Welcome, Guest. Please login or register.

Author Topic: A3640 V3.0 refuses to run speed test on SysInfo V3.24 ?!  (Read 8577 times)

Description:

0 Members and 1 Guest are viewing this topic.

Offline EffyTopic starter

  • Hero Member
  • *****
  • Join Date: Aug 2002
  • Posts: 2053
    • Show only replies by Effy
A3640 V3.0 refuses to run speed test on SysInfo V3.24 ?!
« on: December 24, 2005, 01:41:50 PM »
There is a problem that I can not solve or don“t have enough experience to fix it :

When I run SysInfo v3.24, there is no Speed showing for the CPU nor is there info alongside the FPU. The Amiga crashes just as it is about to list the final scores.However, if I use a CSMKII 060/50 or any of the three different C= 030/6882 cards I have, SysInfo always shows the CPU/FPU info before you run the speed test. This card doesn't!!! When Sysinfo is launched, it should automatically tell you what CPU/FPU is in the Amiga. As I have said, despite doing a fresh Install of the OS, the 040 Card will NOT show the CPU/FPU details. However, if i use ANY of my other cards, the CPU/FPU info always shows. I have tried several versions of the 040 Library & they all do the same thing. SysInfo Crashes just before it lists its final scores. It gets as far as showing comparative speeds against other Amiga Models but crashes with Guru Meditation just as its about to show the Mips/Flops results.

The question is whether the card is faulty or not. Any intelligent answer might be welcome. Other non intelligent answers which are useful are of course also welcome  :lol:

Offline Piru

  • \' union select name,pwd--
  • Hero Member
  • *****
  • Join Date: Aug 2002
  • Posts: 6946
    • Show only replies by Piru
    • http://www.iki.fi/sintonen/
Re: A3640 V3.0 refuses to run speed test on SysInfo V3.24 ?!
« Reply #1 on: December 24, 2005, 01:46:48 PM »
SysInfo is really old and has all sort of bugs in it, some which will crash it randomly on 040 and 060 systems. Crashes are normal with this program. You're better off by not using SysInfo at all.

Try SysSpeed.

PS. MIPS is useless info really, and you can't compare MIPS numbers of different programs. Real life benchmarks give you better idea of the performance.
 

Offline EffyTopic starter

  • Hero Member
  • *****
  • Join Date: Aug 2002
  • Posts: 2053
    • Show only replies by Effy
Re: A3640 V3.0 refuses to run speed test on SysInfo V3.24 ?!
« Reply #2 on: December 24, 2005, 01:52:51 PM »
Is there any reason why it would crash on an V3.0 and not on an V3.1 A3640 ???  :-?

Offline Piru

  • \' union select name,pwd--
  • Hero Member
  • *****
  • Join Date: Aug 2002
  • Posts: 6946
    • Show only replies by Piru
    • http://www.iki.fi/sintonen/
Re: A3640 V3.0 refuses to run speed test on SysInfo V3.24 ?!
« Reply #3 on: December 24, 2005, 02:02:23 PM »
Probably there is.

Just use SysSpeed ok?
 

Offline Kin-Hell

  • Hero Member
  • *****
  • Join Date: Dec 2005
  • Posts: 624
    • Show only replies by Kin-Hell
Re: A3640 V3.0 refuses to run speed test on SysInfo V3.24 ?!
« Reply #4 on: December 24, 2005, 02:42:12 PM »
Effy, the C= 040 card that crashes in Sysinfo is a v3.1
Getting 0lder is Mandatory..... Growing up is an option.
 

Offline EffyTopic starter

  • Hero Member
  • *****
  • Join Date: Aug 2002
  • Posts: 2053
    • Show only replies by Effy
Re: A3640 V3.0 refuses to run speed test on SysInfo V3.24 ?!
« Reply #5 on: December 24, 2005, 02:50:48 PM »
No, the one that causes the crash is a V3.0, the one that is not supposed to work in an A3000. SysInfo does run with the card but when you want to do a speed test then it crashes with a Guru  :-o

Offline EffyTopic starter

  • Hero Member
  • *****
  • Join Date: Aug 2002
  • Posts: 2053
    • Show only replies by Effy
Re: A3640 V3.0 refuses to run speed test on SysInfo V3.24 ?!
« Reply #6 on: December 24, 2005, 03:16:46 PM »
Okay, after a lot of confusement it seems to be that the crashing A3640 is a V3.1 and not a V3.0 as I first thought  :lol: Guess I have too much red wine in my blook  :pint: But still I would like to know why it crashes when you want to know how fast it is  :-?

Offline Piru

  • \' union select name,pwd--
  • Hero Member
  • *****
  • Join Date: Aug 2002
  • Posts: 6946
    • Show only replies by Piru
    • http://www.iki.fi/sintonen/
Re: A3640 V3.0 refuses to run speed test on SysInfo V3.24 ?!
« Reply #7 on: December 24, 2005, 03:25:28 PM »
@Effy

Get over it already. SysInfo is buggy as hell. Use some other program.
 

Offline AmiDude

  • Hero Member
  • *****
  • Join Date: Oct 2005
  • Posts: 903
    • Show only replies by AmiDude
Re: A3640 V3.0 refuses to run speed test on SysInfo V3.24 ?!
« Reply #8 on: December 24, 2005, 03:36:06 PM »
@Effy

Is it a "LC" or "EC" 68040 CPU? Maybe Sysinfo has
problems with some versions, because I've got a full
68040 on my A3640 rev.3.1 and Sysinfo isn't crashing
when running the speed-test.

 :-)
 

Offline EffyTopic starter

  • Hero Member
  • *****
  • Join Date: Aug 2002
  • Posts: 2053
    • Show only replies by Effy
Re: A3640 V3.0 refuses to run speed test on SysInfo V3.24 ?!
« Reply #9 on: December 24, 2005, 04:49:25 PM »
The truth is that this V3.1 is a card that I sold to Kin-Hell. I just asked him to run Showconfig and/or SysSpeed V2.6 ...

Offline Kin-Hell

  • Hero Member
  • *****
  • Join Date: Dec 2005
  • Posts: 624
    • Show only replies by Kin-Hell
Re: A3640 V3.0 refuses to run speed test on SysInfo V3.24 ?!
« Reply #10 on: December 26, 2005, 08:02:06 AM »
@ Piru

With all respect m8, there is nothing to get over here!
SysInfo v3.24 was available back in the early 90's & that programme worked flawlesly on my A4000D back then. Further, it was a C= A3640 3.1 card which I overclocked to 33Mhz using a 66Mhz Timing Crystal. She ran at 23.5Mips, which was 4.5Mips over-stock.
Sysinfo 3.24 never witheld the CPU's info on programme launch or crashed for it's speed test on that 040 Card!?

In the meantime, I've Emailed JJ B @ Amiga France, though I don't expect to hear from him until the New Year now.
As of New Years Eve, I won't be around till jan 14th anyway.
Thanks for all the feedback so far fellas, will yap some more in the New Year perhaps.
Happy New Year folks! :pint:
Getting 0lder is Mandatory..... Growing up is an option.
 

Offline Framiga

  • Hero Member
  • *****
  • Join Date: May 2003
  • Posts: 4096
    • Show only replies by Framiga
Re: A3640 V3.0 refuses to run speed test on SysInfo V3.24 ?!
« Reply #11 on: December 26, 2005, 12:33:27 PM »
EDITED - sorry


 

Offline nasty

  • Sr. Member
  • ****
  • Join Date: Nov 2005
  • Posts: 460
    • Show only replies by nasty
Re: A3640 V3.0 refuses to run speed test on SysInfo V3.24 ?!
« Reply #12 on: December 26, 2005, 12:50:02 PM »
Sorry to go abit off-topic here, but can someone explain  LC" or "EC to me?  My mate tryed to explain to me but it kind of went in one ear and out the other :lol:
Specs:
A1200PT, mediator c/w voodoo 3 2000, sb128, 60G HDD,OS3.9,LITEON CDR, nec 4x4 changer,External Scan Doubler,Cocolino adapter,PS2 keyboard adaptor.

Morphos 2.6, mini mac g4 @ 1.25ghz.
 

Offline Framiga

  • Hero Member
  • *****
  • Join Date: May 2003
  • Posts: 4096
    • Show only replies by Framiga
Re: A3640 V3.0 refuses to run speed test on SysInfo V3.24 ?!
« Reply #13 on: December 26, 2005, 01:06:00 PM »
Have you tryed the patch in Piru' Exec44 archive for SysInfo?

SysInfo 3.23
------------

SysInfo MFlops test access memory memory below 0x4000. This will barf with
exec44 + CyberGuard since this memory is marked invalid. Patch provided
fixes this.

Version information:

SysInfo 3.23 (1993-07-18)
(18-Jul-93)
 

Offline CatHerder

  • Full Member
  • ***
  • Join Date: Nov 2004
  • Posts: 193
    • Show only replies by CatHerder
Re: A3640 V3.0 refuses to run speed test on SysInfo V3.24 ?!
« Reply #14 on: December 26, 2005, 01:49:48 PM »
Quote

nasty wrote:
Sorry to go abit off-topic here, but can someone explain  LC" or "EC to me?  My mate tryed to explain to me but it kind of went in one ear and out the other :lol:


Certainly.

The "LC" means LowCost (literally). It was a 68040 chip produced by Motorola to fit a cost niche that Apple wanted for their MacIntosh computers and that Commodore scooped up to be able to sell 4000's for $65 less. The 68LC040 includes an FPU but lacks a MMU. (Memory Management Unit) It actualy DOES include an FPU. (I listed a whack of 68LC040's on eBay earlier and included incorrect information: out of curiosity yesterday I went and read the specs from Motorola explaining the differences...) I used to know this 10 years ago and forgot...

The "EC" means EmbeddedController, and it lacks both the MMU and the FPU and includes some embedded instruction sets. It uses less power, it runs cooler, and it was cheaper to produce and to buy. (It was a really good choice for the average 3000 desktop user who didn't plan on running Sculpt4D or Lightwave etc.)

The difference between a XC68040RC and an XC68EC040RC is the EC is missing the FPU portion of the chip as well as the MMU. This is very similar to comparing a 68030 Amiga that lacks a 68882 to an Amiga that has a 68030 and a 68882 installed. (The 68882 is the FPU.)

Will your Amiga run slower with an LC chip? Absolutely not. Will it run slower if you use specific applications requiring, or benefitting from an FPU? NO. The only time you will notice ANY problems is if you intend to run Unix on your Amiga (an A3000UX for example because Unix requires a MMU). Other than that, there is no "bad thing" about having an LC chip in an Amiga.

If you want to do renderings, play Quake, or do other applications that are very math intensive, then you want to upgrade your 68EC040 to a 68040 chip. If you only use your Amiga to play games, to edit music, to paint pictures, to type letters, to  do 99% of everything available out there for a stock Amiga then you don't need to worry about having an EC chip.  

People that say the LC chip is not a "real" 68040 (when talking about them in relation to an Amiga) are just misinformed or confused is all. :-)

68040 - includes both CPU FPU and MMU
68LC040 - includes both CPU and FPU
68EC040 - includes only the CPU and some custom embedded controller routines

(This all can be misleading, because there were a few batch runs of Motorola 68LC040 that were full 68040 chips (they were stamped wrong), and there were a few batch runs of LC that were EC chips and there is no way to know without testing each chip yourself.) Sucks, but that's the way it is . :-)
[color=000099]CatHerder[/color][/i]
Go Graphical Website Design is what I do.
My eBay World <- Amiga swag, if any.