Marmes wrote:

Real OPL3 is for purists, most people won't even notice the diference.

Most people won't notice the difference between DosBox and a real MS-DOS machine. That's not the people willing to pay what this card will cost.

Marmes wrote:

Then why choose this AWE64 chip? Why not AWE32?

AWE32, of which I own 3 models with OPL3, had bugs that affected the audio (search for "DMA audio clipping").

There is no perfect Sound Blaster, SB16, AWE32 or AWE64, they are either noisy, have fake OPL3 sound or bugs in the digitalized audio. AWE64 with OPL3 would be the perfect Sound Blaster 16 as it has the DSP 4.16 that fixes all the bugs in MIDI and FX sound.

Check this thread for more info: Sound Blaster 16 Bugs and Deficiencies Summary

dr_st wrote:

Why would you care about the AWE64 Legacy? Just for SPDIF?

Actually, the AWE32 has an internal SPDIF that you can connect to a bracket, like the AWE64 Gold.

why doesn't try OPL4 chip instead of OPL3 to give a reasonable step further but with compability's back?

Kaiser76 wrote:

why doesn't try OPL4 chip instead of OPL3 to give a reasonable step further but with compability's back?

The OPL4 is a larger, more sophisticated chip that offers little to nothing over a YMF289, unless you start adding SRAM and/or another wavetable ROM to go with it. I really don't think having two competing synthesisers of the era on the same card is one of their design goals. It would only serve to further inflate the cost of the final design, with very little to show for it.

Exactly.

640K!enough wrote:

carlostex wrote:

Cheapest options would be to opt for 72pin SIMM slot. If one of those RAM IC's die it's a pain in the ass to replace one.

Unless you get into low-end parts of questionable reliability, these sorts of sockets and headers are almost never cheap. For someone who doesn't already have the appropriate SIMM, this will likely be cheaper in the long run.

Can new RAM ICs be sourced? If reliability is a concern, what about mounting RAM expansion on a daughter board (a la AWE64), only properly documented for future maintainability?

gdjacobs wrote:

Can new RAM ICs be sourced? If reliability is a concern, what about mounting RAM expansion on a daughter board (a la AWE64), only properly documented for future maintainability?

I was referring to the reliability of cheap sockets, not the RAM IC's. As we found out with the ARGUS, receptacles and pin headers for daughter boards also quickly add to the cost.

While new FPM and EDO DRAM parts are still available, they are often not cheap, either. I suspect that they are harvesting their memory from existing SIMMs. As long as they run them through a memory test while they are still on the donor module, as well as once on the card, I don't think it should be a problem. Furthermore, since they are installing as much memory as the Creative chip can handle, there is no need for additional sockets or headers. I think it's a perfectly reasonable design choice.

how have I missed this thread?

Jebus.

thanks for the survey; i also mentioned that a gameport is essential for me. I'm using the cards to connect the joystick to play plane-games, will be difficult without one

Do we have to donate our awe64 to get one? I am currently using a Gold and would never destroy that but I have a number of plain ones. How much will it cost?

640K!enough wrote:

gdjacobs wrote:

Can new RAM ICs be sourced? If reliability is a concern, what about mounting RAM expansion on a daughter board (a la AWE64), only properly documented for future maintainability?

I was referring to the reliability of cheap sockets, not the RAM IC's. As we found out with the ARGUS, receptacles and pin headers for daughter boards also quickly add to the cost.

it might be an idea to make provisions on the board to mount a simm slot instead of on-board ram and just not populate one or the other... unless that's something you wouldn't want to do from an aesthetics point of view ^^

640K!enough wrote:

gdjacobs wrote:

Can new RAM ICs be sourced? If reliability is a concern, what about mounting RAM expansion on a daughter board (a la AWE64), only properly documented for future maintainability?

I was referring to the reliability of cheap sockets, not the RAM IC's. As we found out with the ARGUS, receptacles and pin headers for daughter boards also quickly add to the cost.

While new FPM and EDO DRAM parts are still available, they are often not cheap, either. I suspect that they are harvesting their memory from existing SIMMs. As long as they run them through a memory test while they are still on the donor module, as well as once on the card, I don't think it should be a problem. Furthermore, since they are installing as much memory as the Creative chip can handle, there is no need for additional sockets or headers. I think it's a perfectly reasonable design choice.

I'm surprised that 2.54mm pin headers contribute much to the BOM cost, although direct integration on the board is definitely going to be the cheapest option.

Indeed the lack of real OPL3 and WT header is one of the things that kept me from using them solely.

So...i'm still very interested in this project, i have a donor AWE64 value that i don't even care about so i'd rather have this.

I'm not crazy about the RCA jacks. Wouldn't be better to make use of the donor AWE64 brackets? Or you guys intend to make new brackets for the card?

carlostex wrote:

Indeed the lack of real OPL3 and WT header is one of the things that kept me from using them solely.

So...i'm still very interested in this project, i have a donor AWE64 value that i don't even care about so i'd rather have this.

I'm not crazy about the RCA jacks. Wouldn't be better to make use of the donor AWE64 brackets? Or you guys intend to make new brackets for the card?

what about SPID/F? is that a big deal in your opinion?

https://docs.google.com/forms/d/e/1FAIpQLSeu6 … 12goYQ/viewform

carlostex wrote:

Or you guys intend to make new brackets for the card?

This. 😀

Ohhhh those cards look nice 😁

H_Rush wrote:

carlostex wrote:

Or you guys intend to make new brackets for the card?

This. 😀

Just be aware that this is often not as cheap or easy as it seems, as many other PC card projects have found out. I wouldn't spend a heap on getting PCBs printed until you have finalised the bracket design and have secured it's manufacture.

While I would still be interested in this, even without the RCA plugs or optical output, I'd definitely give it a pass if it ships without a rear bracket.

I'm not sure about the name. Isn't the original awe64 the "legacy"? And yes spdif (minijack, rca or toslink), opl3, and rear bracket are essential to making this worthwhile. Although how will spdif work.....will a mixer take in the emu8k (which itself takes in opl3) and digital sound and output spdif? It would be good if reverb can still be applied to opl3 like with awe32. What about applying emu8k effects to waveblaster output? It would need to be sampled anyway for spdif so why not. That would really make this a killer card...

I did the survey but I'm not sure if the questions available can reflect my preferences.

OPL3, running without TSRs, without protected mode issues, and replacing CQM, if achieved, makes this card the Eiffel Tower, the Rachmaninoff's 3rd, the Michelangelo's Pietà of the Sound Blasters.

Wavetable header is the second best feature. I would prefer if both WT header and gameport can do simultaneous midi but I don't mind losing either.

From the existing AWE64 features I find very important to keep the gameport (for joysticks), Line-in, Line-out and CD-in. Something I'd like is an internal 4-pin line-out connector like the ones used by CDs so it can be routed internaly to the Aux/Line-in of another sound card.

The rest of the features, they are nice, but honestly, I don't really mind:

- Intelligent mode can be done by software, and the AWE64 should not have any hanging note bug.

- Loading soundfonts and emulate GM/GS/MT32 with the AWE drivers doesn't work in protected mode, and it takes ages to load a big one, which has to be done every single time you turn on the machine, and for soundfont use in Windows 9X there are better and cheaper solutions that can work alongside the AWE Legacy. Still, I understand the no-compromises approach of maxing out the RAM.

- 2nd Wavetable header in the back of the card will cause more compatibility issues for many people than is worth, I'd prefer that second MIDI Out to go over the gameport and if there is need for extra daughterboards you can use solutions from Serdashop.

- SPDIF, unless it provides an all-in-one solution for OPL3 + digitised sound + wavetable + CD-In + Line-in (and without CQM), it's going ot be a mess, if you just take the EMU8000 signal it has to be mixed anyway with analogue signals, if this is done, will there be a dedicated mixer for this SPDIF? If you just convert Line-out to digital, is not much easier to use an external device?

- RCA outputs, they look nice, but honestly, I'll just get an RCA-to-jack adaptor and keep using my analogue mixer, I would much prefer the gameport to be built in the card and have a more solid connector for the port I will connect and disconnect the most in the machine. Reusing the existing bracket also lowers the price of the card. Still, I understand the no-compromises approach and the wish for that AWE64 Gold aesthetics.

- Coaxial SPDIF: It would be nice if you can add support for a toslink adaptor like this. You could even use a combo jack and gain more space in the bracket.

I understand other people might have some preferences, this is only my opinion.

Schyz wrote:

- Loading soundfonts and emulate GM/GS/MT32 with the AWE drivers doesn't work in protected mode, and it takes ages to load a big one, which has to be done every single time you turn on the machine, and for soundfont use in Windows 9X there are better and cheaper solutions that can work alongside the AWE Legacy. Still, I understand the no-compromises approach of maxing out the RAM.

What are these options?