Crazy Problem! Does not Compute!!

How many RAM upgrades have you done?

  • 1

    Votes: 0 0.0%
  • none

    Votes: 1 14.3%
  • 2-5

    Votes: 0 0.0%
  • 6-15

    Votes: 3 42.9%
  • 16-100

    Votes: 1 14.3%
  • over 100

    Votes: 2 28.6%

  • Total voters
    7

cart0181

Distinguished
Jun 2, 2007
24
0
18,510
OK, this problem is crazy....
Background:
I have a Dimension 4300, which uses pc133 sdram. It came with 256MB installed and I want to upgrade to 768MB or the maximum 1024MB. There are 2 DIMM slots in this system. I have two sticks of Kingston KVR133X64C2/512 which are "known good." I've tested all the memory in my possession using this PC and Memtest x86 for about 8 hours each with no errors.
The Problem:
As soon as I put 2 dimms in at one time, the computer will freeze solid at the Welcome screen or black screen with mouse pointer. I have tested this several times with all possible combinations. Each stick works if it's by itself in either DIMM0 or DIMM1. I have tried removing the original 256MB and inserting two identical Kingstons. The problem corrects immediately if I remove one stick. Also, the BIOS recognizes everything properly.

Any ideas? :?
 
Double sided vs single sided? In addition to the number of sticks, you also have to worry about the number of banks. (ram with chips on both sides of the stick count as two banks.

I'm surprised the bios saw everything. The boards I've seen that take PC133 ram only support 256MB per stick. (three slots for a total of 768MBs)
 

azimuth40

Distinguished
Feb 13, 2007
102
0
18,680
look in the slot that was unused and see what you see. The socket could have been initially bad, some of the pins could have gotten coated with insulation material, a smoke or grease filled environment will do this. Crucial.com says that you can have 128 or 256 or 512 Dimms and they guarantee theirs to work so there is no reason that the Kingstons should not. Another possibilitywould be the on motherboard powersupply is not up to speed but this is 3.3v memory and should be coming from the main supply with a little filtering. Could have a bad filter Capacitor however that can't handle the strain under load.
 

cart0181

Distinguished
Jun 2, 2007
24
0
18,510
@4745454b
They are all 16 chip/double sided.
@azimuth40
Keep in mind that both slots work independently, just not together. The computer wasn't used in a greasy or smokey environment. The home I got it from was unusually clean.
The bit about the power supply sounds plausible. Whatever is causing this, I don't think it's simple. I wonder how I could test for this as a possibility. I would think Memtest would behave erratically or freeze also. Why just Windows? I'll have to try some more bootable disks tomorrow and see how that goes.

Thank you very much for the replies!! :D
 

cart0181

Distinguished
Jun 2, 2007
24
0
18,510
Well, it looks like I got it fixed. At least I hope so (I'm still paranoid). I did a BIOS update last night from version A04 to A06 and it looks like it took care of it. I booted it 3 times last night without any problems and once more this morning. I even put in the same RAM I originally wanted to put in there. So now I have the original 256MB that came with the PC, plus a "G-tek" 512MB I pulled from another computer. This is what the Dell website has to say about the BIOS updates:

A05 Release Date: 5/20/2002
This BIOS file corrects a bit in one the system registers and allows the Creative Labs Audigy sound card to function correctly when installed in the Dell Dimension 4300 system.
!*!*! WARNING !*!*!
DO NOT USE THIS BIOS ON ANY DIMENSION 4300 THAT DOES NOT HAVE A CREATIVE
LABS "AUDIGY" or "AUDIGY2" SOUND CARD.

The following changes have been made to the A05 BIOS to create A06:
Release Date: 6/6/2003

1. Updated microcode and SMBIOS support for newer processors.
2. Prevented boot-up of system when detecting a 533 FSB CPU.
3. Fixed possible incorrect setting of 'System Time' in Setup.
4. Fixed boot code to prevent problem with no boot devices or with a dormant
devices being removed.
5. Fixed hang problem with Creative Audigy and large sound samples.
6. Fixed issue when resuming from Hibernate with keyboard & mouse swapped.
7. Reduced POST time when no PS/2 keyboard or Mouse installed.
8. Fixed false error on Floppy Drive when running Diags From Bootable
Diags CD without a Floppy Drive.
9. Improved USB code in USB hubs processing. Also, fixed issues with some
USB devices causing a keyboard POST error with a USB keyboard.
10. Adjusted USB Emulation support for special keys, and non-US keyboards.
11. Added two new diagnostic LED codes for better memory failure isolation.
No Memory Detected (ABCD=Grn, Yel, Yel, Yel)
Memory Configuration Failure (ABCD=Grn, Yel, Grn, Yel)
12. Added IDE UDMA support to speed up resume from Hibernate (when enabled).
13. Added support for additional models in the 4300 Series.


As you can see, nothing is mentioned about any weird RAM-related problem that would cause you difficulty upgrading the RAM. In fact, the only thing I could see that had anything to do with lock-ups was the sound fix in rev A05 and that was for 4300s that had the Audigy sound cards. This one has the Live! Value card. Did I miss something there? Well, hopefully this problem won't rear its ugly head again. I hope this helps someone out there, but it's probably about a one in a million shot. :tongue: