lane@dalcs.UUCP (John Wright/Dr. Pat Lane) (09/25/88)
I have two old 4.77 MHz PCs...they're true blue IBM's...one is a "5150" which
is, I believe, a PC-2 and the other a "5160", an XT.
I'm wondering what I can do to jazz them up, particularly with regard to CPU
speed. Could I replace the CPU with an 8MHz 8088 or V20? Would I have to
change the crystal? how about the RAM memory?
Any suggestions on other things I could with these old clunkers (besides boat
anchors :-)
Technical details on the 5150:
Intel 8088 CPU, no co-processor
256K on motherboard (MOSTEK 8424/MK4564N-20 if I'm reading the chips (and my
notes) correctly...some chips have "8344" instead of "8424")
AST Six Pak Plus with 384K (MicroTek 4264-15's ...I presume this is 150nsec
memory while that on the motherboard is 200?)
PERSYST monochrome display adapter/printer port...would the RAM on this card
be an issue?
Process Traveller full height floppy and controller from "TPA"?
Miniscribe 10Meg hard drive (very noisy, even after I fiddled with the static
strap) with an IBM 1501492 controller...no format routine in ROM that
I can find
63.5 Watt power supply...would this not be a candidate for upgrading?
ROM BIOS dated 10/27/82
HD controller BIOS contains "5000059 (c) IBM 1982 08/16/82"
Also in ROM space "IBM PC Basic ver C1.10 copyr. 1981 25-Apr-81"
Technical details on the 5160:
AMD 8088 with no co-processor
256K on mother board...didn't record what the chips were
STB RIO PLUS II multifunction board with 384K RAM, uT4264-15s and MM4164A-15's
Hercules monochone graphics/printer
Tandon TM100-2A full height floppy with controller (don't know what kind)
Tandon TM502 hard disk with an Adaptec controller (BIOS has ACB2000 fmt pgm)
130W power supply
BIOS dated 11/08/82
What about changing the full height floppies for half height devices?
Much appreciate any advise or general info on what I've got here.
--
John Wright ///////////////// Phone: 902-424-3805 or 902-424-6527
Post: c/o Dr Pat Lane, Biology Dept, Dalhousie U, Halifax N.S., CANADA B3H-4H8
Cdn/Bitnet: lane@cs.dal.cdn Arpa: lane%dalcs.uucp@uunet.uu.net
Uucp: lane@dalcs.uucp or {uunet,watmath,utai,garfield}!dalcs!lane hoctor@osiris.cso.uiuc.edu (09/29/88)
>/* ---------- "Speeding up old PC's" ---------- */ >I have two old 4.77 MHz PCs...they're true blue IBM's...one is a "5150" which >is, I believe, a PC-2 and the other a "5160", an XT. > >I'm wondering what I can do to jazz them up, particularly with regard to CPU >speed. Could I replace the CPU with an 8MHz 8088 or V20? Would I have to >change the crystal? how about the RAM memory? > >Any suggestions on other things I could with these old clunkers (besides boat >anchors :-) > ><techical details ommitted> >/* End of text from osiris.cso.uiuc.edu:comp.sys.ibm.pc */ Given the age and ram speed of those old PCs (& XTs), do not use any speedup device that will increase the speed of the motherboard!!!! This will give you nothing but headaches. As far as installing a V20, the improvement is approx. 5% (in real life). The most effective means of speeding up these kind of boxes is to install and ASYNCHRONOUS speedup board. These sort of devices generaly run an 8 or 12 Mhz 286 on a seperate board (do you have an open slot?) that has its own 8 or 16K RAM CACHE. This sort of device will execute code in this high speed ram instead of the PCs slow system memory. These are VERY effective devices. Two such products that I am aware of are the PCSG (Personal Computer Support Group) Breakthru-286 and the Microsoft Mach 20 and Mach 30. If you have no available slots, I believe the Mach 20 and 30 can be had with a disk controller or printer/comm ports. Here where I work, I have purchased 3 of the 12 Mhz Breakthru-286s as well as having an 8Mhz version in my own machine. I can attest to the great improvement. The only problems that I had was with floppy disk accesses, but that is taken care of with a switch in their software. I have no interest in either of the companies. hoctor@osiris
Ralf.Brown@B.GP.CS.CMU.EDU (09/30/88)
In article <7700074@osiris.cso.uiuc.edu>, hoctor@osiris.cso.uiuc.edu writes: }Given the age and ram speed of those old PCs (& XTs), do not use any }speedup device that will increase the speed of the motherboard!!!! The RAM speed is the least of your worries. An original 4.77 MHz PC or XT would run just fine with 400 ns (not a typo--FOUR HUNDRED nanoseconds) memory chips, if any existed. The usual 200 ns chips would be fine up to 10 MHz. -- UUCP: {ucbvax,harvard}!cs.cmu.edu!ralf -=-=-=- Voice: (412) 268-3053 (school) ARPA: ralf@cs.cmu.edu BIT: ralf%cs.cmu.edu@CMUCCVMA FIDO: Ralf Brown 1:129/31 Disclaimer? I |Ducharm's Axiom: If you view your problem closely enough claimed something?| you will recognize yourself as part of the problem.
amlovell@phoenix.Princeton.EDU (Anthony M Lovell) (09/30/88)
In article <3009@dalcs.UUCP>, lane@dalcs.UUCP (John Wright/Dr. Pat Lane) writes: > I have two old 4.77 MHz PCs...they're true blue IBM's...one is a "5150" which > is, I believe, a PC-2 and the other a "5160", an XT. > I'm wondering what I can do to jazz them up, particularly with regard to CPU > speed. Could I replace the CPU with an 8MHz 8088 or V20? Would I have to > change the crystal? how about the RAM memory? > Any suggestions on other things I could with these old clunkers (besides boat > anchors :-) I have 2 suggestions. I performed these upgrades on my machine at my old job and the results were SO impressive, I now have a machine with the same toys. 1st - Get yourself an Intel INBOARDPC/386 card. It gives you 1 MEG ram and a 16 MHz 386. Also neat is a VERY GOOD disk cache program thrown in for nuthin. The machine FLIES. Probably the 8,10 or 12 MHz 286 boards are good if you want to save a few dollars, but these boards are pretty cheap (depending on memory prices). We got the 1st one 7 months ago for a paltry $650! We got to pull ALL RAM off the PC mother board (save 256K), so that's "free" too. Put it in another machine if you'd like. 2nd- I have not seen many hard drives which compare to the speed/capacity per dollar as the FlashCard49 from CompuAdd. ~$550 for a disk on a card with 49 Mbytes of 28 msec disk access. Combined with the disk caching software in the Intel bundle, you have a power-user machine from your old clunker for under $1500. Look around. If you can get the Intel device at any price NEAR this, do it. You'll leave the others in the dust without taking a soaking on the sale of your still viable PC boxes. -- amlovell@phoenix.princeton.edu ...since 1963.
zgel05@apctrc.UUCP (George E. Lehmann) (10/05/88)
In article <2343761d@ralf> Ralf.Brown@B.GP.CS.CMU.EDU writes: >The RAM speed is the least of your worries. An original 4.77 MHz PC or XT >would run just fine with 400 ns (not a typo--FOUR HUNDRED nanoseconds) memory >chips, if any existed. The usual 200 ns chips would be fine up to 10 MHz. Pardon my naivete, but did someone just change the definition of Mhz? Or do these old machines have one wait state (two clock cycles per memory access) built into the motherboard? -- George Lehmann, ...!uunet!apctrc!zgel05 Amoco Production Co., PO BOX 3385, Tulsa, Ok 74102 ph:918-660-4066 Standard Disclaimer: Contents are my responsibility, not AMOCO's.
berger@clio.las.uiuc.edu (10/08/88)
No - but I think that person is mistaken. I'd like to see how he
calculates that 400 ns rams should be sufficiently fast.
Mike Berger
Department of Statistics
University of Illinois
berger@clio.las.uiuc.edu
{convex | pur-ee}!uiucuxc!clio!bergerRalf.Brown@B.GP.CS.CMU.EDU (10/08/88)
In article <16800379@clio>, berger@clio.las.uiuc.edu writes: }No - but I think that person is mistaken. I'd like to see how he }calculates that 400 ns rams should be sufficiently fast. Here we go again.... (I think this should go in news.announce.newusers as a "commonly asked question"....) Rule-of-Thumb memory speed: (many memory chips actually perform better) max_access_time_ns = (1000 / clock_speed_MHz) * clocks_per_access / 2 where clocks_per_access is FOUR for 808[68] and TWO plus wait states for 80[23]86 machines. So, the calculation for an original PC goes: 1000/4.77 * 4 / 2 = 210 * 4 / 2 = 420ns and for a 10/1 AT 1000/10 * 3 / 2 = 100 * 3 / 2 = 150ns -- UUCP: {ucbvax,harvard}!cs.cmu.edu!ralf -=-=-=- Voice: (412) 268-3053 (school) ARPA: ralf@cs.cmu.edu BIT: ralf%cs.cmu.edu@CMUCCVMA FIDO: Ralf Brown 1:129/31 Disclaimer? I |Ducharm's Axiom: If you view your problem closely enough claimed something?| you will recognize yourself as part of the problem.
johne@hpvcla.HP.COM (John Eaton) (10/10/88)
<< < No - but I think that person is mistaken. I'd like to see how he < calculates that 400 ns rams should be sufficiently fast. ---------- Very simple. The 8088 and 8086 each take four clock cycles per memory access. At 4.77 Mhz thats 210 ns per clock or 840 ns for a complete cycle. The Address out is valid 110 ns after the start of T1 and the Data is required 30 ns before the end of the third. That gives you 490 ns to go through the address multiplexers, ram access and data buffers. If you drop in a 8 mhz part that goes up to 550 ns. I forget the exact tap values used in the IBM memory control circuit but I believe that it was fast enough to support 400 ns ram. John Eaton !hpvcla!johne
corbin@pinocchio.Encore.COM (Steve Corbin) (10/11/88)
In article <16800379@clio> berger@clio.las.uiuc.edu writes: > >No - but I think that person is mistaken. I'd like to see how he >calculates that 400 ns rams should be sufficiently fast. > > Mike Berger > Department of Statistics > University of Illinois I just took a quick look at the data books and here's what I got: It takes 4 clocks ticks for a bus cycle on the 8086/8088. The maximum address valid time is 110ns and the data setup to the trailing edge of T3 is 30ns. Without going into the PC schematics you can assume that there is probably an address buffer and a data buffer in the path. Assuming a 74LS373 for buffering/latching this adds another 54ns (2 * 27ns). At 4.77mhz each clock tick is 209.6ns for a total of 629ns to perform a read. 629-110-30-54 = 435ns. Depending upon how adversely the design of the control path affects the access time, it is entirely possible that 400ns rams will work. Disclaimer: The real answer can be found by a detailed timing analysis of the actual design which is far from what I did. Stephen Corbin UUCP: {bu-cs,decvax,necntc,talcott}!encore!corbin Internet: corbin@multimax.ARPA
stevel@eleazar.dartmouth.edu (Steve Ligett) (10/11/88)
There was discussion of how fast the ram chips have to be for a PC. Including: In article <3850@encore.UUCP> corbin@pinocchio.UUCP (Steve Corbin) writes: >In article <16800379@clio> berger@clio.las.uiuc.edu writes: >>No - but I think that person is mistaken. I'd like to see how he >>calculates that 400 ns rams should be sufficiently fast. > It takes 4 clocks ticks for a bus cycle on the 8086/8088. ... then many important lines that I've had to delete to get to the answer: > 435ns. > >Depending upon how adversely the design of the control path affects the >access time, it is entirely possible that 400ns rams will work. Here, let me try to fool inews into thinking I wrote Stephen's disclaimer: Disclaimer: The real answer can be found by a detailed timing analysis of the actual design which is far from what I did. >Stephen Corbin UUCP: {bu-cs,decvax,necntc,talcott}!encore!corbin Yes, that is correct as far as it goes. A look at the schematics reveals that there's 2 ls245s in the path, and well as the address mux circuit. The timing for the RAS and CAS isn't optimized for the circuit (just off-the-shelf parts). I didn't look at the parity checking. So, I'd be willing to believe (without a more detailed analysis) that 300 ns rams might work, and I'm pretty sure that the 250 ns ones in my PC work. Argh, now to try to trim this article down so that I can send it. (later ... I hope that does it, I've butchered the other articles.) Forgive me Mike and Stephen. Steve Ligett steve.ligett@dartmouth.edu or (decvax harvard ihnp4 linus)!dartvax!steve.ligett
leonard@bucket.UUCP (Leonard Erickson) (10/13/88)
In article <234e0a42@ralf> Ralf.Brown@B.GP.CS.CMU.EDU writes:
<Rule-of-Thumb memory speed: (many memory chips actually perform better)
<
< max_access_time_ns = (1000 / clock_speed_MHz) * clocks_per_access / 2
<
<where clocks_per_access is FOUR for 808[68] and TWO plus wait states for
<80[23]86 machines.
<
<So, the calculation for an original PC goes:
<
< 1000/4.77 * 4 / 2 = 210 * 4 / 2 = 420ns
<
<and for a 10/1 AT
<
< 1000/10 * 3 / 2 = 100 * 3 / 2 = 150ns
Sorry, but your formula produces bogus answers. As an example, it says that
a 10 Mhz XT clone can use 200 ns RAM. When you compare this with reality,
you find that you need at least 120 ns RAM to get anything resembling
reliable operation.
I could give other examples but the bottom line is that your formula gives
values that are roughly double the "real" answer...
--
Leonard Erickson ...!tektronix!reed!percival!bucket!leonard
CIS: [70465,203]
"I used to be a hacker. Now I'm a 'microcomputer specialist'.
You know... I'd rather be a hacker."Ralf.Brown@B.GP.CS.CMU.EDU (10/13/88)
}<and for a 10/1 AT }< }< 1000/10 * 3 / 2 = 100 * 3 / 2 = 150ns }I could give other examples but the bottom line is that your formula gives }values that are roughly double the "real" answer... Your experience must be exclusively with systems using extremely slow logic chips. My 9.8/1 AT clone works just fine with 150ns chips.... -- UUCP: {ucbvax,harvard}!cs.cmu.edu!ralf -=-=-=- Voice: (412) 268-3053 (school) ARPA: ralf@cs.cmu.edu BIT: ralf%cs.cmu.edu@CMUCCVMA FIDO: Ralf Brown 1:129/31 Disclaimer? I |Ducharm's Axiom: If you view your problem closely enough claimed something?| you will recognize yourself as part of the problem.
leonard@bucket.UUCP (Leonard Erickson) (10/16/88)
In article <235494b5@ralf> Ralf.Brown@B.GP.CS.CMU.EDU writes:
<}<and for a 10/1 AT
<}<
<}< 1000/10 * 3 / 2 = 100 * 3 / 2 = 150ns
<
<}I could give other examples but the bottom line is that your formula gives
<}values that are roughly double the "real" answer...
<
<Your experience must be exclusively with systems using extremely slow logic
<chips. My 9.8/1 AT clone works just fine with 150ns chips....
You should have paid more attention to what I posted. *my* example was a 10
Mhz *XT*, not an AT. And II've checked with some others. 10Mhz XT motherboards
WILL NOT work with 150 ns RAM.
ps. you really need to learn *not* to edit out the credit lines from your
followups...
--
Leonard Erickson ...!tektronix!reed!percival!bucket!leonard
CIS: [70465,203]
"I used to be a hacker. Now I'm a 'microcomputer specialist'.
You know... I'd rather be a hacker."ralf@b.gp.cs.cmu.edu (Ralf Brown) (10/17/88)
In article <1109@bucket.UUCP> leonard@bucket.UUCP (Leonard Erickson) writes: }In article <235494b5@ralf> Ralf.Brown@B.GP.CS.CMU.EDU writes: }<Your experience must be exclusively with systems using extremely slow logic }<chips. My 9.8/1 AT clone works just fine with 150ns chips.... } }You should have paid more attention to what I posted. *my* example was a 10 }Mhz *XT*, not an AT. And II've checked with some others. 10Mhz XT motherboards }WILL NOT work with 150 ns RAM. What could possibly be so slow that 150ns chips work with the 300ns cycle time of a 10MHz AT but not with the 400ns cycle time of a 10MHz XT? }ps. you really need to learn *not* to edit out the credit lines from your }followups... PS: I normally don't, but in this case I decided to post AFTER sending email, so I copied the screen contents into a file. I had stripped the "in article foo, you wrote", which is why the posted article didn't contain credit lines. Perhaps you need to learn not to get uptight about a one-time oversight? -- {harvard,uunet,ucbvax}!b.gp.cs.cmu.edu!ralf -=-=- AT&T: (412)268-3053 (school) ARPA: RALF@B.GP.CS.CMU.EDU |"Tolerance means excusing the mistakes others make. FIDO: Ralf Brown at 129/31 | Tact means not noticing them." --Arthur Schnitzler BITnet: RALF%B.GP.CS.CMU.EDU@CMUCCVMA -=-=- DISCLAIMER? I claimed something?
misha@nsc.nsc.com (Michael Umansky) (10/19/88)
In article <1109@bucket.UUCP> leonard@bucket.UUCP (Leonard Erickson) writes: >You should have paid more attention to what I posted. *my* example was a 10 >Mhz *XT*, not an AT. And II've checked with some others. 10Mhz XT motherboards >WILL NOT work with 150 ns RAM. >^^^^^^^^ NOT TRUE! Last week I put together an XT clone running at 10Mhz. It has 256-150ns and 64-150ns DRAMs and it has been up and running without any problems for over a week now. Maybe you had a bad experience with a bad batch of 150ns DRAMs. -- NAME: Michael Umansky; {amdahl|decwrl|hplabs|pyramid|sun}!nsc!misha WORK: National Semi; 2925 Copper Road; MS/7C266; Santa Clara, CA 95051 HOME: 4331 Lincoln Way; San Francisco, CA 94122 PHONE: (408) 721-8109 (work); (415) 564-3921 (home)
leonard@bucket.UUCP (Leonard Erickson) (10/19/88)
In article <3312@pt.cs.cmu.edu> ralf@b.gp.cs.cmu.edu (Ralf Brown) writes: <In article <1109@bucket.UUCP> leonard@bucket.UUCP (Leonard Erickson) writes: <}In article <235494b5@ralf> Ralf.Brown@B.GP.CS.CMU.EDU writes: <}<Your experience must be exclusively with systems using extremely slow logic <}<chips. My 9.8/1 AT clone works just fine with 150ns chips.... <} <}You should have paid more attention to what I posted. *my* example was a 10 <}Mhz *XT*, not an AT. And II've checked with some others. 10Mhz XT motherboards <}WILL NOT work with 150 ns RAM. < <What could possibly be so slow that 150ns chips work with the 300ns cycle time <of a 10MHz AT but not with the 400ns cycle time of a 10MHz XT? I think I've found the answer to all of this... In the Byte special issue on IBM there is a note about this. The 150ns is the *access* time. But DRAM requires a "recharge" time before the next access that is only slightly shorter than the access time. So reading one byte will work. Reading widely scattered bytes will work. But reading sequential bytes will quite likely fail. See page 102 of "BYTE IBM Special Edition" for details. I'd say the question isn't why our machines require the faster RAM, but rather how your machine manages to work with the slow RAM. -- Leonard Erickson ...!tektronix!reed!percival!bucket!leonard CIS: [70465,203] "I used to be a hacker. Now I'm a 'microcomputer specialist'. You know... I'd rather be a hacker."
del@Data-IO.COM (Erik Lindberg) (10/20/88)
In article <1109@bucket.UUCP> leonard@bucket.UUCP (Leonard Erickson) writes: >In article <235494b5@ralf> Ralf.Brown@B.GP.CS.CMU.EDU writes: ><Your experience must be exclusively with systems using extremely slow logic ><chips. My 9.8/1 AT clone works just fine with 150ns chips.... > >You should have paid more attention to what I posted. *my* example was a 10 >Mhz *XT*, not an AT. And II've checked with some others. 10Mhz XT motherboards >WILL NOT work with 150 ns RAM. > I have a 10 Mhz XT MB that works just fine with 150 ns RAM. I didn't check out the math to figure out where the error in caclulation is, but I know from years of experience that 10 Mhz (AT *OR* XT) should require < 120ns access times. Those same years of experience bring to mind the obvious solution to the question of how some machines can work at 10 Mhz using 150ns chips. It's called "wait states" folks! For those of you new to the term, it involves tugging on the "ready" line to the CPU which tells it to wait for slow memory to catch up. That is why my 10 Mhz clone runs with 150ns chips. It is also why it only gets a Norton SI of 1.9 when obviously: 10Mhz > 4.77Mhz * 2 Back in the good old days when you bought a memory board it had a jumper on it which allowed you to select 0, 1, 2, or sometimes more, wait states for that board. This allowed you to select the ram to match your budget, while still driving the CPU as fast as possible. PC clone board manufacturers seem to have dropped that, the probably figured we weren't smart enough to select the correct jumper. PS. I don't have the original article referred to, but I dug this up: >In article <235494b5@ralf> Ralf.Brown@B.GP.CS.CMU.EDU writes: ><}<and for a 10/1 AT ><}< ><}< 1000/10 * 3 / 2 = 100 * 3 / 2 = 150ns >< Certainly the arithmetic is correct here, but I don't see how this bears any resemblance at all to the timing data found in the Intel data book. Access time is not a simple function of the clock cycle time. There are such things as set up times and data sample times that must be observed. It is left as an exercise for the interested reader to calculate the actual access time required. -- del (Erik Lindberg) uw-beaver!tikal!pilchuck!del
amlovell@phoenix.Princeton.EDU (Anthony M Lovell) (10/26/88)
As I'd advised earlier on the same topic, the Intel 386 plug-in board is the way to go if you get a good price. Right now, I spotted a DEAL. Intel 386PC board w/ 1 Meg RAM, disk cache software, AND windows 386 cost : $799! This is from PC Connection, offer thru Nov 30. The value is obscene, though I have no affiliation with Intel whatsoever. Board works in XT, PC (true IBM) and gives you 16MHz of zip for a song. -- amlovell@phoenix.princeton.edu ...since 1963.
dennis@raphel.UUCP (Dennis Vogel) (11/03/88)
In article <4110@phoenix.Princeton.EDU>, amlovell@phoenix.Princeton.EDU (Anthony M Lovell) writes: > > Board works in XT, > PC (true IBM) and gives you 16MHz of zip for a song. Anyone have any idea if this board works with Taiwanese XT clones? Dennis R. Vogel AT&T Bell Laboratories Somerset, NJ