skank@iastate.edu (Skank George L) (05/26/91)
Hello folks. My A3000 is exhibiting a phenomenon that seems strange to me. I recently had 4Megs of memory added. The origional 1Meg of memory is now in chip ram, bringing the total chip ram to 2Megs. Here's the problem, each time I boot the machine I get a different amount of chip ram available. Usually it's about 900K, which is what I had before the upgrade. Sometimes I get 1.1, 1.4, and as much as 1.9Megs of chip ram available. What's going on? I can imagine that some system structures are allocated differently each time the machine boots, but we're talking about almost a Meg of memory that seems to come and go more or less at will. Could I have bad memory chips? Could I have a bad something (I don't know what) else? Is my machine possessed? The machine in question is an A3000 currently running Workbench 2.02. Any help would be appreciated. Thanks! --George -- George L. Skank |Fast cars, fast women, fast computers... /// Senior, Electrical Engineering |Amiga! /// Iowa State University, Ames, IA | \\\ /// skank@iastate.edu |Phone: (515) 233-2165 \\X//
jesup@cbmvax.commodore.com (Randell Jesup) (05/28/91)
In article <1991May26.064123.18564@news.iastate.edu> skank@iastate.edu (Skank George L) writes: >Hello folks. My A3000 is exhibiting a phenomenon that seems strange to me. >I recently had 4Megs of memory added. The origional 1Meg of memory is now >in chip ram, bringing the total chip ram to 2Megs. Here's the problem, >each time I boot the machine I get a different amount of chip ram available. >Usually it's about 900K, which is what I had before the upgrade. Sometimes >I get 1.1, 1.4, and as much as 1.9Megs of chip ram available. My first suspicion (as a software guy) would be a bad memory chip (one that's marginal, so it fails the chip-ram existence test at different places). It also could be making intermittent contact, perhaps (less likely). -- Randell Jesup, Jack-of-quite-a-few-trades, Commodore Engineering. {uunet|rutgers}!cbmvax!jesup, jesup@cbmvax.cbm.commodore.com BIX: rjesup Disclaimer: Nothing I say is anything other than my personal opinion. "No matter where you go, there you are." - Buckaroo Banzai
skank@iastate.edu (Skank George L) (05/28/91)
In article <21950@cbmvax.commodore.com> jesup@cbmvax.commodore.com (Randell Jesup) writes: >In article <1991May26.064123.18564@news.iastate.edu> skank@iastate.edu (Skank George L) writes: >>Hello folks. My A3000 is exhibiting a phenomenon that seems strange to me. >>I recently had 4Megs of memory added. The origional 1Meg of memory is now >>in chip ram, bringing the total chip ram to 2Megs. Here's the problem, >>each time I boot the machine I get a different amount of chip ram available. >>Usually it's about 900K, which is what I had before the upgrade. Sometimes >>I get 1.1, 1.4, and as much as 1.9Megs of chip ram available. > > My first suspicion (as a software guy) would be a bad memory chip >(one that's marginal, so it fails the chip-ram existence test at different >places). It also could be making intermittent contact, perhaps (less likely). > >-- >Randell Jesup, Jack-of-quite-a-few-trades, Commodore Engineering. >{uunet|rutgers}!cbmvax!jesup, jesup@cbmvax.cbm.commodore.com BIX: rjesup >Disclaimer: Nothing I say is anything other than my personal opinion. >"No matter where you go, there you are." - Buckaroo Banzai Dandy. :-( How does one go about testing this? Are there programs to test this or is this enough of a hardware thing that I should take it back to my dealer to test? --George -- George L. Skank |Fast cars, fast women, fast computers... /// Senior, Electrical Engineering |Amiga! /// Iowa State University, Ames, IA | \\\ /// skank@iastate.edu |Phone: (515) 233-2165 \\X//
daveh@cbmvax.commodore.com (Dave Haynie) (05/29/91)
In article <1991May26.064123.18564@news.iastate.edu> skank@iastate.edu (Skank George L) writes: >Hello folks. My A3000 is exhibiting a phenomenon that seems strange to me. >I recently had 4Megs of memory added. The origional 1Meg of memory is now >in chip ram, bringing the total chip ram to 2Megs. Here's the problem, >each time I boot the machine I get a different amount of chip ram available. >Usually it's about 900K, which is what I had before the upgrade. Sometimes >I get 1.1, 1.4, and as much as 1.9Megs of chip ram available. What's going >on? Sounds to me like you have at least one flakey chip. That's not necessarily a bad chip, you might have one with a bent-under pin, or other such atrocity. Basically, as I understand it, when Exec wakes up from a cold boot it goes out and calculates how much Chip RAM is available, on a reasonably fine grain (maybe 64K or 128K chunks). It doesn't really expect to find flakey Chip RAM. So in testing memory, it'll most likely go until it finds a chunk that doesn't test out. If you're not crashing, this indicates that the first meg is still fine, and that you never get to the second meg. That's actually a good sign, as it indicates that nothing is likely being shorted out and you _probably_ don't have a severely dead chip, as these things tend to drag down the whole chip bus. You should have that second bank of Chip RAM looked at carefully. >George L. Skank |Fast cars, fast women, fast computers... /// -- Dave Haynie Commodore-Amiga (Amiga 3000) "The Crew That Never Rests" {uunet|pyramid|rutgers}!cbmvax!daveh PLINK: hazy BIX: hazy "That's me in the corner, that's me in the spotlight" -R.E.M.
hull@hao.hao.ucar.edu (Howard Hull) (05/30/91)
In article <1991May28.073221.8614@news.iastate.edu> skank@iastate.edu (Skank George L) writes: > > Dandy. :-( How does one go about testing this? Are there programs to >test this or is this enough of a hardware thing that I should take it back >to my dealer to test? > > --George How do you test memory? You can use MemBoardTest from Fred Fish Disk 158, though the length of some of the A3000 memory addresses will write beyond the edge of some of the gadget fields. The Contents file says it includes source in Modula2, so if you have a Modula2 compiler, you can probably fix it in short order. But it's just a cosmetic flaw and doesn't prevent the program from working on $7FFFFFC memory. Can your dealer do it? Yeah, by any of several methods, including: Chip Swap = Some Bucks, but you get the suspect parts back if you ask for 'em Board Swap = More Bucks but you get new I/O chips (sort of like trading in a 6-month old Mercedes just to get clean ash trays). A3000 Swap = A good deal if you have a Commodore PET to throw in for trade :-) Fish 168 or equiv = You don't get to know what he found, so you might as well do it yourself so that after this you'll know what's going on, unless of course your A3000 is under warranty - then you go to the dealer in any case. Any CBM Authorized Amiga dealer should be able to accomplish it, easily. Howard Hull hull@ncar.ucar.edu