🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Thread View: comp.sys.ibm.pc.hardware.misc
11 messages
11 total messages Started by David Longley Fri, 15 Dec 1995 00:00
Endeavour, P120 & 256K CELP Benchmarks ??
#721
Author: David Longley
Date: Fri, 15 Dec 1995 00:00
36 lines
1999 bytes
Do the following benchmakes seem correct for a 

Pentium 120
256K CELP
32MB EDO RAM

The Cache figures seem low?

public domain version COMPTEST  2.59 

                      computer type: IBM AT or compatible
                                CPU: Intel Pentium
                    clock frequency: 122.73 Mhz
                          bus width: 32 bit
                          CPU-cache: 1. level: 1 KB, 2. level: 4 KB

maximum RAM thruput (without cache): 59473 KB/s (effective wait states: 3.5)
       CPU cache thruput: 1. level: 596591 KB/s, 2. level: 477273 KB/s

                      system memory: 640 KB
                  available for DOS: 638 KB
   permanently used by DOS and TSRs: 136 KB

                    extended memory: 31744 KB (INT 15h thruput: 36300 KB/s)
                    expanded memory: 32064 KB (EMS 4.0, thruput: 32199 KB/s)

                          other RAM: C8000-C9FFF (  8 KB)
                                     CC000-DBFFF ( 64 KB) EMS-frame
                                     EC000-ECFFF (  4 KB)
                                     
                    BIOS-extensions: C0000-C7FFF ( 32 KB) VGA-BIOS
                                     CA000-CB7FF (  6 KB)
                                     ED000-EDFFF (  4 KB)

-- 
David Longley
Re: Endeavour, P120 & 256K CELP Benchmarks ??
#715
Author: Rajesh Varia
Date: Sat, 16 Dec 1995 00:00
43 lines
2269 bytes
In article <819061281snz@longley.demon.co.uk>
           David@longley.demon.co.uk "David Longley" writes:

> Do the following benchmakes seem correct for a 
> 
> Pentium 120
> 256K CELP
> 32MB EDO RAM
> 
> The Cache figures seem low?
> 
>                           CPU-cache: 1. level: 1 KB, 2. level: 4 KB
> 
> maximum RAM thruput (without cache): 59473 KB/s (effective wait states: 3.5)
>        CPU cache thruput: 1. level: 596591 KB/s, 2. level: 477273 KB/s
> 
>                     extended memory: 31744 KB (INT 15h thruput: 36300 KB/s)
>                     expanded memory: 32064 KB (EMS 4.0, thruput: 32199 KB/s)
> 
> -- 
> David Longley
> 
It seems to be reporting the incorrect cache sizes.
The following is for my P100, 256k 15ns sync, 32 meg 70ns FPR.
I am also currently testing a new M/B which gives 30% better preformance
with the same memory.

                          CPU-cache: 1. level: 8 KB, 2. level: 256 KB

maximum RAM thruput (without cache): 43034 KB/s (effective wait states: 4.1)
                  CPU cache thruput: 1.level: 367133 KB/s, 2.level:129925 KB/s

                    extended memory: 14336 KB (INT 15h thruput: 29801 KB/s)
                    expanded memory: 14656 KB (EMS 4.0, thruput: 27661 KB/s)

                  Dhrystones/second: 79941.9 (CPU: 219.2-fold of XT)
    Double-Precision Kilowhetstones: 44927.5 (FPU: 462.8-fold of XT w/ 8087)
            Double-Precision MFLOPS: 3.884 (FPU: 312.1-fold of XT w/ 8087)

 COMPTEST  2.59  (c) 1988-1993 N.J. 

-- 
Rajesh Varia     raj@c4rv.demon.co.uk
Re: Endeavour, P120 & 256K CELP Benchmarks ??
#717
Author: John Beardmore
Date: Sat, 16 Dec 1995 00:00
32 lines
1503 bytes
In article <819061281snz@longley.demon.co.uk>
           David@longley.demon.co.uk "David Longley" writes:

>Do the following benchmakes seem correct for a 
>
>Pentium 120
>256K CELP
>32MB EDO RAM
>
>The Cache figures seem low?
>
>public domain version COMPTEST  2.59 
>
>                      computer type: IBM AT or compatible
>                                CPU: Intel Pentium
>                    clock frequency: 122.73 Mhz
>                          bus width: 32 bit

Wot ??  Only 32 bit ??


>                          CPU-cache: 1. level: 1 KB, 2. level: 4 KB

Naah !

From what I've seen, comptest is quite good with old machines but falls
apart totally on new kit.

I don't know if there are newer releases that are pentium ready ?

-- 
John Beardmore
Re: Endeavour, P120 & 256K CELP Benchmarks ??
#708
Author: Rajesh Varia
Date: Sun, 17 Dec 1995 00:00
16 lines
1220 bytes
In article <30d41bbb.6452768@distort.demon.co.uk>
           jezhall@frimley.demon.co.uk "Jeremy Hall" writes:

> Rajesh Varia <raj@c4rv.demon.co.uk> wrote:
> >I am also currently testing a new M/B which gives 30% better preformance
> >with the same memory.
> >
> What is the motherboard and where can you get it in the uk. TIA.
> 
Not yet, this board was given to me by a possible importer.

> Jeremy Hall
> 

-- 
Rajesh Varia     raj@c4rv.demon.co.uk
Re: Endeavour, P120 & 256K CELP Benchmarks ??
#718
Author: Jeremy Hall
Date: Sun, 17 Dec 1995 00:00
7 lines
903 bytes
Rajesh Varia <raj@c4rv.demon.co.uk> wrote:
>I am also currently testing a new M/B which gives 30% better preformance
>with the same memory.
>
What is the motherboard and where can you get it in the uk. TIA.

Jeremy Hall
Re: Endeavour, P120 & 256K CELP Benchmarks ??
#719
Author: obiwan
Date: Sun, 17 Dec 1995 00:00
10 lines
848 bytes
In article <819123345snz@c4rv.demon.co.uk>,
   Rajesh Varia <raj@c4rv.demon.co.uk> wrote:

>  ...
> COMPTEST  2.59  (c) 1988-1993 N.J. 
> ^^^^^^^^^^^^^^^^^

Use Comptest 2.60.

ftp://ftp.aimnet.com/pub/users/rickl/
Re: Endeavour, P120 & 256K CELP Benchmarks ??
#710
Author: Rajesh Varia
Date: Mon, 18 Dec 1995 00:00
20 lines
1330 bytes
In article <819154806snz@wookie.demon.co.uk>
           wookie@wookie.demon.co.uk "John Beardmore" writes:

> In article <819061281snz@longley.demon.co.uk>
>            David@longley.demon.co.uk "David Longley" writes:
> 
> >                      computer type: IBM AT or compatible
> >                                CPU: Intel Pentium
> >                    clock frequency: 122.73 Mhz
> >                          bus width: 32 bit
> 
> Wot ??  Only 32 bit ??
> 
> -- 
> John Beardmore
> 
As you well know, the external bus on the pentium is 32-bit.

-- 
Rajesh Varia     raj@c4rv.demon.co.uk
Re: Endeavour, P120 & 256K CELP Benchmarks ??
#711
Author: John Beardmore
Date: Mon, 18 Dec 1995 00:00
18 lines
1369 bytes
In article <819232967snz@c4rv.demon.co.uk>
           raj@c4rv.demon.co.uk "Rajesh Varia" writes:

>In article <30d41bbb.6452768@distort.demon.co.uk>
>           jezhall@frimley.demon.co.uk "Jeremy Hall" writes:
>
>> Rajesh Varia <raj@c4rv.demon.co.uk> wrote:
>> >I am also currently testing a new M/B which gives 30% better preformance
>> >with the same memory.
>> >
>> What is the motherboard and where can you get it in the uk. TIA.
>> 
>Not yet, this board was given to me by a possible importer.

How does it do it ?

-- 
John Beardmore
Re: Endeavour, P120 & 256K CELP Benchmarks ??
#712
Author: Mr. Who
Date: Tue, 19 Dec 1995 00:00
44 lines
2424 bytes
Rajesh Varia wrote:
> 
> In article <819154806snz@wookie.demon.co.uk>
>            wookie@wookie.demon.co.uk "John Beardmore" writes:
> 
> > In article <819061281snz@longley.demon.co.uk>
> >            David@longley.demon.co.uk "David Longley" writes:
> >
> > >                      computer type: IBM AT or compatible
> > >                                CPU: Intel Pentium
> > >                    clock frequency: 122.73 Mhz
> > >                          bus width: 32 bit
> >
> > Wot ??  Only 32 bit ??
> >
> > --
> > John Beardmore
> >
> As you well know, the external bus on the pentium is 32-bit.
> 
> --
> Rajesh Varia     raj@c4rv.demon.co.uk

Sorry, nice try.  The Pentium has a 64-bit data bus.  Think about
this with me for just one minute.  72 Pin Simms are 32 bits wide
(+ parity).  These _must be installed into Pentium MBs in pairs so
that they can be interleaved into a single 64 bit wide data path.

If you are referring to the PCI bus, then yes in its current
incarnation (rev 2.01) it is a 32 bit bus.  However with an
upcoming enhancement to the bus it will _also_ be 64 bits wide.

Thought I'd straighten out the 32-bit and 64-bit -ness of it all.


-- 
+-------------------------------------------+
| Mr. Who is:                               |           
| johnk@sigg.com (Business)                 |
| jkotches@interaccess.com (Personal)       |
|                                           |
| New:  The Mr. Who home page               |
| http://homepage.interaccess.com/~jkotches |
+-------------------------------------------+
Re: Endeavour, P120 & 256K CELP Benchmarks ??
#713
Author: Rajesh Varia
Date: Tue, 19 Dec 1995 00:00
30 lines
1902 bytes
In article <30D66364.747C@interaccess.com>
           jkotches@interaccess.com "Mr. Who" writes:

> Rajesh Varia wrote:
> > 
> > As you well know, the external bus on the pentium is 32-bit.
> > 
> > --
> > Rajesh Varia     raj@c4rv.demon.co.uk
> 
> Sorry, nice try.  The Pentium has a 64-bit data bus.  Think about
> this with me for just one minute.  72 Pin Simms are 32 bits wide
> (+ parity).  These _must be installed into Pentium MBs in pairs so
> that they can be interleaved into a single 64 bit wide data path.
> 
> If you are referring to the PCI bus, then yes in its current
> incarnation (rev 2.01) it is a 32 bit bus.  However with an
> upcoming enhancement to the bus it will _also_ be 64 bits wide.
> 
> Thought I'd straighten out the 32-bit and 64-bit -ness of it all.
> 
OK, I missed out a word, the external data bus is 32-bit (ie the one
to the peripherals).  Moving to the newer 64-bit PCI bus will not
solve the resistriction.

The memory bus is indeed the full 64-bit.  The internal architecture is 
a mixture of 32-bit & 64-bit.

-- 
Rajesh Varia     raj@c4rv.demon.co.uk
Re: Endeavour, P120 & 256K CELP Benchmarks ??
#707
Author: David Longley
Date: Sat, 23 Dec 1995 00:00
107 lines
4825 bytes
In article <819657956snz@c4rv.demon.co.uk>
           raj@c4rv.demon.co.uk "Rajesh Varia" writes:

> In article <819650704snz@longley.demon.co.uk>
>            David@longley.demon.co.uk "David Longley" writes:
> 
> > Hmmmm  this directory does not exist. It does down to  users  and 
> > that's  it. As far as I can make out there is no such thing as  a 
> > version 2.60!
> > -- 
> > David Longley
> > 
> I've mailed it to you.
> 
> -- 
> Rajesh Varia     raj@c4rv.demon.co.uk
> 
Hmmmm...that's better:

32MB EDO
120 MHz P120
"256K cache on a stick" 
DS 64 VRAM PCI - 2MB
ADAPTEC 2940 PCI
Conner 1.2GB EIDE
Seagate Hawk 2GB SCSI


public domain version MMM COMPTEST  2.60 Page 1 

                      computer type: IBM AT or compatible
                                CPU: Intel Pentium
                    clock frequency: 122.73 Mhz
                          bus width: 32 bit
                          CPU-cache: 1. level: 8 KB, 2. level: 256 KB

maximum RAM thruput (without cache):  59467 KB/s (effective wait states: 3.5)
        CPU cache thruput: 1. level: 415020 KB/s, 2. level: 116675 KB/s

                      system memory: 640 KB
                  available for DOS: 640 KB
   permanently used by DOS and TSRs: 126 KB

                    extended memory: 31744 KB (INT 15h thruput: 38079 KB/s)
                    expanded memory: 32064 KB (EMS 4.0, thruput: 35853 KB/s)

                          other RAM: CD000-CFFFF ( 12 KB)
                                     D4000-DBFFF ( 32 KB)
                                     DC000-EBFFF ( 64 KB) EMS-frame
                                     EC000-ECFFF (  4 KB)
                                     EE000-EFFFF (  8 KB)
                                     F8000-F9FFF (  8 KB)
                                     
                    BIOS-extensions: C0000-C7FFF ( 32 KB) VGA-BIOS
                                     C8000-CC7FF ( 18 KB)
                                     ED000-EDFFF (  4 KB)
                                     
public domain version MMM COMPTEST  2.60 Page 2 

               parallel ports: 1
                 serial ports: 2 (COM1: 16550A, COM2: 16550A)
     mathematical coprocessor: Intel Pentium (clock frequency:121.97 MHz)
                        mouse: NOT INSTALLED
                games adaptor: INSTALLED

                         DOS drives: 6  (A:, B:, C:, D:, E:, F:)
                      floppy drives: 1  (1 x 1.44 MB 3+")
                         hard disks: 2

                 graphics card: Video Graphics Array (VGA)
         video-RAM wait states: 12
speed of video output via BIOS: 372869 characters/sec
 speed of video output via DOS: 166667 characters/sec (without ANSI driver)
                   DOS version: 6.22

                  Dhrystones/second: 94728.5 (CPU: 259.8-fold of XT)
    Double-Precision Kilowhetstones: 72727.3 (FPU: 749.1-fold of XT w/ 8087)
            Double-Precision MFLOPS: 6.107 (FPU: 490.7-fold of XT w/ 8087)

public domain version MMM COMPTEST  2.60 Page 3 

hard disk 1               cylinders: 618
                   read/write heads: 64
                  sectors per track: 63
                   storage capacity: 1275789312 Byte (1216.69 MB)

           track-to-track seek time:   7.30 ms
                  average seek time:  13.70 ms
                  maximum seek time:  23.00 ms
                    maximum thruput: 6702 KB/sec


hard disk 2               cylinders: 261
                   read/write heads: 255
                  sectors per track: 63
                   storage capacity: 2146798080 Byte (2047.35 MB)

           track-to-track seek time:   0.10 ms
                  average seek time:   0.10 ms
                  maximum seek time:   0.00 ms
                    maximum thruput: 8363 KB/sec (using disk cache)



Does this seem about right?
-- 
David Longley
Thread Navigation

This is a paginated view of messages in the thread with full content displayed inline.

Messages are displayed in chronological order, with the original post highlighted in green.

Use pagination controls to navigate through all messages in large threads.

Back to All Threads