tlm@pur-phy (Timothy Lee Meisenheimer) (03/16/88)
[ This is for you wicked lind eater. ] This is to see if anyone has a suggestion or a similar experience: Machine: Amiga 2000 still under warrenty. I'm having problems with my floppies getting corrupted while I'm doing file transfers - say using VT100 (kermit) and DNET (which is getting stable and usefull even at 1200 baud). I've had the same problem transfering files from ramdisk to df0: but it seems to hit me more when I'm down loading. Needless to say it is very disconcerting knowing I could lose my floppy at any moment! (About once a day!) So...., I had the drive checked out (under warrenty) and it had alignment problems that couldn't be fixed with the scope so it got replaced :-). But noooooooo! I get about 50k transfered an BAMMMMMmmm the disk does the usual horrible grind, grind, brrraaat with a system request about a read/write error and I'm stuck again!!!! I started checking the disks (had three in a row and it doesn't matter who made them) with sectorama (directory check) and found block x00370 was where it always started the corruption. Since it's independent of program, and floppy drive, do I have a flakey Denise (or whatever the name is) on my mother board? Of course the fan has just started to sound like a Mach truck on a cold winter morning when I first turn on the machine - Aarrrrrrgggghhhhhhh!!!!! It goes away in about 10 minutes but the bearing are probably going. The shop is going to call CBM about it but I'd still appreciate any insights from the net. The frustration level is starting to rise to very warm, and I'm getting tired of the hassle and would like to get some real work done! - On another frustrating note: I just recieved Manx 3.6 etc and like that SDB! But when I compile I keep getting system requests for the boot disk - I don't boot up on a compiler disk, I have most everything in memory - I can click cancel and it goes on its merry way. I've got the environment stuff 'set' correctly, and have reassigned C:,S:,T: and SYS: etc. In fact I reeassign everything (FONTS: etc.) and more and still get those requests!!! Who and what is it (make, cc, as, ln) looking for!! I have noticed that if you compile with +ff of +fi the compiler needs to suck in the floating point libraries (why??????) if they haven't already been brought in but this is a different issue. I've just got the latest version of sc (spreadsheet) running and that is how I came about all this. -Sorry to take up so much bandwidth. -Thanks in advance. tlm@newton.physics.purdue.edu
jesup@pawl18.pawl.rpi.edu (Randell E. Jesup) (03/17/88)
In article <1021@pur-phy> tlm@newton.physics.purdue.edu.UUCP (Timothy Lee Meisenheimer) writes: >I had the drive checked out (under warrenty) and it had alignment >problems that couldn't be fixed with the scope so it got replaced :-). >But noooooooo! I get about 50k transfered an BAMMMMMmmm the disk >does the usual horrible grind, grind, brrraaat with a system request >about a read/write error and I'm stuck again!!!! I started checking the >disks (had three in a row and it doesn't matter who made them) with >sectorama (directory check) and found block x00370 was where >it always started the corruption. Since it's independent of program, Actually, I think it might matter who makes them. My brother has an amiga, and is a naive user. He kept getting read errors on his WordPerfect disk (copied from master), even after he re-diskcopied it. So he tried it on another disk, same problem. And a third, as well. It turned out that 5 out of the 10 disks he bought were bad, and if you did a format on them, they didn't pass. Try formatting some of these disk you have problems with. Next question: do you run any programs all the time, like maybe an old copy of the shell? Anything that does a Delay(0) (like an old copy of the dillon shell) will cause random trashed of the disk, usually track 40. 0x0370 is 880 decimal, or track 40 sector 0 (what normally gets hit by Delay(0)). Try a different download program, don't run ANY background stuff, and I'll bet it stops. Not 100% sure, but it is a good bet. >on my mother board? Of course the fan has just started to sound like >a Mach truck on a cold winter morning when I first turn on the machine - >Aarrrrrrgggghhhhhhh!!!!! It goes away in about 10 minutes but the >bearing are probably going. The shop is going to call CBM about it >but I'd still appreciate any insights from the net. The frustration My fan is like that too, but it was like that from the beginning. Since I bought it mailorder, I haven't bothered doing anything about it yet. It does sound like rocks in the fan for 10 min or so. > tlm@newton.physics.purdue.edu // Randell Jesup Lunge Software Development // Dedicated Amiga Programmer 13 Frear Ave, Troy, NY 12180 \\// beowulf!lunge!jesup@steinmetz.UUCP (518) 272-2942 \/ (uunet!steinmetz!beowulf!lunge!jesup) BIX: rjesup (-: The Few, The Proud, The Architects of the RPM40 40MIPS CMOS Micro :-)
kenchiu@phoenix.Princeton.EDU (Kenneth Chiu) (03/17/88)
In article <1021@pur-phy> tlm@newton.physics.purdue.edu.UUCP (Timothy Lee Meisenheimer) writes: >I just recieved Manx 3.6 etc and like that SDB! But when I compile >I keep getting system requests for the boot disk - I don't boot up >on a compiler disk, I have most everything in memory - I can >click cancel and it goes on its merry way. I've got the environment >stuff 'set' correctly, and have reassigned C:,S:,T: and SYS: etc. >In fact I reeassign everything (FONTS: etc.) and more and still >get those requests!!! I don't have these problems, and I do something similar. Most likely your forgetting something. Use 'assign' to see if anything is still assigned to your boot disk. Also try 'set'. Oh, and for more information, try 'info'. That will tell you if you have any outstanding locks on your boot disk. Might give you some clues. >I have noticed that if you compile with +ff of +fi the compiler needs >to suck in the floating point libraries (why??????) if they haven't >already been brought in but this is a different issue. First, these are link libraries. Actual code is linked into the final binary. Second, even if they were the other kind of library, the shared Amiga kind, they might get purged if no one was using them and the system needed the memory. Ken Chiu
cjp@antique.UUCP (Charles Poirier) (03/18/88)
In article <2079@phoenix.Princeton.EDU> kenchiu@phoenix.Princeton.EDU (Kenneth Chiu) writes: >In article <1021@pur-phy> tlm@newton.physics.purdue.edu.UUCP (Timothy Lee Meisenheimer) writes: >>I just recieved Manx 3.6 etc and like that SDB! But when I compile >>I keep getting system requests for the boot disk - I don't boot up >>on a compiler disk, I have most everything in memory - I can >>click cancel and it goes on its merry way. I've got the environment >>stuff 'set' correctly, and have reassigned C:,S:,T: and SYS: etc. > >forgetting something. Use 'assign' to see if anything is still assigned to >your boot disk. Also try 'set'. Oh, and for more information, try 'info'. I had a similar problem. Moved all my compiler files to hard disk, did "set"s and "assign"s and "info"s out the wazoo, triple-checked my makefile. Using Manx 3.4a. "Make" asks for sys1: each time, cancellable. I still can't figure it out -- unless maybe the commented lines in my makefile are being read somehow. Anyway, work around this by assigning sys1: as something innocuous like nil: or ram:. Bingo, no requestors. -- Charles Poirier (decvax,ihnp4,attmail)!vax135!cjp "Docking complete... Docking complete... Docking complete..."
jim@b11.UUCP (Jim Levie ) (03/19/88)
In article <540@imagine.PAWL.RPI.EDU>, jesup@pawl18.pawl.rpi.edu (Randell E. Jesup) writes: > >on my mother board? Of course the fan has just started to sound like > >a Mach truck on a cold winter morning when I first turn on the machine - > >Aarrrrrrgggghhhhhhh!!!!! It goes away in about 10 minutes but the > >bearing are probably going. The shop is going to call CBM about it > >but I'd still appreciate any insights from the net. The frustration > > My fan is like that too, but it was like that from the beginning. > Since I bought it mailorder, I haven't bothered doing anything about it yet. > It does sound like rocks in the fan for 10 min or so. > Three of the four Amiga 2000's that we have developed noisy fans within 30 days. About 10 minutes with a screwdriver and some 3-in-1 oil cured the problem. I don't think that the bearings were proberly lubricated during manufacture. I realize that price is important in picking hardware conponents but I think that using a sleeve-bearing fan in the 2000 is a mistake. There is little difference in price between a sleeve and ball bearing fan, but a very large difference in the lifetime. Considering the damage that a non-operational fan can do to the innards of a computer it seems to be a poor compromise. -- =+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+ Jim Levie REMTECH Inc Huntsville, Al The opinions expressed above are just that. Ph. (205) 536-8581 email: uunet!ingr!b11!jim
kenchiu@phoenix.Princeton.EDU (Kenneth Chiu) (03/22/88)
In article <2114@antique.UUCP> cjp@antique.UUCP (Charles Poirier) writes: >I had a similar problem. Moved all my compiler files to hard disk, did >"set"s and "assign"s and "info"s out the wazoo, triple-checked my >makefile. Using Manx 3.4a. "Make" asks for sys1: each time, >cancellable. That's strange, because I have no problem of this sort, and I'm also using 3.4a. The only thing that I can think of, that may or may not have anything to do with the problem, is that if you don't 'set' something, the system will try to load the pseudo-library off disk. Doesn't explain why it should want SYS1 instead of your boot disk, though. Ken Chiu
scott@applix.UUCP (Scott Evernden) (03/22/88)
In article <2114@antique.UUCP> cjp@antique.UUCP (Charles Poirier) writes: >>In article <1021@pur-phy> tlm@newton.physics.purdue.edu.UUCP (Timothy Lee Meisenheimer) writes: >>>I just recieved Manx 3.6 etc and like that SDB! But when I compile >>>I keep getting system requests for the boot disk -... >I had a similar problem. Could it be that the compiler is trying to get to "mathieeedoubbas.library"? I know that SDB only works with pgms compiled +fi meaning it does IEEE arithmetic; maybe the compiler's built the same way?? Don't forget, Manx dropped their own mx IEEE (runtime) libs in 3.6. -scott