halff@utah-cs.UUCP (Henry M. Halff) (06/30/86)
Several clones advertise that they use a 8088-2 processor. Can anyone tell me what the difference is between an 8088-2 and an 8088 is? Thanks, hh -- Henry M. Halff Halff Resources, Inc. halff@utah-cs.ARPA 4918 33rd Road, N., Arlington, VA 22207
rob@gould9.UUCP (rob) (07/01/86)
--
Rob Gwaltney San Diego,CA
{cbosgd,ihnp4,sdcsvax,ucla-cs}!gould9!rob
{seismo!s3sun,hplabs!hp-sdd,sun!suntan}!gould9!rob
rob%gould9.uucp@NOSC.ARPA jnl@inuxh.UUCP (07/02/86)
> > -- > > What does -2 mean in 8088-2? > > > > Rob Gwaltney San Diego,CA > > > {cbosgd,ihnp4,sdcsvax,ucla-cs}!gould9!rob > {seismo!s3sun,hplabs!hp-sdd,sun!suntan}!gould9!rob > rob%gould9.uucp@NOSC.ARPA *** REPLACE THIS MESS WITH YOUR MESS *** Intel 8088 comes in different speeds. -3 means 4.77 MHz and -2 means 8.00 MHz John N. Le (AT&T IS/CPL - Indpls, IN) ..!inuxh!jnl
pavlov@hscfvax.UUCP (840033@G.Pavlov) (07/02/86)
In article <3842@utah-cs.UUCP>, halff@utah-cs.UUCP (Henry M. Halff) writes: > Several clones advertise that they use a 8088-2 processor. Can > anyone tell me what the difference is between an 8088-2 and an > 8088 is? > An "8088-2" includes functions/circuits that an 8088-based machine requires a additional (support) chips for. While an 8088-2 -based machine also requires support chips, the overall logic chip count is reduced substantially - thereby reducing cost, size, and (usually) failure points. g. pavlov, fstrf, amherst, ny.
kim@mips.UUCP (07/07/86)
>In article <3842@utah-cs.UUCP>, halff@utah-cs.UUCP (Henry M. Halff) writes: >> Several clones advertise that they use a 8088-2 processor. Can >> anyone tell me what the difference is between an 8088-2 and an >> 8088 is? >> >An "8088-2" includes functions/circuits that an 8088-based machine requires a >additional (support) chips for. While an 8088-2 -based machine also requires >support chips, the overall logic chip count is reduced substantially - thereby >reducing cost, size, and (usually) failure points. > g. pavlov, fstrf, amherst, ny. Hold on, there! The 8088-2 part is functionally the same as the 8088. It is spec'd to run at up to 8 MHz, whereas the 8088 is only spec'd to run up to 5 MHz. G. Pavlov's reply sounds like s/he was describing the 80188 part, which does indeed pull alot of the required outboard support logic onto the chip (clock gen, timers, rupt controller, DMA, etc). The 80188 also provides a *few* additional instructions (push-all/pop-all, variable length shifts, etc). The clones that Mr. Halff refers to likely run at 8 MHz, though they probably have a switch or mode in which they run at 4.77 MHz (necessary for *true* compatibility with brain-damaged s/w that uses s/w timing-loops). To anticipate a probable question ... yes, you can replace an 8088 with an 8088-2, *BUT* it will not increase the performance of your system, as the processor will still be clocked at the 4.77 MHz rate (for a true PClone). Increasing the clock on a PClone is an "iffy" proposition, as the rest of the system was probably not designed for higher speed operation. A very reasonable way to boost vanilla PClone performance a bit (10-15%, overall), is to replace the 8088 with a NEC V20 processor. While the clock will still be 4.77 MHz, the NEC chip performs many operations in fewer cycles than does the 8088. As a bonus, the V20 includes the additional instructions that the 80188 has, plus it includes several "new" instructions that provide bit operations, packed decimal arithmetic, and a few others. The V20 also has an "emulation- mode" for the 8080 uP instruction set, and can execute 8080 code directly (of course appropriate control and dispatch s/w is needed to use this in a system). There is an EXCELLENT two-part article on the V20 (and the V30, which is the 8086 replacement) in the Nov/Dec '85 and Jan/Feb '86 issues of "Micro/Systems Journal" if anyone is looking for more information. Currently, V20's go for about $20 (look at the ads in the back pages of BYTE), though I've seen them locally for around $15. As a suggestion, if you plan to make this upgrade, try to get the 8 MHz version of the V20 (even if it will only be run at 4.77 MHz); it seems that there *is* a slight difference in the spec for the clock duty-cycle that *could* cause a problem (erratic operation, etc.) if the 5 MHz part is used (the NEC part-numbers for the 8 MHz parts are: uPD70108C-8 and uPD70108D-8 for plastic and ceramic packages, respectively). Only problem I've had using a V30 in place of an 8086 is a couple of programs that use those brain-damaged s/w timing-loops I mentioned earlier. Gurrrr!!! Disclaimer: I don't work for (or have any connection with) Intel, NEC, BYTE, or Micro/Systems Journal in any way except that I've used their chips and read both magazines. I *do* think that $15-$20 for a 10%-15% performance boost on a 8086/8088 based machine is one of the best values around today. /kim -- UUCP: {decvax,ucbvax,ihnp4}!decwrl!mips!kim DDD: 408-720-1700 x231 USPS: MIPS Computer Systems Inc, 930 E. Arques Av, Sunnyvale, CA 94086 CIS: 76535,25
herman@ti-csl (07/08/86)
> Several clones advertise that they use a 8088-2 processor. Can > anyone tell me what the difference is between an 8088-2 and an > 8088 is? Clock rate. The 8088-2 is an 8 MHz part. The standard 8088 has a maximum clock rate of 5 MHz. ================================================================ *hardcopy* *electr{onic, ic}* Herman Schuurman ARPA: herman%TI-CSL@CSNET-RELAY.ARPA POB 225474 M/S 238 CSNET: herman@TI-CSL Texas Instruments Inc. USENET: {ut-sally,convex!smu,texsun,rice}!ti-csl!herman Dallas, Texas 75265 VOICE: (214) 995-0845