VOGONS


First post, by anthony

User metadata
Rank Member
Rank
Member

hi
does anyone will be interested in awe64 clone with such features as 72 pin simm, midi header, spdif out and may be more?
looks like card will not be cheap, considering hard to get creative chips and 4 layer design
and very much looks like creative will phohibit it distribution (don't know why, but i asked them about such type of clone. don't get answer yet)

upd. how it looks:
awe64.jpg

Last edited by anthony on 2017-05-22, 11:25. Edited 1 time in total.

Reply 1 of 25, by Jorpho

User metadata
Rank l33t++
Rank
l33t++

Have you ever designed anything remotely like this before? Even with off-the-shelf parts the price would likely be astronomical.

anthony wrote:

and very much looks like creative will phohibit it distribution (don't know why, but i asked them about such type of clone. don't get answer yet)

The AWE32 emulation in PCem requires a ROM dump from an AWE32, if I'm not mistaken. Any such clone card would also probably require a ROM dump from an original card to ensure compatibility.

Unless you have something to indicate significant progress I doubt anyone at Creative would take you seriously.

Reply 3 of 25, by anthony

User metadata
Rank Member
Rank
Member

Well, I see misunderstanding there. I'm not planning chip design. So no fancy hybrids like sb pro on awe64 are intended. I'm talking about using existing ct8912 chip with all it's pro's and con's. Not a big deal to create pcb with new features

Reply 4 of 25, by havli

User metadata
Rank Oldbie
Rank
Oldbie

72-pin SIMM compatible AWE64 sounds interestig... how much interesting depends on the price of course. 😊

HW museum.cz - my collection of PC hardware

Reply 5 of 25, by cyclone3d

User metadata
Rank l33t++
Rank
l33t++

But what would be the point of just having a 72-pin SIMM compatible AWE64?

You could just manufacture the SIMMCONN adapter for the regular AWE64 cards.

AWE64 cards are pretty cheap and I've already converted the plans for the SIMMCONN to a newer format.

See here:
Simmconn AWE64 Adapter

Adding SPDIF to an existing card is also possible and not really complicated.

However, the AWE64-Gold is apparently the only one that will output both MIDI and sound through SPDIF. The others only output MIDI through SPDIF. It has to do with the mixer chip used.

See here:
https://groups.google.com/forum/#!topic/rec.m … nth/mhigB5IHypc

Yamaha modified setupds and drivers
Yamaha XG repository
YMF7x4 Guide
Aopen AW744L II SB-LINK

Reply 6 of 25, by anthony

User metadata
Rank Member
Rank
Member

Point is: I hate ugly solutions (like creative offer to memory enhancement on awe64), so i decide to build for myself card with desired options like memory and midi header. Card is working now, i'm happy, nothing less, nothing more.

About spdif out, I wish to test it, but have no one device with RCA digital audio input. Will try to test it anyway

Reply 7 of 25, by SRQ

User metadata
Rank Member
Rank
Member
anthony wrote:

Point is: I hate ugly solutions (like creative offer to memory enhancement on awe64), so i decide to build for myself card with desired options like memory and midi header. Card is working now, i'm happy, nothing less, nothing more.

About spdif out, I wish to test it, but have no one device with RCA digital audio input. Will try to test it anyway

You actually accomplished this??? Prooooof.

Reply 10 of 25, by Jepael

User metadata
Rank Oldbie
Rank
Oldbie
anthony wrote:

photo uploaded.

i'll show analog audio part, if diy audio ethusiasts here exists, i'd like to see any advices on improvements over schemetic. i'm pretty sure there is plenty space for it

I can take a look if you want, just post a link to schematics.

BTW, why not use optical Toslink connector for the SPDIF?
Also, is the coaxial SPDIF output properly buffered/isolated so it's not TTL levels?

Reply 11 of 25, by anthony

User metadata
Rank Member
Rank
Member

here is schematic. tried to replace ka3402 to opa4132. but overal volume became to low

spdif is not ttl

Attachments

  • Filename
    awe64.pdf
    File size
    196.86 KiB
    Downloads
    293 downloads
    File comment
    Schematic
    File license
    Fair use/fair dealing exception

Reply 12 of 25, by gdjacobs

User metadata
Rank l33t++
Rank
l33t++

Also, being able to use the EMU chip via a hardware MPU-401 interface rather than the TSR would be excellent.

All hail the Great Capacitor Brand Finder

Reply 13 of 25, by Jepael

User metadata
Rank Oldbie
Rank
Oldbie
anthony wrote:

here is schematic. tried to replace ka3402 to opa4132. but overal volume became to low

spdif is not ttl

Holy smoke, you reverse-engineered and re-engineered the whole thing with signal names and all?
That is a very impressive task, just for everyone to know.

Yeah, replacing bipolar opamps directly with fet opamps might require some other changes as well, because in general, bipolars can accept and drive signals of higher amplitude with same supply voltage than fet opamps.

SPDIF coax output is acceptable. Using multiple buffers in parallel to boost output current is not the optimal solution, but it is cheap, it is a common chip and the output should work fine in home use nevertheless. As a bonus it could be isolated with a transformer, but most people don't need it.

By the way, what is the purpose of that 405x analog switch circuitry, some kind of output gain selection or output filter bandwidth selection?

Reply 14 of 25, by anthony

User metadata
Rank Member
Rank
Member

exact purpose of analog switch is not clear enough for me. as i remember, back in the days, some people replaced this part with external commutator. as well as ka3402 to opa4132. but knowledge is lost.

quite a lot of ct8920 pins functioning in analog part is unknown, i named they as bypass in/out. and a lot of nc pins too. wish to know more about it, but looks like it's impossible until chip re-engeneering

Reply 15 of 25, by anthony

User metadata
Rank Member
Rank
Member
gdjacobs wrote:

Also, being able to use the EMU chip via a hardware MPU-401 interface rather than the TSR would be excellent.

it's possible by placing programmable logic on isa bus before ct8920, but requre cpld/fpga programming. i'm able to place it on right place, but not able to programm it, so far.

Reply 16 of 25, by Jepael

User metadata
Rank Oldbie
Rank
Oldbie
anthony wrote:

exact purpose of analog switch is not clear enough for me. as i remember, back in the days, some people replaced this part with external commutator. as well as ka3402 to opa4132. but knowledge is lost.

quite a lot of ct8920 pins functioning in analog part is unknown, i named they as bypass in/out. and a lot of nc pins too. wish to know more about it, but looks like it's impossible until chip re-engeneering

Hmm I can try to figure out the analog switch purpose from the schematic.

The caps are both bypasses for reference voltages and AC coupling caps between different stages that might require different DC bias. Also some might be caps for filters. If you have an oscilloscope, you might be able to determine something how it works by looking if you see audio on capacitor, and what audio, and at what amplitude, and does it change when using the mixer.

Edit:
It looks like the analog switch feeds an inverse proportion of one channel to another. Could this be the "3D stereo enhancement" effect? At least that can be solved by using mixerset to change the effect on/off while monitoring switch control signal.

Reply 17 of 25, by ab0tj

User metadata
Rank Member
Rank
Member
anthony wrote:

it's possible by placing programmable logic on isa bus before ct8920, but requre cpld/fpga programming. i'm able to place it on right place, but not able to programm it, so far.

That is a very interesting idea... Use a CPLD to trap ISA bus transfers before they go into the AWE64 chip, and send them off to different chips. Perhaps a MPU401 intelligent mode interface, and an OPL3 chip?

Reply 18 of 25, by anthony

User metadata
Rank Member
Rank
Member

switch chip has quite interesting behaviour. by default control signal is low, but after init in dos it became high (with significant background noise). in windows it always low, even if initing in dos was done. and background noise is barely hearing. will try to connect it to ground and see what will happend in dos

Perhaps a MPU401 intelligent mode interface, and an OPL3 chip?

yep, who able to rule bus, able to rule whole card )

Reply 19 of 25, by anthony

User metadata
Rank Member
Rank
Member

could be anyone so kind to tell me, what format used to store instruments in base rom for awe card? rom in attachment. and is there any possibility to convert this file http://www.vogonsdrivers.com/getfile.php?file … menustate=56,53 , for instance, in rom format?

Filename
base_rom.rar
File size
857.12 KiB
Downloads
180 downloads
File comment
awe rom
File license
Fair use/fair dealing exception