P3B-F And 256 MB SD-Ram Sticks

  • Thread starter Thread starter Geetar
  • Start date Start date
G

Geetar

hi guys,

im desperately trying to get a ram stick with 256mb of capacity to run in my
p3b-f rev. 1.3 motherboard. i tried every manufacturer around. including
infineon and kingston. bios is already up to latest. what else to do?
thx
g.t.
 
hi guys,

im desperately trying to get a ram stick with 256mb of capacity to run in my
p3b-f rev. 1.3 motherboard. i tried every manufacturer around. including
infineon and kingston. bios is already up to latest. what else to do?

You are buying the wrong type of dimm. Consistently, it appears.

So go where you should have gone in the first place: www.crucial.com
Where they have a clue.

The Crucial part number you should order is CT32M64S4D7E.
The information page for that dimm on your board is here:

http://www.crucial.com/store/MPartspecs.Asp?mtbpoid=32146837A5CA7304&WSMD=P3B-F&WSPN=CT32M64S4D7E

/daytripper
 
The problem we ran into with P3BF's is they don't support double-sided
DIMMs.

Any BX chipset board requires that 256MB DIMMs be double-sided.

Daytripper is spot on with his Crucial.com suggestion, though I'm not
sure I'd hire him for phone support.
 
Any BX chipset board requires that 256MB DIMMs be double-sided.

Daytripper is spot on with his Crucial.com suggestion, though I'm not
sure I'd hire him for phone support.

It's true. I don't suffer fools well.

/daytripper (besides, you scamps couldn't afford me ;-)
 
With close to 2000 of these suckers at 80 sites, with various BIOS and board
revisions, it's best to expand the RAM with the least problematic answer.
 
With close to 2000 of these suckers at 80 sites, with various BIOS and board
revisions, it's best to expand the RAM with the least problematic answer.

Puhlease, you're making a spectacle of yourself. There were never any bios or
board revision dependencies on operating a gigabyte of memory on a P3B-F. All
you had to do was buy the proper dimms. Period.

While I have a mere four P3B-F systems running today, each system has four
double-sided 256MB dimms. The dimms were even bought from three different
manufacturers, only two systems are running a single brand of memory.

No big whoop. It just works. Like it always did.

Honestly, if you couldn't get double-sided 256MB dimms to work in a P3B-F,
with the result that your customers were throttled to 512MB or less across
close to 2000 systems, you probably shouldn't have had anything to do with
those systems in the first place...

/daytripper (hth ;-)
 
Trip,

Have you ever seen any problem trying to run @133 FSB with a full
gigabyte of memory on the P3B-F? I have one that will and one that
won't. The only obvious diffrence is the memory. The one that will
uses Mushkin and the one that won't runs 768 meg of Crucial but will
not run a full gig. The boards are same revision and bios. I guess I
will try swapping the memory at some point but have some reservations
about doing that, since both run as is.
 
Trip,

Have you ever seen any problem trying to run @133 FSB with a full
gigabyte of memory on the P3B-F? I have one that will and one that
won't. The only obvious diffrence is the memory. The one that will
uses Mushkin and the one that won't runs 768 meg of Crucial but will
not run a full gig. The boards are same revision and bios. I guess I
will try swapping the memory at some point but have some reservations
about doing that, since both run as is.

I built each of these systems using PC133 double-sided DIMMs, eventually
ending up with the four fully loaded systems a bit more than a year ago. No
problems. None of them have any problems, even the two systems with mixed
pairs.

I've recommended PC133 dimms for the P3B-F ever since they hit price parity
years ago (never mind that they've been *cheaper* than "PC100" dimms for well
over a year now). I've yet to have a P3B-F normally-clocked system have a
problem with that configuration...

cheers

/daytripper
 
Ron said:
Trip,

Have you ever seen any problem trying to run @133 FSB with a full
gigabyte of memory on the P3B-F? I have one that will and one that
won't. The only obvious diffrence is the memory. The one that will
uses Mushkin and the one that won't runs 768 meg of Crucial but will
not run a full gig. The boards are same revision and bios. I guess I
will try swapping the memory at some point but have some reservations
about doing that, since both run as is.

I frequently see that problem at 133 FSB on the P2B variants with 4
memory slots (different board, same chipset & manufacturer) - 768MB
works fine, but when you install the 4th DIMM the BIOS counts a
seemingly random amount of memory on every boot, but never counts the
full gig.

A trick that usually works for me is to go into BIOS setup, then
immediately exit without saving. The BIOS then counts the memory again,
and sees the full gig every time. It's a slight nuisance, but not too
much hassle if you rarely reboot. YMMV.

P2B
 
I frequently see that problem at 133 FSB on the P2B variants with 4
memory slots (different board, same chipset & manufacturer) - 768MB
works fine, but when you install the 4th DIMM the BIOS counts a
seemingly random amount of memory on every boot, but never counts the
full gig.

A trick that usually works for me is to go into BIOS setup, then
immediately exit without saving. The BIOS then counts the memory again,
and sees the full gig every time. It's a slight nuisance, but not too
much hassle if you rarely reboot. YMMV.

Ah - I missed the OP's *133mhz* part. Big difference between getting four
double-rank dimms to play nice on a P3B-F at 100mhz and doing the same at
133mhz. Which is pretty much the same behavior as you'd see on the 4-dimm P2B
variants at 133mhz.

Gotta be PC133, obviously, and be the best you can find at that...

/daytripper
 
daytripper said:
Ah - I missed the OP's *133mhz* part. Big difference between getting four
double-rank dimms to play nice on a P3B-F at 100mhz and doing the same at
133mhz. Which is pretty much the same behavior as you'd see on the 4-dimm P2B
variants at 133mhz.

Gotta be PC133, obviously, and be the best you can find at that...

/daytripper

Might not need to be the best you can find... I've tried a wide variety
of DIMMs, including the Mushkin the OP says works for him, and never had
a P2B see a gig at 133 without help from the 'BIOS trick' - but I do
have two P2B-DS systems running a gig of no-name PC133, and the DIMMs
don't even match each other! I leave the BIOS set to complain about
processor voltage on boot so I don't forget to 'help' it count the
memory. Still beats me why the trick works, but it does :-)

133Mhz does seem to be the limit, though. I can run 768MB of Corsair
PC150 C2 at 150Mhz on my P2B-S boards, but no way will they see more
than that past 133. The BIOS trick doesn't work at 140Mhz (and isn't
required at 124Mhz).

P2B
 
Other than the brand of memory the only other readily seen difference
is the voltage readings. On the system that will not run a full
bibabyte of memory, the negative 12volt and negative 5volt readings
are a little low. I'm not sure if this is significant or not. I will
load it back up with 1 gig and see if P2B's bios trick works. I
suspect this has something to do with power loadings and such, since
the board is running already and may allow more initial power to the
components. Just speculation on my part. Also wonder if the same
result could be obtained by using the reset button?
 
Anyone know what a "bibabyte" is? I must have flipper-flux. Obviously
"gigabyte" was the intended word.
 
Actually, the IEEE-correct word would be "gibibyte" ;-).
The negative voltages tend to be not very accurate, since they are
hardly used (voltage stabilization typically requires some minimum load
to work correctly). IIRC one is only used for serial ports (?), the
other one isn't used at all with todays board. Latest ATX power supply
specification drops them completely AFAIK (or maybe not both of them,
too lazy to look it up).

Roland
 
Ron said:
Other than the brand of memory the only other readily seen difference
is the voltage readings. On the system that will not run a full
bibabyte of memory, the negative 12volt and negative 5volt readings
are a little low. I'm not sure if this is significant or not. I will
load it back up with 1 gig and see if P2B's bios trick works. I
suspect this has something to do with power loadings and such, since
the board is running already and may allow more initial power to the
components. Just speculation on my part. Also wonder if the same
result could be obtained by using the reset button?

Voltage certainly plays it's part - although the memory runs on 3.3v
(nominal), regulated on the motherboard, so it's unlikely small
differences in power supply output voltages would be relevant to the issue.

P2B-S boards typically supply about 3.4v to the memory, while P2B-DS
boards (for some unknown reason) are down around 3.2v. I found it
necessary to increase memory voltage to 3.4v on -DS boards (resistor
change) before the BIOS trick would work.

I doubt it has to do with initial power supply loading either as even a
stripped system (CPU, RAM, PCI video only - no drives) requires the BIOS
trick to see all the memory, and hitting the reset switch doesn't work
no matter how long the board has been running.

Still a mystery to me....
 
Back
Top