Path: utzoo!attcan!uunet!samsung!gem.mps.ohio-state.edu!rpi!uwm.edu!rutgers!att!tsdiag!ocpt!pechter From: pechter@ocpt.ccur.com (Bill Pechter ) Newsgroups: comp.sys.ibm.pc Subject: Re: DMA controller chip error Summary: Fastback Plus DMA Keywords: Fastback, DMA Message-ID: <21@ocpt.ccur.com> Date: 19 Nov 89 01:18:27 GMT References: <941@ac.dal.ca> Organization: CONCURRENT COMPUTER CORP. OCEANPORT NJ Lines: 40 In article <941@ac.dal.ca>, bertfram@ac.dal.ca writes: > I just bought a copy of Fastback +. When I configured it the other day > it informed me that I have an error in my DMA controller chip. I'm not > a happy camper. They suggested that I buy a fix from them. This is the > first I've heard of this problem, and it hasn't seemed to affect anything > in the past. This is a very common problem in XT's and the Fastback Manual goes into it in depth... There are 2 reasons for the problem: 1. Defective DMA chips (most of these in original PC's) which don't let you do read and write DMA at the same time. 2. Badly built clones that have problems with DMA at high speed. (This is my speculation -- the bad DMA chips should be out of circulation by now. I think this problem is bios related) There are 2 ways to get around this -- run with slow or medium DMA instead of fast DMA, or retry the test at a slower cpu speed with DMA at the fast rate. Fastback Plus at 4.77 MHZ with high DMA beats slow DMA at a 10 meg speed I think (your mileage may vary 8 -} ). Also -- the Fastback Plus tests are sensitive to TSR's. I also improved my DMA rate by changing my refresh rate a bit. Bill (running FB Plus for 2 years on an AT&T 6300 which wouldn't run old Fastback with any kind of reliability...) -- Bill Pechter -- Home - 103 Governors Road, Lakewood, NJ 08701 (201)370-0709 Work -- Concurrent Computer Corp., 2 Crescent Pl, MS 172, Oceanport,NJ 07757 Phone -- (201)870-4780 Usenet . . . rutgers!pedsga!tsdiag!scr1!pechter ** MS-DOS is CP/M on steroids, bigger, bulkier and not much better **