burton@mitisft.Convergent.COM (Philip Burton) (09/23/89)
In getting a no-name 386 system, it seems that the motherboard choice is very critical. Several months ago, PC Magazine reviewed about 110 386 systems. Among the manufacturers who used a "name brand" third party motherbaord, it seemed that most used AMI. A few used Micronics, and almost no one used Mylex. However, if I read the ads for motherboards in newspapers in the Bay Area, stores seem to carry Micronics and Mylex more than AMI. Comments anyone?
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (09/25/89)
In article <1235@mitisft.Convergent.COM>, burton@mitisft.Convergent.COM (Philip Burton) writes: | In getting a no-name 386 system, it seems that the motherboard choice is very | critical. Several months ago, PC Magazine reviewed about 110 386 systems. | | Among the manufacturers who used a "name brand" third party motherbaord, it | seemed that most used AMI. A few used Micronics, and almost no one used | Mylex. I think you will find that the choice of MB is not nearly as critical as it was at the time the 386 clones first became available. Any of the three brands you mention seem to run UNIX/Xenix just fine (hardware compatibility) and the Award, AMI, and Pheonix BIOS sets seem to allow good DOS functionality. I have heard a lot of bad things about other boards, but because they are often sold under several names and with various BIOS chips, I don't want to say "this is a bad one." If you look at the cost of boards in the back of _PC Week_ or _Computer Shopper_ you will see that AMI sells for a higher price. I don't know if this is because they cost more in qualtity or not (I got quotes on Q11 and they did) or if people are willing to pay more for them. If money was no object I would buy AMI (personal opinion, I have had one since Dec 86). Either of the other will probably be perfectly fine. Unless you're doing an upgrade of an existing machine, you might be able to get a complete machine for within $100 of the same base price, and add options as needed. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon
cliffhanger@cup.portal.com (Cliff C Heyer) (09/26/89)
All I can say is do *lots* of homework before you buy! For example, someone out there is supposed to be making a board with SCSI directly connected to the CPU bypassing the AT bus...so you can get "real" I/O up to 900KB/sec rather than 250KB/sec like all the other boards..
ralf@b.gp.cs.cmu.edu (Ralf Brown) (09/27/89)
In article <22538@cup.portal.com> cliffhanger@cup.portal.com (Cliff C Heyer) writes: }there is supposed to be making a board }with SCSI directly connected to the }CPU bypassing the AT bus...so you can }get "real" I/O up to 900KB/sec rather }than 250KB/sec like all the other Except that the AT bus is not a bottleneck until you get way beyond current disks and controllers. Running at the equivalent of 8 MHz/1 ws with a 16-bit controller, you can pump 5.3 megabytes per second across the bus.... (3 clocks for every 2 bytes, at 125 ns per clock cycle). HD controllers with hardware caches regularly get data transfer rates of over 3 megs per second (and that includes BIOS and other overhead at least once for every 64K transferred). -- {backbone}!cs.cmu.edu!ralf ARPA: RALF@CS.CMU.EDU FIDO: Ralf Brown 1:129/46 BITnet: RALF%CS.CMU.EDU@CMUCCVMA AT&Tnet: (412)268-3053 (school) FAX: ask DISCLAIMER? |"Humor is laughing at what you haven't got when you ought to What's that?| have it." -- Langston Hughes
caf@omen.UUCP (Chuck Forsberg) (09/27/89)
In article <22538@cup.portal.com> cliffhanger@cup.portal.com (Cliff C Heyer) writes:
:All I can say is do *lots* of homework
:before you buy! For example, someone out
:there is supposed to be making a board
:with SCSI directly connected to the
:CPU bypassing the AT bus...so you can
:get "real" I/O up to 900KB/sec rather
:than 250KB/sec like all the other
:boards..
Hmmmm, I get 900 kb/sec (according to CORETEST) on a number of
286 and 386 boxes using either an ESDI 2322 or SCSI 1540 Adaptec controllers.