wewallac@watsol.waterloo.edu (Bill Wallace) (02/19/91)
Does anyone have suggestions for how to get a large amount of memory in a A3000? I am interested in having 64 Megbytes, 96 or 128 would be better. Also, running a single, debugged, program, does anyone know how long between crashes on an A3000? Finally, does anyone have experience with large hard disks on an A3000? I am talking about 1 Gb or larger hard disk. Bill Wallace New OED wewallace@watsol.waterloo.edu
rwm@atronx.OCUnix.On.Ca (Russell McOrmond) (02/19/91)
In a message posted on Mon, 18 Feb 1991 21:48:27 GMT,
wewallac@watsol.waterloo.edu (Bill Wallace) wrote:
BW>running a single, debugged, program, does anyone know how long between crashes
BW>on an A3000? Finally, does anyone have experience with large hard disks on
Well, WITH buggy programs (IE: Stuff I've written myself, and a lot of other
BETA software) I go about 2 weeks before crashing sometimes.
I'm quite happy with the Stability of the new O.S. It's quite a bit better
than 1.2/1.3 .
---
Opinions expressed in this message are my Own. I represent nobody else.
Russell McOrmond rwm@Atronx.OCUnix.On.Ca {tigris,alzabo,...}!atronx!rwm
FidoNet 1:163/109 Net Support: (613) 230-2282
Amiga-Fidonet Support 1:1/109 Gateway for .Amiga.OCUnix.On.Cadaveh@cbmvax.commodore.com (Dave Haynie) (02/19/91)
In article <1991Feb18.214827.10775@watdragon.waterloo.edu> wewallac@watsol.waterloo.edu (Bill Wallace) writes: >Does anyone have suggestions for how to get a large amount of memory in a >A3000? I am interested in having 64 Megbytes, 96 or 128 would be better. At this exact moment, you'd have to build it yourself. So far, at least, there aren't any Zorro III memory boards that I know of out on the market. Any company wishing to design one could base it on the example I did for last June's DevCon in Atlanta. That example is fully tested, we have four working samples. The board supports up to 32 MB. It has been slightly over a year since we started sending out A3000/Zorro III docs. It takes somewhere between two weeks (the total design, layout, and debug time for the example memory board) and a year or so to get a producable memory design going, depending on how important it is. The main issue, I imagine, is marketing. MOST A3000 owners don't need memory beyond the 18MB on the motherboard. And there aren't comparatively all that many A3000 owners. So anyone building an A3000 memory board for real commercial production will probably need to consider the A2000 as well, which makes the design more complex. >Also, running a single, debugged, program, does anyone know how long between >crashes on an A3000? I ran one for a month once, a variety of test programs going between three different hard disks on the A3000 (Hardframe, A2091, and internal), but had to shut it off since I needed the system for other things. At some times of the year, we don't have uninterrupted power here for more than a month, but there's no other reason that setup wouldn't still be running if I hadn't taken it down. >Bill Wallace >New OED -- Dave Haynie Commodore-Amiga (Amiga 3000) "The Crew That Never Rests" {uunet|pyramid|rutgers}!cbmvax!daveh PLINK: hazy BIX: hazy "What works for me might work for you" -Jimmy Buffett
blgardne@javelin.es.com (Blaine Gardner) (02/19/91)
wewallac@watsol.waterloo.edu (Bill Wallace) writes: > Finally, does anyone have experience with large hard disks on >an A3000? I am talking about 1 Gb or larger hard disk. I've run gigabyte drives by H-P and Fujitsu on the A3000. With DiskSpeed 3.1 the H-P hit 1.5 megabytes/second, and the Fujitsu 1.9 megabyes/sec. Even running in high intensity mode with both DMA and CPU stress turned didn't slow the results by more than 5-10%. Impressive performance! The only problem I ran into is a bug in 2.02 that limits partitions to less than 512 megabytes. I understand that this has already been fixed. -- Blaine Gardner @ Evans & Sutherland 580 Arapeen Drive, SLC, Utah 84108 blgardne@javelin.sim.es.com or blgardne%javelin@dsd.es.com ...dsd.es.com!javelin!blgardne or {decwrl, utah-cs}!esunix!blgardne DoD #0046 My other motorcycle is a Quadracer. BIX: blaine_g
bernie@metapro.DIALix.oz.au (Bernd Felsche) (03/02/91)
In <1991Feb18.214827.10775@watdragon.waterloo.edu> wewallac@watsol.waterloo.edu (Bill Wallace) writes: >Does anyone have suggestions for how to get a large amount of memory in a A3000? >I am interested in having 64 Megbytes, 96 or 128 would be better. Also, >running a single, debugged, program, does anyone know how long between crashes >on an A3000? Finally, does anyone have experience with large hard disks on >an A3000? I am talking about 1 Gb or larger hard disk. Just a suggestion: Make sure that it has at least parity, or even better EDAC. There might only be a 1/100 million chance of a bit error, but the more bits you have the greater tha chance. There's nothing worse than the system falling over after 18 hours of ray-trace. The extra cost is well worth it. No I don't know of anybody who make the things for Zorro III. Disks, you can get almost anywhere. HP make some neat ones, wit high transfer rates, as are the nes "Elite" drives which are all too rare. -- Bernd Felsche, _--_|\ #include <std/disclaimer.h> Metapro Systems, / sale \ Fax: +61 9 472 3337 328 Albany Highway, \_.--._/ Phone: +61 9 362 9355 Victoria Park, Western Australia v Email: bernie@metapro.DIALix.oz.au
sck@barn.watson.ibm.com (SC Kennedy) (03/05/91)
For memory I don't know... But for Drives.... I have been playing around with several ST1600N drives from Seagate. They work great in a 2500, I have had upto 4 of them connected to my machine and could operate nicely. Problems I did find were mostly AmigaDOS related, and were'nt problems as much as "how do I deal with this?" ie: When you have 4 drives that format out to 1.4 Gig and you partition them in 100 MG increments, how long does it take you to mount all of them. or better yet: Where did I leave my source??? Oh, yeah disk2, partition 6, sub-directory, 14-42-523. Simple... IMHO Seagate has a very decent drive. But, don't get the out of form factor version. I had a whale of a time trying to get that one to be recognized by the 2091. ------------------------------------------------------------------------ Scott C. Kennedy (sck@shed.watson.ibm.com) | "All we are saying ... Distributed High Performance Computing | is give peace a chance..." I.B.M. Thomas J. Watson Research Facility | John Lennon - Dec. 8, 1980 ------------------------------------------------------------------------
daveh@cbmvax.commodore.com (Dave Haynie) (03/05/91)
In article <1991Mar2.053412.2223@metapro.DIALix.oz.au> bernie@metapro.DIALix.oz.au (Bernd Felsche) writes: >In <1991Feb18.214827.10775@watdragon.waterloo.edu> wewallac@watsol.waterloo.edu (Bill Wallace) writes: >>Does anyone have suggestions for how to get a large amount of memory in a A3000? >Just a suggestion: Make sure that it has at least parity, or even >better EDAC. There might only be a 1/100 million chance of a bit >error, but the more bits you have the greater tha chance. There's >nothing worse than the system falling over after 18 hours of >ray-trace. Well, error correcting memory might be useful. Parity memory would give you a larger target for that random bit drop, and guarantee the crash, since the only real recourse for a parity error, other than ignoring it, is to force a system error (UNIX can sometimes deal with it -- a parity error on a code page could be fixed simply by reloading that code page). A parity error in the data space of a ray trace would probably, at wort, give you an ugly image, at best, be undetectable. Certainly such errors elsewhere, as in big financial models or some-such, are more dangerous. But such errors are extremely rare. >The extra cost is well worth it. No I don't know of anybody who make >the things for Zorro III. To date, no one has announced a Zorro III memory board, far as I know at least. But who knows, there's an Amiga show in NYC this month, and another next month. Shows are often a good time to introduce new products. >Bernd Felsche, _--_|\ #include <std/disclaimer.h> -- Dave Haynie Commodore-Amiga (Amiga 3000) "The Crew That Never Rests" {uunet|pyramid|rutgers}!cbmvax!daveh PLINK: hazy BIX: hazy "What works for me might work for you" -Jimmy Buffett