GUS Daily Digest Wed, 26 Jan 94 0:07 Volume 10: Issue 26 Today's Topics: DMA one more time GUS Daily Digest V10 #20 GUS Daily Digest V10 #25 (5 msgs) GUS DRAM problems Howling dog stuff and Gravis offer MIDI functionality with GUS MIDI Patch OPTi chipset explanation PC200mkII Keyboard SQ4 and PQ3 Voice Assist Standard Info: - Meta-info about the GUS can be found at the end of the Digest. - Before you ask a question, please READ THE FAQ. ---------------------------------------------------------------------- Date: Tue, 25 Jan 1994 14:49:35 -0500 (EST) From: "K.S. Holly" Subject: DMA one more time OK, one last question..... I have the UMC chipset that doesn't do 16 bit DMA. Even though not having 16 bit DMA may not be a big deal, I am going to go back to the company that sold me the computer because as far as I am concerned, this motherboard is NOT 100% IBM compatible so therefore I did not get what I paid for. Now my question: Opti must have admitted to their screw up since they put out a fix for it (or did the fix come from somewhere else?). Has UMC publishes anything that verifies the problem? I know that when I go back to the computer store, they will not believe me that there is a problem until I prove it to them. If I could get some documentation or some other info it would help my cause immensely. Any help will be much appreciated. Thanks Kevin ------------------------------ Date: Tue, 25 Jan 94 11:32:59 PST From: Scott Bobo Subject: Re: GUS Daily Digest V10 #20 Some folks are looking for GUS DRAM and are having trouble locating compatable parts at a fair price. How about $4 per part and I pay shipping? E-mail if interested. These are TI 100ns parts (the same as what was shipped on my GUS). I don't have many, so you may want to act fast. scott@mothra.rose.hp.com ------------------------------ Date: Tue, 25 Jan 94 00:43:38 PST From: chrism@huey.nic.bc.ca (Chris Mackay) Subject: Re: GUS Daily Digest V10 #25 > After a reboot, the first started working, but the latter three > banks always fail. SETGUS DRAM diagnostics start (!?) at 351K and > finish with a message that "DRAM FAILED (351K)" or some such. I've > tried shuffling DMAs and IRQs, to no avail. The DRAMs I used are > Sanyo 256x4 70ns, model number LC324256BP-70. Probably a stupid question, but did you plug the chips in the right way?? When I first got my GUS, I ripped some 256x4 chips off an old EMS board I had and plugged 'em in. Anyways, I booted up and everything beyond the standard 256k chips failed. So I took the GUS back out, and found that I had put them in upside down. I'm lucky I didn't fry the GUS or the chips(though the chips were real hot!) ------------------------------ Date: Tue, 25 Jan 94 11:40:40 EST From: bprince@mcs.capital.edu (Brian Prince) Subject: Re: GUS Daily Digest V10 #25 ------------------------------ Date: Tue, 25 Jan 1994 12:38:36 EST From: dth2@Lehigh.EDU (Dave Horoschak) Subject: Re: GUS Daily Digest V10 #25 Is there any way to convert PAT files into WAV files? I am interested because I would like to use my GUS to sample sounds and then put them up on my Maui card. Or, is there any documentation on the make-up of PAT files? Last, does anyone know if there are any plans on upgrading the max amount of RAM on the GUS? Thanks a lot. Later... - Dave ------------------------------------------------------------------------------ Dave Horoschak -aka- dth2@lehigh.edu Chief engineer of WLVR 91.3 FM. Treasurer & Project Engineer - Lehigh Formula SAE ------------------------------------------------------------------------------ ------------------------------ Date: Tue, 25 Jan 1994 12:52:53 -0500 (EST) From: Phat H Tran Subject: Re: GUS Daily Digest V10 #25 > Date: Mon, 24 Jan 1994 07:31:20 -0800 (PST) > From: mikebat@netcom.com (Mike Batchelor) > Subject: Re: DOS4GW and QEMM's Stealth > > a one-byte patch for DV-compatibility, though). I am going to try Shadow > Caster in DV today, now that I have 32 bit AIL drivers for it, and the GUS > offloads sound overhead from the CPU, so it ought to fly pretty good! :) No, it won't fly any better than with any other souncard. The AIL interface restricts the GUS to a synth and streaming DAC. Music is played on the synth, while digital sound effects are sent to the DAC as they are played. There are no provisions to take advantage of the GUS' hardware properly. > Date: Mon, 24 Jan 1994 17:45:45 -0500 (EST) > From: Damian Frank > Subject: GUS DRAM problems > > I just got my GUS a week ago, and am quite happy with it. I just today > received my eagerly awaited shipment of DRAMS, and as soon as I could, > plugged them in, with the usual precautions (grounding etc). Unfortunately, > as soon as I could boot and run GUSDRAM, it showed the banks as there ("good > good), all of them (even the first!) failed rapidly. Neither GUSDRAM nor the DRAM diagnostics in Setgus are 100% reliable. They will sometimes erringly report DRAMs as bad which are quite okay. If you don't hear clicks or staticky noises in your patches, then you should be able to use your DRAMs with confidence. > Date: Mon, 24 Jan 1994 14:54:15 +0200 (IST) > From: Guy Tal > Subject: Playmidi vs. Media-Player > > One of you mentioned a while back that he noticed a difference between > the sound as played by Playmidi and the one he gets while running > Windows. > I have encountered a bigger problem - I have downloaded the file > beetho27.zip from one of the ftp sites, and while it plays just fine > using Playmidi, it won't play at all using Media Player on windows (it > does work with Recording Session, however). Any ideas ? Can you upload beetho27.zip to Epas (or mail it to me?) You're not the only one to report problems with it and I want to check it out to see what's so special about it. Phat. ------------------------------ Date: Tue, 25 Jan 94 12:43:35 MST From: Juggler Subject: Re: GUS Daily Digest V10 #25 > From: Damian Frank > Subject: GUS DRAM problems > > After a reboot, the first started working, but the latter three banks always > fail. SETGUS DRAM diagnostics start (!?) at 351K and finish with a message > that "DRAM FAILED (351K)" or some such. I've tried shuffling DMAs and IRQs, > to no avail. The DRAMs I used are Sanyo 256x4 70ns, model number > LC324256BP-70. I got those EXACT same chips the other day myself and got the EXACT same errors!! I took my chips back and got some 80ns, different brand, and popped them in with NO problems. The idiot at the computer store wanted me to bring in my GUS so he could put them in and show me they worked. I told him these chips are compatible, cuz I tried EVERYTHING with them to no avail. You need to send those chips back and get some new ones. I don't think they are GUS compatible... --- - = Juggler = - [ IRC: jugular - vrave: jug ] IH23@utep.bitnet IH23%utep@utepvm.ep.utexas.edu --------------------------------------------------------------------- "Sometimes I wonder about the charmic implications of our actions" -Farmner Vincent in "Hell Motel" --------------------------------------------------------------------- ------------------------------ Date: Tue, 25 Jan 94 13:49:57 MST From: tschmidt@pefs1.micron.com (Tom Schmidt) Subject: Re: GUS DRAM problems In reply to Damian Franks posting on Mon, 24 Jan 1994 17:45:45 -0500 (EST), : From: Damian Frank : Subject: GUS DRAM problems : : I just got my GUS a week ago, and am quite happy with it. I just today : received my eagerly awaited shipment of DRAMS, and as soon as I could, : plugged them in, with the usual precautions (grounding etc). Unfortunately, : as soon as I could boot and run GUSDRAM, it showed the banks as there ("good : good), all of them (even the first!) failed rapidly. : : After a reboot, the first started working, but the latter three banks always : fail. SETGUS DRAM diagnostics start (!?) at 351K and finish with a message : that "DRAM FAILED (351K)" or some such. I've tried shuffling DMAs and IRQs, : to no avail. The DRAMs I used are Sanyo 256x4 70ns, model number : LC324256BP-70. : : What I'd like to know is whether there's something I'm doing wrong, or if I : have to go and wrestle with the memory company and get some replacements. : Any help would be greatly appreciated. This problem has been posted several times before. The Sanyo chips simply do not work with the GUS. Apparently, they are not JEDEC standard 256Kx4 DRAMs. Try to get them exchanged for one of the vendors listed below. Here is my 256Kx4 DRAM vendor cross reference that I have posted before. It has been updated some, and I have requested that it be placed in the FAQ. ------------ This is a list of vendors that make JEDEC standard 256Kx4 Fast Pagemode DRAMs. I am not saying that all of these will work in the GUS because the timing and power consumption of every vendors design is slightly different. Nor have I personally tested each of these vendors in the GUS. If anyone has pros or cons about any of these vendors, or knows of additional vendors parts that work with the GUS, please post them to the digest or E-mail me so that we can update this list and get it into the FAQ. This is only a cross reference list that I put together with a little research into various DRAM data books and the IC Master 1993 master selection guide. For simplicity, I am leaving the speed grade designation off (a -10, -8, -7, or -6 extention at the end of the part number). Manufacturer Part Numbers of 256Kx4 Fast Pagemode DRAMs -------------------- ------------------------------------------ Micron Semiconductor MT4C4256 AT&T M441024 EDI EDI44256C Fujitsu MB81C4256P Goldstar GM71C4256A Hitachi HM514256AP Hyundai HY51C4256 or HY53C4256 Intel 21014 Mitsubishi M5M44C256BP Mosel-Vitelic V53C104 Motorola MCM514256AP NEC uPD424256C OKI MSM514256 Panasonic MN41C4256 Samsung KM44C256BP Sharp LH64256 Siemens HYB514256 Texas Instruments TMS44C256 Toshiba TC514256BP You may have noticed that the Sanyo part (LC324256BP) and the NMB part (AAA1M304P) are not included in this list. This was done because several GUS users have reported problems in the digest when trying to use them to upgrade the GUS. I also could not find the Sanyo part in the IC Master or other data books, so I am not exactly sure whether or not it is an JEDEC standard 256Kx4 Fast Pagemode DRAM. I repeat: I am not recommending any one part type or vendor. _____ ___ Tom L. Schmidt | | / \ Manager - Component Characterization | | \___ Micron Semiconductor, Inc. | | \ 2805 E. Columbia Rd. Mail Stop 376 | |____\___/ Boise, ID 83709-9698 tschmidt@vax.micron.com or tschmidt@pefs1.micron.com Note: These opinions are mine and are not necessary shared by Micron P.S. If anyone has been able to get the Sanyo or NMB parts to work, please let me know. Also, if you were not able to return them for one reason or another and have not use for them, please send them to me and I will test them to see what they really are and I will post the results. ------------------------------ Date: Tue, 25 Jan 94 13:03:27 CST From: dwaldron@ccmail.rrddts.donnelley.com Subject: Howling dog stuff and Gravis offer I've been out of this group for awhile. I'm thinking about ordering Power Chords upgrade or pro, also Mr. Drumstix Music Studio. Anyone care to comment on these? Its an old offer, does anyone have an email address for Howling Dog Systems? Also, from reading the FAQ, I can order a 1 meg upgrade through Gravis, does there exist an order form that I can download and print? Finally, I missed the DOOM craze, is there a way to get better sound effects out of doom 1.1? Must I wait for doom 1.2? Doug Waldron Dazed and Confused Incorporated ------------------------------ Date: Tue, 25 Jan 1994 15:18:29 GMT From: Clarke Brunt Subject: Re: MIDI functionality with GUS >I am in the process of buying a MIDI-adapter and a Roland PC-200mkII >keyboard for my 1Mb GUS. The keyboard has velocity-sensitive keys, >pitch-bender/modulation key, reverb and chorus level key plus some >other neat features. Will these features be supported by the GUS >Windows driver? (I have the new driver installed). It doesn't do reverb or chorus. I believe that these are 'signal processing' type effects, and the GUS does not have a signal processor. It does do velocity, pitch bend, and modulation though (and sustain as well). ------------------------------ Date: Tue, 25 Jan 1994 22:09:41 -0600 (CST) From: Fitzgerald Jr Steele Subject: MIDI Patch Hello. I just got the new Windows Drivers (better late than never) as well as the Xwing sound files from EPAS (xwngsnds.zip). I found out that I am missing the Midi Patch tremstr.pat. It doesn't seem to be in any of my installation disks (I checked v1.21 2.06L and 2.06a). I thought I remember seeing this file on epas but it's not there, nor is it in any of the mirrors. Does anyone know where I can get this patch? Are there any Midi Libraries on the 'Net? Incidently, should this have been posted to the music digest? I wasn't sure. --------------------==========*********==========---------------------- Jerry Steele "Give me a lever, and a place to stand fjsteele@icaen.uiowa.edu and I will move the world" fsteele@umaxc.weeg.uiowa.edu -- Archemedes --------------==============*************==============---------------- ------------------------------ Date: Tue, 25 Jan 1994 11:22:09 -0800 (PST) From: david.roberts@amd.com (Dave Roberts) Subject: OPTi chipset explanation > Date: Sun, 23 Jan 1994 21:02:08 -0400 (EDT) > From: Roger Frederi Clark > Subject: Chipsets yet again... > > I've seen some posts recently about this subject, and I have a couple of > questions that some guru might be able to answer out there. It's explained below. > Ever since I purchased the GUS(15 months ago) and found out about 8 vs. 16 bit > DMA, I've been trying to get a 16-bit DMA to work because it's purportedly > faster(yes, I'm a speed-freak). You shouldn't notice the difference unless you're recording to disk. If you just playback, the 8-bit DMA channels are just fine. > Yesterday I tried once again. In SETGUS, I change DMA from 3 to 5,6,or 7, > go to test, and "On board parity error". I know about the problem with > the Opti chipsets, but my chips advertise UMC on them. Specifically, part > numbers UM82C481AF, UM82C482AF, and UM82C206AF. On a post a couple of > days ago, I noticed the "82C206" part of the last chip number as being > suspect. Does that mean 16-bit DMA is out for me too even without Opti? > If so, then explain this one. Okay, people have been very confused about the whole "OPTi chipset problem" for a long time. As explained below, the problem is not OPTi's at all, but does involve them. Here's the real scoop as near as I can tell. This information comes from the person sitting across the aisle from me here at work. He used to work at OPTi, so this is about as official as you can get. The chip in question is known as a 82C206. This chip is standard in the chipset business and compatible versions of it are made by a number of companies. For those of you who care, this chip contains all the ISA-bus standard motherboard peripherals (e.g., DMA controller, interrupt controller, real-time clock, etc.). Now, a version of this chip made by UMC in Taiwan contained (still contains?) a bug in the DMA controller. In particular, at the end of a DMA cycle a signal on the ISA bus is not driven to its inactive state, but rather allowed to float. As a result, the timing specification for that signal is not met. This problem only occurs when 16-bit DMA is used, not 8-bit. There are three ways around this problem: 1. Make the GUS use 8-bit DMA rather than 16-bit. You probably won't see any problems with this. 8-bit DMA is perfectly capable of keeping up with the GUS. 16-bit DMA is more efficient, however, so if you are also doing a lot of other stuff on the bus, this can help (e.g., recording to disk). For simple playback, no problem. 2. Solder a resistor onto your motherboard to pull the floating signal line to its inactive state faster. This will cause the signal to meet its timing spec. Requires that you know how to drive a soldering iron and feel comfortable doing very minor surgery on your motherboard. If this is a name-brand PC and the warranty is still valid, send it back and have them send you another one. The file gus0013.zip in the GUS archives contains specific information about the problem and tells you where to solder the resistor. Gus0013.zip can be found on theoris.rz.uni-konstanz.de in /pub/sound/gus/gravis/patch. Note that it is not on archive.epas.utoronto.ca (because only newer GUS files are kept there) or wuarchive.wustl.edu (because wuarchive crashed as lost a lot of archived information -- all the GUS stuff, as near as I can tell). 3. Buy a new motherboard. If you just got your motherboard, tell the manufacturer that it's defective (check for NMI being disabled, while you're at it -- might as well kill two birds with one stone) and have them give you a new one. Now, some details about OPTi, UMC, and how to tell if you have a problem. Now, I said above that the problem is really UMC's, not OPTi's. Why is OPTi's name connected with the problem. Well, OPTi buys 82C206s from a couple of people, UMC among them. Those relabeled UMC parts are the ones that cause the problem, and unfortunately most people who started to notice the situation had UMC parts relabeled as OPTi rather than UMC. As a result, this started getting called the "OPTi problem" in GUS circles. Now, there have been a number of people who also report that their OPTi chipset has no problem. This is consistent with the above paragraph. As I said, OPTi buys '206s from a number of people. Those from other manufacturers do not have the problem and work just fine. Now, the big question is, how do you tell what you have, an OPTi/UMC '206 or an OPTi/other '206? To tell the difference, look at the top of the '206 chip on your motherboard and note where it was manufactured. Chances are it will be one of two places, Taiwan or Korea. The ones from Taiwan are from UMC and *may* have the problem (if UMC hasn't already fixed it -- I don't know if they ever have). If you have one from Korea, you're golden and should have no problems (those are made by Samsung). Okay, everybody got all that? Dave Roberts Advanced Micro Devices, Inc. I/O and Network Products Division david.roberts@amd.com ------------------------------ Date: Tue, 25 Jan 94 08:46:20 EST From: oehler@ispvax.dnet.ge.com Subject: Re: PC200mkII Keyboard Good Morning all, Tor Fargi, God of Thunder (toref@dcs.gla.ac.uk) asks whether the Roland PC200mkII modulation, pith bending and chorus controllers would work under windows. Well, I recently aquired this keyboard and have played a bit with it. Here's what I know: Pitch Bending - Supported, works like a champ Modulation - Supported, work great, too Chorus - Haven't tried it yet. I will guess, however, that the chorus feature is not supported by the windows drivers because the GUS has no chorus effect in hardware (like the WaveBlaster does). I'll give it a shot tonight and post again tomorrow with the results. Have a day, Pete Oehler ------------------------------ Date: Tue, 25 Jan 1994 13:35:29 ADT From: "Craig Galbraith" Subject: SQ4 and PQ3 Quick question... I can't seem to get the gus drivers from GUSDRV.EXE for sierra games to work with Space Quest IV and Police Quest III. I keep getting an error that it can't find "patch.004" or "4.pat". I used uminstal.exe like it said and copied over the proper files...maybe these games aren't new enough to use with the drivers...I dunno. Thanks. . Craig Galbraith j5oa@unbsj.ca are you phishsperienced? ------------------------------ Date: Tue, 25 Jan 1994 21:42:33 -0500 From: David A Denis Subject: Voice Assist I borrowed Voice Assist from my roomate, and get can't get it to work. It is for wINDOWs so I thought it would work fine. But, I click on it, and it says SB card neeeded. I try to install th e SB drivers adn it won't work at 220 and 7 IRQ. Its like it won't rwecognize it at all. Any ideas on how to get it to run? Sorry for the typos, I can't backspace for some reason. Having it fixed tomorrow. Dave ------------------------------ End of GUS Daily Digest V10 #26 *******************************