madd@bu-cs.BU.EDU (Jim Frost) (08/07/89)
I have a Northgate 386-16 machine with 4Mb of memory, which is woefully inadequate for my purposes. I would like to expand it to at least 8Mb, but have been unable to find an expansion board. Northgate apparently used to have such a board, but our sales representative said they no longer sell it (seems stupid with people starting to use OS/2, really stupid with all us UNIX people, but what do I know I just buy the stuff). The sales rep suggested I look in Computer Shopper, which had myriads of AT expansion boards but I'm not too keen on loosing the memory speed. We found one dealer who had three kinds of 32-bit expansion boards but we could not determine which, if any, would work in our machine. If anyone knows where to find a 32-bit memory expansion board for the Northgate 386 machine, please email me. If anyone knows what the 32-bit bus used by Northgate is called, that would probably be useful too. Please email replies to madd@std.com or uunet!skuld!madd. Thank you, jim frost software tool & die madd@std.com
burton@mitisft.Convergent.COM (Philip Burton) (08/08/89)
In article <35950@bu-cs.BU.EDU> madd@bu-it.bu.edu (Jim Frost) writes: >I have a Northgate 386-16 machine with 4Mb of memory, which is . . (deleted) . which had myriads of AT expansion boards but I'm not too keen on . . >Please email replies to madd@std.com or uunet!skuld!madd. > >Thank you, > >jim frost >software tool & die >madd@std.com A somewhat related question is whether or not you can take 16 bit memory boards, say an AST ADVantage, with 120/150 ns memory, and run it successfully on a 20 or 25 MHz system. Is this possible? How much does performance suffer? Does a RAM cache on the motherboard help any? PHIL BURTON Convergent Technologies 408 435 3791
madd@bu-cs.BU.EDU (Jim Frost) (08/08/89)
In article <1151@mitisft.Convergent.COM> burton@mitisft.convergent.COM writes: |In article <35950@bu-cs.BU.EDU> madd@bu-it.bu.edu (Jim Frost) writes: |>I have a Northgate 386-16 machine with 4Mb of memory, which is |. |. (deleted) |. |which had myriads of AT expansion boards but I'm not too keen on |. |. |A somewhat related question is whether or not you can take 16 bit memory boards, |say an AST ADVantage, with 120/150 ns memory, and run it successfully on a |20 or 25 MHz system. Is this possible? How much does performance suffer? |Does a RAM cache on the motherboard help any? Yes, it can be done, but most 386 machines run the I/O bus at 8MHz while the 32-bit memory bus runs at CPU speed -- typically 16MHz, 20Mhz, 25MHz, or 33MHz. Even at the slowest speed you're talking about at least half of the performance. A cache will help, often quite a bit, but it's still not going to be the same as 32-bit memory. jim frost software tool & die madd@std.com
davidsen@sungod.crd.ge.com (ody) (08/09/89)
In article <1151@mitisft.Convergent.COM> burton@mitisft.convergent.COM writes: | A somewhat related question is whether or not you can take 16 bit memory boards, | say an AST ADVantage, with 120/150 ns memory, and run it successfully on a | 20 or 25 MHz system. Is this possible? How much does performance suffer? | Does a RAM cache on the motherboard help any? I've been running a mixed 32/16 bit system for several years and have done quite a bit of testing. Here's how it works. If you're running DOS and using the high memory for RAMdisk, the cache won't help much, if at all. If you run QEMM or other software to use the extended mem as expanded mem, it will help quite a bit. To get the average access penalty you can perform a calculation like this: access to fast memory = 2 cycles for 32 bits, slow memory 2 cycles + 1 wait state (16MHz, 2 for 20, 3 for 25, using 120ns memory). Since a fetch takes two accesses on the 16 bit bus, double that. I used 85% cache hits, because (a) I saw that number in a lot of places, and (b) I measured 84.4% on a simulator using a C compile and an awk program. <_ cycles -> < effective > CPU MHz fast slow clocks waits 16 2 6 2.60 0.60 20 2 8 2.90 0.90 25 2 10 3.20 1.20 From this I conclude that a 20MHz 386 running 64kb cache and memory on a 16 bit bus will be about as fast as an 1w/s machine without cache. The penalty will be less for 16 bit data transfers (all instructions are 32 bits, I believe). These figures are for AMI 16 MHz 386, running Xenix/386. Any other hardware or software will change them (but not much). These calculations reasonably match actual tests I have made. bill davidsen (davidsen@crdos1.crd.GE.COM) {uunet | philabs}!crdgw1!crdos1!davidsen "Stupidity, like virtue, is its own reward" -me
wek@point.UUCP (Bill Kuykendall) (08/11/89)
> <_ cycles -> < effective > >CPU MHz fast slow clocks waits >16 2 6 2.60 0.60 >20 2 8 2.90 0.90 >25 2 10 3.20 1.20 Well, my tests didn't include such impressive calculations, but 2 1/2 years ago when I first got my 16mz '386 (Intel MB) I did quite a bit of testing with 16-bit vs 32-bit memory running Desqview & Qemm to multitask benchmark programs. The bottom line seemed to be that 32-bit memory was nearly *4 times* as fast as 16-bit memory. Desqview and Qemm have improved considerably since then, as has the '386 architecture (with various caching and interleaving schemes) so perhaps my tests are no longer relevant. Bill Kuykendall ...ddsw1!point!wek