shock__ wrote:

Whoever can name the origin of the codename correctly first gets a free desoldered InterWave that was used in development (ideal for keychains or huge profits on eBay) 😜

The Dirty Harry movie seems too obvious?

That might still be correct (I've got to admit I haven't seen that one), but is not what it refers to 😉

Wasn't it something like Advanced Replica GUS ?

shock__ wrote:

Whoever can name the origin of the codename correctly first gets a free desoldered InterWave that was used in development (ideal for keychains or huge profits on eBay) 😜

Advanced Replica GUS - Member QBiN's idea (17th site of this post)... Later was shock's idea to change R form Replica to Reaserch because he didn't like "Replica" part...

5u3 suggested Argus - the mythological giant or Advanced Reverse-engineered Gravis UltraSound - and shock Agreed to it on 2015-04-02 00:35 because it sound better than AREGUS 😎 so that is my answer: Advanced Reverse-engineered GUS -

Batyra wrote:

shock__ wrote:

Whoever can name the origin of the codename correctly first gets a free desoldered InterWave that was used in development (ideal for keychains or huge profits on eBay) 😜

Advanced Replica GUS - Member QBiN's idea (17th site of this post)... Later was shock's ider to change R form Replica to Reaserch...

5u3 suggested Argus - the mythological giant or Advanced Reverse-engineered Gravis UltraSound - ang shock Agreed to it on 2015-04-02 00:35 because it sound better than AREGUS 😎

I think the code name shock__ is taking about is the one silk screened on his card, "Sudden Impact" not "ARGUS".

The Goat got it right.
ARGUS is the project name (advanced, reversed GUS) 😉

"Sudden Impact" the codename for the specific current layout. What is it a nod to?

the Goat wrote:

Batyra wrote:

shock__ wrote:

Whoever can name the origin of the codename correctly first gets a free desoldered InterWave that was used in development (ideal for keychains or huge profits on eBay) 😜

Advanced Replica GUS - Member QBiN's idea (17th site of this post)... Later was shock's ider to change R form Replica to Reaserch...

5u3 suggested Argus - the mythological giant or Advanced Reverse-engineered Gravis UltraSound - ang shock Agreed to it on 2015-04-02 00:35 because it sound better than AREGUS 😎

I think the code name shock__ is taking about is the one silk screened on his card, "Sudden Impact" not "ARGUS".

Hehe... in this cace my guess is that UFC game... because building ARGUS is quite a struggle 😀 (kidding)

shock__ wrote:

Elongating the pads sounds like a great option in my book, I didn't think of yet! That would certainly require less of an effort than the 42 pin socket (which would require a major re-routing for the SOJ-RAMs), while certainly cherishing the DIY-aspect of the project. You're speaking of SMD 1.27mm headers, right?

I was actually referring to these dual-row receptacles, simply because they are readily available at Digi-Key in 44-pin form (whereas they don't stock the 22-position, single-row part). If making the pads long enough to accomodate one of those isn't feasible, I guess we'll be able to live with the alignment headaches of getting two separate headers to mate correctly; I'll trust your judgement.

shock__ wrote:

About !RAS that signal is indeed connected to all 4 !RAS* signals on the SIMM. Which seems to work without any catastrophic consequences (to card and SIMM sticks) and reliable operation, even when used with incorrect types (that was one of the things Elianda and I tested a while back). Might still be worth checking into as I noticed the SOJ-RAMs often get quite hot (which might be normal ... or not).

At the time, he mentioned that at least one SIMM had chips that became warm (and the memory wasn't detected), which isn't something I'd consider normal for an old FPM or EDO SIMM. I'm left wondering if it wasn't a result of multiple banks trying to drive the bus at the same time. Of course, he could just have had a proprietary or 3.3V SIMM that wasn't happy in that socket.

I was under the impression that a separate RAS# signal was used to select the second rank of a dual-rank SIMM. Have a look at this old, archived Micron data sheet, if you're interested. It should be noted, also, that some oddball SIMMs still play other, non-standard games with the RAS# lines.

I didn't check, but do you have a pull-up resistor on the select lines for the SOJ DRAMs, so that they are not active when the bank jumpers are configured for SIMM use?

I have been forgetting to mention one thing that I found both interesting and disappointing: a few months ago, while looking over the ROMMaker source code, I had a theory that the SBOS chunk it generates has some limitations. Specifically, it only ever tries to map a certain number of instruments. I took a few minutes to test that behaviour before removing the official ROM, and my theory was correct. What that means is that, when playing music via SBOS emulation, only the standard GM instruments can be selected, so that any software that would normally use drum kits other than 0 (Standard), will revert to standard, and any melodic instruments in other banks (not that we have any of those now) would also revert to their standard GM counterparts. I may try to see if this can be fixed later, but not likely before the production run. Anyone interested in the additional features would have to generate a new set of ROM files and re-program the instrument ROM(s) when that's done.

Mega-Em is still a mystery, as it refused to run in the few seconds that I spent with it. I'll get back to it later.

Some of the clicking that we were hearing during MIDI playback was also related to the loop points. I have already fixed some of them, and will ultimately have to check every one of the 250 or so files that make up the PPLT set. It is beginning to look like it would be a good idea for me to eventually share a complete, fixed FFF/DAT bank of the PPLT set that can be used in building the ROMs for the cards. Alternatively, and because it offers greater flexibility, I can just distribute the few fixed PAT files, and let each person building the card source the rest of the patch set; comments/preferences?

640K!enough wrote:

shock__ wrote:

Elongating the pads sounds like a great option in my book, I didn't think of yet! That would certainly require less of an effort than the 42 pin socket (which would require a major re-routing for the SOJ-RAMs), while certainly cherishing the DIY-aspect of the project. You're speaking of SMD 1.27mm headers, right?

I was actually referring to these dual-row receptacles, simply because they are readily available at Digi-Key in 44-pin form (whereas they don't stock the 22-position, single-row part). If making the pads long enough to accomodate one of those isn't feasible, I guess we'll be able to live with the alignment headaches of getting two separate headers to mate correctly; I'll trust your judgement.

Alignment like this? Might take a bit to even out all collisions, but might be possible.

shock__ wrote:

Alignment like this? Might take a bit to even out all collisions, but might be possible.

With the dual-row receptacle, I figured you could just elongate the pads, so that there is only a 4 mm gap or so between the rows of pads for the SOP chip.

If you're set on the single-row receptacles, simply making the pads about 5 mm long should be sufficient.

I think I get it now 😉
I've elongated the pads to the outside (rather than centered to the landing areas of the TSOP44), which would enable one to either solder single row SMD headers or dual headers (on each side - adding redundancy but making acquisition easier, adding a single dual row header might still be an option, but don't get your hopes too high on that for now) while an alignment aid can be easily made from ordinary 2.54mm breadboard.
In case of using single row or the 'outer' dual row headers I can easily see enough space on the daughterboard to potentially add either more ROM or space for active components to allow out of circuit flashing via USB incorporating a microcontroller (that could be on the board directly) - I'll leave that up for someone else to design tho.

Once that's done I'll concentrate on some "behind the scenes" stuff. For example the current layout is tied to a messy version of the schematic (without values for the parts) while there's a tidier and more complete schematic linked to an older layout.

shock__ wrote:

What are the general feelings on that, since flash support might become a reality at one point in the future, but isn't at a point yet where I'd happily add it (circuit probably too complex to fit, adding redundancy on cards, cost factor probably being higher when installed on multiple cards). I'd like to hear some opinions on that - whether we should wait longer or are ready to go (feel free to reply here publically or via PM).

As for me, I'm 500% for the maximum funcionality and the money doesn't matter. However the overdue is a factor indeed, so I'd vote for 1 card in any condition ASAP and the other 2 with the possible future enhancements.

Batyra wrote:

For me logo size is OK - it's elegant somehow. Larger may me more "obvious"... I like how it is now.

Totally agree, the current design is OK.

However let me reiterate the older full checklist of other ideas that have been submitted by then but not yet implemented (just to check them twice):
* Make MONOIN and MONOOUT available as pin-headers on the board (marked as PCSPK_IN and PCSPK_OUT)
* Add swappable OPAMPs
* Consider a daughterboard based on the Music Quest MPU clone by keropi (in this case some jumpers/headers for the joystick DB15 should be integrated)

As soon as you proceed to the first batch, please consider once again my suggestion: does your plant provides for any silkscreen colors other than white? If yes, then I'd definitely suggest using yellow for all silkscreen inscriptions - just look how cool it is on the Roland SCB-55 and VideoLogic video cards!

And guys. The possibility for a 4Mb ROM will make the whole project 500% more interesting. Please keep up the good work! 😉

And another quick addition: I want a perfect ultimate GUS without any-any-any cost compensation soooooooo much that I'm even ready to offer some subsidies to everyone in order to eliminate any "cost efficient concerns".

^I agree with hard1k. For me, in this type of project, quality and funcionality etc are the biggest value.

Final proposal for ROM headers: https://i.imgur.com/CfALryK.jpg (Vias on the pads [symbolizing the headers] are just for visualisation), lines on all 4 extremes of the footprint are visual alignment aids when soldering the headers in place.
(Adding a single dual row header would result in too much fragmentation of the ground plane)

Let me re-iterate, yellow silkscreen (not offered by the plant) and an additional MIDI interface (just get keropi's card) won't happen. 4MB ROM might be hackable, that's one of the reasons the headers have been added (someone else will have to design that - even though once again, I seriously wonder how that's suddenly important after no one hacked something in that direction in the past 20 years of the GUS PnP being around). Op-Amps are swapable by means of smd reworking (remember it's a DIY project - targeted to be cheap, as the motivation was to design an accessible card for hobbyists who aren't willing to pay the $100+ for classic GUS cards).
Mono In/Out headers are possible but might cause additional noise (Mono In is tied to ground on the GUS PnP and Philips PCA761AW) and might cause catastrophic consequences when hooking up something incorrectly. Does the Interwave driver offer any mixing of that input/output anyways? Therefore addition of those is unlikely as well.

Since general interest to start the run now seems to be low, I'll likely do an 'open prototype' for anyone interested (likely anything from $90-$30 per card, depending on interest - hard to estimate). Design might be final or change in the future (I'll gladly include most of the denied features from above if you can present a hacked prototype that's working), with me now setting the latest point of the full production run to be in June 2018.

Either way ... time to start linking the schematic to the current layout 😀 I'll upload the schematic (as .jpegs) once that's been done.

I think I'm ok with the features as is. Like you say shock the idea was for a DIY GUS card for hobbyists at a sensible price.
It's been a while since it was discussed but how will the open prototype or later iterations come? Just the board with a interweave chip? And did the rights to the chip rom get sorted out as a royalty as part of the cost?

Cheers.

PCB + presoldered InterWave - if one wants I'd deliver the board without the InterWave soldered in place, included on it's own. I might have a small stock of angled SIMM connectors and SOJ-RAMs which I'd be willing to include for the price I paid while stock lasts (~$1.50 per connector, RAMs are salvaged therefore would be cheap)

Eye & I never replied back to my knowledge, even tho multiple people reached out. They probably don't own the sole rights to the ROM contents so they can't make a decision on parts they don't own.
640K!enough is currently building a replacement ROM that shouldn't have any copyright issues (or might provide a tool which can generate a ROM from a soundbank that can be easily obtained). Alternatively one could always warez the ROM contents read out from another InterWave card (or wait til someone uploads them on the internet).

Since there has been some discussion of the instrument sets lately, I thought I would provide a few more samples, for the sake of comparison. In this case, I have recorded the same MIDI file, with all of the same settings, recording and volume levels. I did not edit the resulting files in any way, except to trim a few seconds of silence at either end; not even the "Normalize" function was used.

The first recording was the standard TECHNO78.MID, as shipped with every UltraSound card, played with the official GSFULL4M.FFF (a.k.a. amd_4m, a.k.a. Voice Crystal 4 MiB GM/GS set version 3.0, which was awarded the Fat Labs Seal of Approval). The second is the work-in-progress set, based on the PPLT 1.61 patch set.

Any comments or comparisons would be welcome. If you are interested in comparative recordings of a particular MIDI file, feel free to ask.

shock__ wrote:

Eye & I never replied back to my knowledge, even tho multiple people reached out. They probably don't own the sole rights to the ROM contents so they can't make a decision on parts they don't own.

Given the progress so far, if the sentiment is still that the Eye & I set is superior, we could make one last attempt to contact them. I think, perhaps, we were asking the wrong questions. As shock__ points out, they don't own the entirety of the ROM image that shipped on the board, and are probably not even in a position to provide us with a ROM image. If, instead, we were to ask about pricing for permission to use their version 4.1, 4 MiB set (which is for sale on their Web site in SF2 format for $14.95) to produce our own ROM image, for inclusion on a run of up to 500 cards (based on the number of InterWave ICs shock__ had), we might get a different answer.

Of course, it could just be that such a small community project with a small budget isn't worth their time. Opinions, comments, could we possibly pool an amount and see if they'd prefer to do a lump licensing deal? If necessary, we could even produce and ship assembled, pre-programmed, 100% legal ROM daughterboards (no sines bank, as that is a separate copyright can of worms). What are the thoughts on this?

As it stands, with some optimisation, it looks like the PPLT set, once complete, might fit into an 8 MiB ROM; possibly 4 MiB at 8-bit sampling resolution.

As promised:
current version of the schematic. If you have any issues reading something, feel free to give me a ring (I know page 1 is a mess of labels).
Pulldowns for the SOJ-RAMs have not been implemented yet, I'll actually test those on one of the prototypes within the next few days first.