kiki@uhunix1.uhcc.Hawaii.Edu (J. Wine) (05/15/91)
These interesting chips were found in my STe: [From motherboard labeled GX-211 VO 4990] pins package label * = soldered ------------------------------------------------------------------------------ a) 144 quad flat pack C300589-001, c. 1988, -19, J9030 * b) 84 " plcc C301712-001, c. 1990, -31 I60608-01, 9029EBB c) 68 " plcc C101643, c. 1989, -31 I60611-01, 9033ETB d) 68 " plcc MC68000FN8, 2C91E, QEYX9025 e) 48 dip MK68901N-04, D9028 * f) 40 dip C398739-001A, c. 1990, -31 I60511-01, 9031PCA * g) 40 dip Yamaha, YM2149F, 9039 GAGG The chips with a 'C' prefix (a, b, c, and f) are designed by Atari. Dr. Tsuji of Waseda Univ. informed me that they were GALs (gate-array logic) and were fabricated in Japan. The WD 1772 floppy controller is located under the disk drive (Sony drive w/ 5V only power input), so there is very little room for an alternate board to plug into the 28-pin socket. The Yamaha sound chip is socketed for some reason, while the 2 TOS ROMs are soldered. Information about the many of the Atari STe chips is not available to the gen- eral public, so I have some questions that might be answered here. 1) The old MMU and GLUE are supposedly integrated into the MLUE. Since the former chips had 68 pins apiece, what, if any, enhancements does the MLUE have? Since the new MLUE has about 12 extra pins (144 - (2*68) + unused ROM selects), did the designers create an additional memory multiplex add- ress line so that 4 MB SIMMs could be used? Who has info on the GAL chips? 2) Reading/writing the area of memory between $400000 (end of 4MB DRAM memory) and $CFFFFF causes a bus error, however addressing $D00000 (start of 13MB) to $DFFFFF seems to be OK. Is this memory protection specfic to the STe? Although a local dealer claimed the new Mega STe has the same addressing scheme, it would seem that it would support the entire 16MB space. So can anyone verify that the Mega STe uses a different MLUE and has an extended addressing capability? 3) So far people have informed me that the Atari developer documentation isn't a good value, even at $125. What are the alternatives??? Supposedly TOS was developed in Germany and the chips made in Japan, so all the docs lack any cohesion. What will happen when Atari brings out a system designed in Britain by an Indian engineer with a floppy chip from Israel, monitors from Korea and Netherlands, custom chips from Japan, German TOS, a Unix GUI from France and manufactures it in Taiwan. Then will it be voted number one again in Hong Kong? I received the developer kit registration form and it requires that all infor- mation received from Atari not be disclosed. In fact, it prohibits you from disclosing the fact that you are prohibited from disclosing facts. If you disagree with this policy, please send mail to Bob Rehbock, Director of Tech- nical Services, c/o Atari Corp, 1196 Borregas Ave, Sunnyvale, CA 94088-3427. I can understand putting a restriction on the divulgence of Atari's prospect- ive products and plans, but all their current systems should have complete and open documentation. If Atari dominated the industry, then their stance might be understandable, but the trend is toward open systems and I feel it would be to their benefit if they would disseminate their documentation without excess- ive restrictions. Jack P.S. Let's go TEAM Motorola!!!
ekrimen@ecst.csuchico.edu (Ed Krimen) (05/15/91)
In article <13023@uhccux.uhcc.Hawaii.Edu> kiki@uhunix1.uhcc.Hawaii.Edu (J. Wine) writes: >fabricated in Japan. The WD 1772 floppy controller is located under the disk >drive (Sony drive w/ 5V only power input), so there is very little room for >an alternate board to plug into the 28-pin socket. The Yamaha sound chip is >socketed for some reason, while the 2 TOS ROMs are soldered. > Just to compare and contrast, when I opened up my STe to upgrade the RAM, I noticed that I had an Epson drive and the 2 TOS ROMs were *socketed*, under the drive. > >2) Reading/writing the area of memory between $400000 (end of 4MB DRAM memory) > and $CFFFFF causes a bus error, however addressing $D00000 (start of 13MB) > to $DFFFFF seems to be OK. Is this memory protection specfic to the STe? > Although a local dealer claimed the new Mega STe has the same addressing > scheme, it would seem that it would support the entire 16MB space. So can > anyone verify that the Mega STe uses a different MLUE and has an extended > addressing capability? > I read in one of the on-line magazines a couple of months ago that Jim Allen had developed a 14meg upgrade for the STe and a 10meg upgrade for the MegaSTe. I couldn't understand the difference, since I would expect the MegaSTe to be able to be upgraded with more RAM, but I haven't heard anything since. Perhaps these are related. -- ||| Ed Krimen [ekrimen@ecst.csuchico.edu or al661@cleveland.freenet.edu] ||| Video Production Major, California State University, Chico / | \ SysOp, Fuji BBS: 916-894-1261 TWO WEEKS UNTIL GRADUATION!!
ytsuji@wucc.waseda.ac.jp (Y.Tsuji) (05/15/91)
I don't think I said all chips inside STe were made in Japan. The famous new DMA chips are made in Taiwan. The GALs in there are custom made chips with relatively few transistors. Not to be confused with ordinary GALs which can be treated by hobbyists like EPROMs. Y. Tsuji.
kiki@uhunix1.uhcc.Hawaii.Edu (Jack W. Wine) (05/16/91)
In article <1991May15.055715.4021@ecst.csuchico.edu> ekrimen@ecst.csuchico.edu (Ed Krimen) writes: > >Just to compare and contrast, when I opened up my STe to upgrade the RAM, >I noticed that I had an Epson drive and the 2 TOS ROMs were *socketed*, under >the drive. That's interesting, because I was thinking that Atari wouldn't be allocating any more development time to the 1040 STe machines, so they decided to solder the ROMs. Although in the March/April Atari Explorer, Greg Pratt said that a multi-tasking TOS was due soon, so I suppose it will require a ROM upgrade? I was a little happy that the Yamaha sound chip was socketed because they make a similar chip, the YM-2203, with enhanced capabilities (FM sound, 13-bit D/A, prog. multiple modulation and feedback) that is software compatible and near- ly pin compatible. If there is interest, I can post more specs later. Atari packed a lot of value in the STe models and I think the people of Hong Kong were right when they voted it the computer of the decade. >I read in one of the on-line magazines a couple of months ago that Jim Allen >had developed a 14meg upgrade for the STe and a 10meg upgrade for the MegaSTe. >I couldn't understand the difference, since I would expect the MegaSTe to >be able to be upgraded with more RAM, but I haven't heard anything since. >Perhaps these are related. The 14MB upgrade sounds like inserting 4 4MB SIMMs into the STe SIMM socket, with the last 2MB lost because of the ROM address space which begins at $E00000 (start of 14MB). The 10MB version sounds like 2 4MB and 2 1MB SIMMs, but I'm not sure why it would be particular to the Mega STe. Jack
csbrod@immd4.informatik.uni-erlangen.de (Claus Brod) (05/16/91)
kiki@uhunix1.uhcc.Hawaii.Edu (Jack W. Wine) writes: >I was a little happy that the Yamaha sound chip was socketed because they make >a similar chip, the YM-2203, with enhanced capabilities (FM sound, 13-bit D/A, >prog. multiple modulation and feedback) that is software compatible and near- >ly pin compatible. If there is interest, I can post more specs later. I am interested a lot. Tell us more about this chip. ---------------------------------------------------------------------- Claus Brod, Am Felsenkeller 2, Things. Take. Time. D-8772 Marktheidenfeld, Germany (Piet Hein) csbrod@medusa.informatik.uni-erlangen.de Claus Brod@wue.maus.de ----------------------------------------------------------------------
Roger.Sheppard@bbs.actrix.gen.nz (05/16/91)
In article <1991May15.055715.4021@ecst.csuchico.edu> ekrimen@ecst.csuchico.edu (Ed Krimen) writes: > In article <13023@uhccux.uhcc.Hawaii.Edu> kiki@uhunix1.uhcc.Hawaii.Edu (J. Wine) writes: > I read in one of the on-line magazines a couple of months ago that Jim Allen > had developed a 14meg upgrade for the STe and a 10meg upgrade for the MegaSTe. > I couldn't understand the difference, since I would expect the MegaSTe to > be able to be upgraded with more RAM, but I haven't heard anything since. > Perhaps these are related. > -- > ||| Ed Krimen [ekrimen@ecst.csuchico.edu or al661@cleveland.freenet.edu] > ||| Video Production Major, California State University, Chico > / | \ SysOp, Fuji BBS: 916-894-1261 > TWO WEEKS UNTIL GRADUATION!! Well the latest STE's here in NZ. now have Brass inserts (3) for the Screws that hold down the Main PCB. these can be seen if you turn over the machine, as they come though the bottom, also with a newer type Epson disk drive, now has a blue plasic cover over the motor, this drive will give you 84 tracks. The Sony Drive will give you 82 track and the Mitsumi will give you 81. The early Epson dives can have problems with a PC conector to drive PCB, (Drive will not stop), just remove this connector and clean with contact cleaner, problem was intermitent index pulses. I don't like any of the newer Epson drives, they use a rack and pinion mechanism to move the head, this can cause hairs to get caught in the rack and pinion gears. -- Roger W. Sheppard 85 Donovan Rd, Kapiti New Zealand...
stephen@oahu.cs.ucla.edu (Steve Whitney) (05/20/91)
In article <1991May15.055715.4021@ecst.csuchico.edu> ekrimen@ecst.csuchico.edu (Ed Krimen) writes: ... > >I read in one of the on-line magazines a couple of months ago that Jim Allen >had developed a 14meg upgrade for the STe and a 10meg upgrade for the MegaSTe. >I couldn't understand the difference, since I would expect the MegaSTe to >be able to be upgraded with more RAM, but I haven't heard anything since. I imagine that the reason for a smaller maximal memory upgrade for MegaSTEs is that the VME bus requires some address space. I don't know if 4 Meg is available on the VME bus or not, but that might explain it. Can't put RAM where the bus is on the memory map! >-- > ||| Ed Krimen [ekrimen@ecst.csuchico.edu or al661@cleveland.freenet.edu] > ||| Video Production Major, California State University, Chico > / | \ SysOp, Fuji BBS: 916-894-1261 > TWO WEEKS UNTIL GRADUATION!! -- Steve Whitney - UCLA CS Grad Student (())_-_(()) Soon to be working at Silicon Graphics | (* *) | Internet: stephen@cs.ucla.edu UCLA Bruin--> { \_@_/ } GEnie: S.WHITNEY `-----'