robertl@killer.UUCP (Robert Lord) (08/02/87)
I was just wondering, since I had seen some talk of using 386's in AT's,
what the compatibility problems would be? I am not really interested in
the functions of the 386 per say, just the speed increase. Soposedly,
Intel is coming out with a 20Mhz (whew!) 80386 chip later this year, and
I would like to drop it into me AT to make it into a fast desktop. If I
did this, would I have any problems with existing software/hardware?
Would I be able to use software designed for the 386? Would all the
bios routines be the same?
Thanks in advance,
Robert Lord
P.S. Could someone please drop me a copy of the message of how to turn your
AT into a 386, I seem to have lost it....dyer@spdcc.COM (Steve Dyer) (08/03/87)
The best way to turn your AT or AT clone into a 386 box is to buy the Intel
Inboard 386/AT card. You can go cheap and buy it without any memory
for about $1400 through mail order
This gives you about a 2X speedup
over an 8mhz AT. You can add 32-bit memory in 1mb increments, up to 3mb.
32-bit memory in place of 16-bit memory gives you a 3X speedup over an
8 mhz AT. Finally, using a true 32-bit C compiler generating native 386
instructions (such as MSC 5.0) combined with 32-memory yields about a 5X
speed increase over an 8mhz AT running 16-bit instructions. The board
has a provision for a 387 math processor if you want that.
Eventually you'll be able to get a generic 386 motherboard cheap, but they
don't seem to have crossed that price threshold yet and they aren't widely
available.
Intel provides very competent phone support, and they'll be around at least
as long as the 386 chip.
--
Steve Dyer
dyer@harvard.harvard.edu
dyer@spdcc.COM aka {ihnp4,harvard,linus,ima,bbn,m2c}!spdcc!dyercbenda@unccvax.UUCP (carl m benda) (08/04/87)
In article <1272@killer.UUCP>, robertl@killer.UUCP (Robert Lord) writes: > Intel is coming out with a 20Mhz (whew!) 80386 chip later this year, and > I would like to drop it into me AT to make it into a fast desktop. If I > did this, would I have any problems with existing software/hardware? > Would I be able to use software designed for the 386? Would all the No No and Finally No. The problem with what you described is the fact that your AT was probably designed to run at NO more than 10MHz. This is because the memory is probably no faster than 120ns RAM. What this means is that if you replace your 80286 with a 80386 inserted into a Chetah Adapter, it will perform at a MIPS rate which is no better than your current configuration. What you will gain is the addressing capabilaty of the 80386, namely, 4G bytes, and 1Meg segments, instead of 80286 which is 16Meg and 64K segments. With this in mind, a 20Mhz 80386 chip would be wasted in your machine. What you would really like to do is keep everything in your AT but your mother board. Replace the mother board with an 80386 motherboard which will allow you to have a 20Mhz machine with 32 bit word memory moves, your AT can only do 16bit memory moves. Here again memory for this machine will be exspensive, I believe with a 20Mhz clock rate the memory needs to be at least as fast as 70ns. This is a guess since my ALR 386 uses 80ns ram and has a 16Mhz clock rate. hope this helps /Carl ...decvax!mcnc!unccvax!cbenda
enped@wolf.UUCP (Eric Pederson) (08/04/87)
Most of the speed ratings put 386 machines with the same bus and the same clock speed slightly above the compared 286 machines. The added speed isn't neccesarily worth the extra cost in getting the 386 add-on. You'd want it most likely for the ability to run multiple copies of DOS, a nice linear UNIX, etc. etc. You can speed up your AT machine with a faster disk and a decent cache program.
burton@parcvax.Xerox.COM (Philip M. Burton) (08/04/87)
In article <503@wolf.UUCP> enped@wolf.UUCP (Eric Pederson) writes: > >Most of the speed ratings put 386 machines with the same bus and the same >clock speed slightly above the compared 286 machines. The added speed >isn't neccesarily worth the extra cost in getting the 386 add-on. You'd >want it most likely for the ability to run multiple copies of DOS, a nice >linear UNIX, etc. etc. You can speed up your AT machine with a faster >disk and a decent cache program. I agree. My preference for a new term solution would be to replce the 286 with a 386, and leave everything else alone. EGA, AST multifunction card, disk controller, etc. Has anyone actually installed the Cheetah upgrade? Does anyone know about the rumored Intel special 386 that would plug into an AT motherboard directly? -- Philip Burton Xerox Corporation 408 737 4635 ... usual disclaimers apply ...
davidsen@steinmetz.steinmetz.UUCP (William E. Davidsen Jr) (08/05/87)
In article <214@spdcc.COM> dyer@spdcc.COM (Steve Dyer) writes: >The best way to turn your AT or AT clone into a 386 box is to buy the Intel >Inboard 386/AT card. You can go cheap and buy it without any memory >for about $1400 through mail order Hum. I saw a board advertized in PC Week, 386 based, motherboard replacement, 2MB of memory installed, 32 bit slots. $1375, has sockets (and jumpers I assume) for either the 80287 or 80387. I have no idea of the quality of this board (I hope to soon), but I think this may be a good time to select a good replacement motherboard and buy it. Without getting into details which are hardware dependent, the Intel add-on board does no currently work with all systems, failing to initialize on boot. The original Intel motherboards won't run with an 80387 in the socket according to PC Week. A system without a 32bit memory bus is likely to be a hinderance in the future. Memory prices are sitting about $200/MB, and I suspect that they will take another big drop in the next 15 months, making it cost effective to add more than 16MB if you can connect it. A good time to take time. -- bill davidsen (wedu@ge-crd.arpa) {chinet | philabs | sesimo}!steinmetz!crdos1!davidsen "Stupidity, like virtue, is its own reward" -me
anton@utai.UUCP (08/06/87)
In article <786@unccvax.UUCP> cbenda@unccvax.UUCP (carl m benda) writes: >In article <1272@killer.UUCP>, robertl@killer.UUCP (Robert Lord) writes: >> Intel is coming out with a 20Mhz (whew!) 80386 chip later this year, and >> I would like to drop it into me AT to make it into a fast desktop. If I >> did this, would I have any problems with existing software/hardware? >> Would I be able to use software designed for the 386? Would all the > > >No No and Finally No. The problem with what you described is the fact that >your AT was probably designed to run at NO more than 10MHz. This is because >the memory is probably no faster than 120ns RAM. What this means is that if >you replace your 80286 with a 80386 inserted into a Chetah Adapter, it will >perform at a MIPS rate which is no better than your current configuration. > Since when did MIPS rate directly depend on the memory speed? This is not true for highly popelined architectures like the 386. Besides MIPS is not a good test of system performance. >What you will gain is the addressing capabilaty of the 80386, namely, 4G >bytes, and 1Meg segments, instead of 80286 which is 16Meg and 64K segments. !!!! 386 supports 4G segments. Are you one of those other uP lovers trying to proliferate myths about 386? >With this in mind, a 20Mhz 80386 chip would be wasted in your machine. > The problem with today's fast CPU's is that there is no cheap RAM to keep up with them. Hence, a cache is necessary so that memory will not degrade CPU performance. Since, an average instruction on the 386 takes longer than 1 bus cycle(=2 CPU cycles) it is possible to fetch the next instruction while the first is being executed using the pipelines in the 386. Hence, a 20Mhz CPU will not be a waste, it will still execute complex code faster while waiting for the memory a lot. The biggest problem with putting a 386 without a cache into an AT board is that the AT board has a 16 bit data bus. Therefore, each 386 32 bit fetch will take 2 bus cycles. A smaller problem is that most AT boards have slow (120nsec) memories and slow address decode logic. The biggest advantage of the 80x86 processors is their fast memory cycle. Each memory fetch requires only 2 CPU cycles vs. 3 or 4 for other uP's. However, to solve the memory problems above, the pipelining of in the CPU provides the address of the next fetch about 1/2 cycle before the next bus cycle. This allows for address decoding while the information from the previous fetch is on the bus. >What you would really like to do is keep everything in your AT but your >mother board. Replace the mother board with an 80386 motherboard which >will allow you to have a 20Mhz machine with 32 bit word memory moves, your >AT can only do 16bit memory moves. Here again memory for this machine will >be exspensive, I believe with a 20Mhz clock rate the memory needs to be >at least as fast as 70ns. This is a guess since my ALR 386 uses 80ns ram >and has a 16Mhz clock rate. > The solution suggested here is definitly the best in terms of performance. But some caution is advised. Consider a 20Mhz CPU. From above we get the maximum time to decode an address and fetch a word on the bus as 2.5cycles*50nsec clock=125nsec without wait states. This is not enough, even with the fastest decode logic and 80 ns RAM. In fact this is about half the time that is actually needed.(Have you ever wondered why there are no 0 wait state AT clones faster then 10Mhz). The cheapest solution is to let the 386 run with lots of wait states (2?). The next chepest is to use special RAM. We will call this the Compaq solution. Here you can address a whole bunch of words *in sequence* very quickly (i.e. no address decode neccessary). With pipelining in 386 this probably works out to an average of 1 wait state per bus cycle.( What is the biggest chunk of inline code and data that you have written?). This is where Motorola's idea of having separate Instruction and Data caches pays off. The best solution is to have a large (64K) fast (35ns) cache. The difference between 100ns and 80ns main memory RAMs becomes insignificant. The memory cycle as seen by the CPU becomes (assuming 80% hit ratio) 80%*35ns+20%*100ns=48ns which is barely acceptable at 20Mhz. Because of 386's pipelining the improvment over the Compaq solution is about 20%(as suggested by Intel literature and determined by tests of cache based 386 machines vs. Compaq in recent issues of Byte). Caches are expensive but so is nibble mode RAM for COmpaq. I hope that you realize what the moral of the story is. When you buy a 386 or any other fast CPU machine it is the memory sub- sytem that you are paying for. Consider a cached 386(as above) in a 10 MHz 0 wait state AT bus. AT bus cycle is 2.5cycles*100ns/cycle=250ns. 32bit cycle on the AT bus will be 500ns i.e.2 fetches. From the formula above the 386 cycle will be 80%*35ns+20%*500ns=128ns or almost good enough to run with 1 wait state. >hope this helps >/Carl >...decvax!mcnc!unccvax!cbenda
tankus@hsi.UUCP (Ed Tankus) (08/07/87)
In article <407@parcvax.Xerox.COM>, burton@parcvax.Xerox.COM (Philip M. Burton) writes: > In article <503@wolf.UUCP> enped@wolf.UUCP (Eric Pederson) writes: > > > >Most of the speed ratings put 386 machines with the same bus and the same > >clock speed slightly above the compared 286 machines. The added speed > >isn't neccesarily worth the extra cost in getting the 386 add-on. You'd > >want it most likely for the ability to run multiple copies of DOS, a nice > >linear UNIX, etc. etc. You can speed up your AT machine with a faster > >disk and a decent cache program. > > I agree. My preference for a new term solution would be to replce the 286 > with a 386, and leave everything else alone. EGA, AST multifunction card, > disk controller, etc. > > Has anyone actually installed the Cheetah upgrade? > . . . > > -- > Philip Burton > Xerox Corporation 408 737 4635 > ... usual disclaimers apply ... A friend of mine has the Cheetah Adaptor Board installed in his true blue m/n 339 AT. Initial installation caused a speed DECREASE. This is because Cheetah inserts an extra wait state in the board to compensate for the 386's 32-bit access vs the 286's 16-bit access. The board comes with a crystal and an adaptor for your 287. Your 287 must be removed and placed in this adaptor in order to be used. He had no problems with standard software packages such as word processors, etc. Caveats: My friend's applications are special. He is a statistician and researcher and thus requires additional features not available in a standard box. He has added the following features in addition to the Cheetah board: Ariel XCELX speed injector to increase overall system speed to at least 10Mhz. Microway 287Turbo board (JUST fits in Cheetah adaptor) to increase 287 clock speed to 12Mhz. Orchid Eccell board fully loaded with 100ns RAM. Motherboard RAM replaced with 100ns chips. PC-MOS/386 single user (soon to be 5 or 25 user). He has measured the system speed once (using the Ariel utilites that come with the XCELX product) to be 9.5 MHz. He has since claimed to go a bit beyond this but has not quantified those results. Under PC-MOS, he has generally found most DOS products to be completely compatible. He has trouble only with the following: Freelance - did not work at all. Drivers are supposed to be available shortly. Systat - a statistics package. Systat files install with a ".sys" extension. PC-MOS thinks these files are device drivers, not program files. He has to talk to the software developer about this one! Bernoulli Box - 10+10 does not install. It requires AT LEAST the 4.11 driver from Iomega. Beats me what he's got, but it's obviously not the latest and greatest. Generally, terminate-and-stay-resident (TSR) packages may have a problem, though Sidekick worked correctly. It may have to do more with where they are loaded in memory (PC-MOS) and when, I'm just not sure. Sorry for being so looooooooooooong winded. Hope this adds something to the discussion. Cheers! -- Ed. Net : {noao!ihnp4!yale!}!hsi!tankus Snail: Health Systems Int'l, 100 Broadway, New Haven, CT 06511 Bell : (203) 562-2101
ray@madnix.UUCP (Ray P. Hill) (08/08/87)
In article <407@parcvax.Xerox.COM>, burton@parcvax.Xerox.COM (Philip M. Burton) writes: > > Has anyone actually installed the Cheetah upgrade? > Has anyone attempted to use any Cheetah products with XENIX? I have a 286 class AT clone and would love to upgrade to a 386 machine and 386 XENIX. I don't care if its faster or not I just want to get away from large/huge model (ie. segmentation). Ray Hill -- UUCP: {harvard,seismo,rutgers}!uwvax!nicmad!madnix!ray {ihnp4,decvax,terminus}/