3 Speech Things mobygamer has - maybe an old version uses discrete components and not that array

🎵 🎧 PCMIDI MPU , OrpheusII , Action Rewind , Megacard and 🎶GoldLib soundcard website

Reply 95 of 484, by MobyGamer

User metadata
Rank Member
Rank
Member
keropi wrote:

maybe there is still a point to open all 3 Speech Things mobygamer has - maybe an old version uses discrete components and not that array

Was hoping it wouldn't come to that 😀 but ok. Give me a few hours to do this.

Reply 96 of 484, by lissajous

User metadata
Rank Newbie
Rank
Newbie
Scali wrote:
lissajous wrote:

Why not do that just to test? Play a single high quality 8-bit 44.1 or 48 kHz sample through the hardware just to see what it sounds like. It's actually surprising how close you can get to perceived CD quality within those constraints. If oversampling and noise shaping are also performed then you can comfortably exceed measured CD quality.

Not sure why we should take the Amiga as reference for an 8-bit DAC?

It's pretty clear I'm not talking about the Amiga here I thought. I just mentioned it before because I think it sounds better than anything I've heard linked to in this thread so far and because it's my personal reference for decent 8-bit audio. I think a well-designed LPT DAC could certainly exceed it in quality without being prohibitively expensive.

Reply 97 of 484, by keropi

User metadata
Rank l33t++
Rank
l33t++
MobyGamer wrote:

Was hoping it wouldn't come to that 😀 but ok. Give me a few hours to do this.

2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg2q2emiv.jpg

🎵 🎧 PCMIDI MPU , OrpheusII , Action Rewind , Megacard and 🎶GoldLib soundcard website

Reply 98 of 484, by MobyGamer

User metadata
Rank Member
Rank
Member

Okay, here are all three of mine disassembled:

https://goo.gl/photos/PGaVfpzbvTYyTGdB6

https://goo.gl/photos/NYfmWroJ71xD32e97 # This is the one I recorded the reference video with

https://goo.gl/photos/MZzrVvLaJpD5waAe8

Let me know if the pictures don't show up. While the middle one has a more complicated board, all three appear to use the same resistor package.

I still want to help, but I would prefer to not sacrifice one of these to the great clone gods in the sky. Is there any way I can help further without destroying one? I own a multimeter (even if I'm not that qualified to use it).

Reply 99 of 484, by keropi

User metadata
Rank l33t++
Rank
l33t++

MEGA thanks for the pics, sadly all of them use the same array thingie... it is interesting that the 2nd one has pads for ICs on the back side it seems as an alternative way to assemble one or something

If you own a multimeter you can try and measure resistance on the pins:

1i0ci.jpg

Keep one probe on pin1 and start measuring with the other probe the rest of the pins and see if you get the same reading with every pair. If that doesn't produce constant results then keep one probe at pin11 (the opposite end) and measure the rest of the pins.
It must be a common resistor network since there are 11 pins, that would be my guess. It's just a matter of finding the common pin.

🎵 🎧 PCMIDI MPU , OrpheusII , Action Rewind , Megacard and 🎶GoldLib soundcard website