E2180 + Abit IP35-E - which is limiting FSB?

Jimbo Mahoney

Jimbo Mahoney

Jimbo Mahoney

Soldato
Joined
18 Oct 2002
Posts
6,672
[Preamble]
-----------------------------------------------------------
I'm aware that high FSB is not the key performance factor. I was bored with no new toys to play with, so I decided to tinker with the current setup again and found some new things.

As some of you may know, on the Intel platform, the Static Read Delay, aka tRD, has a fairly hefty impact on memory latency and bandwidth. This is also what is known as the 'Performance Level' shown in MemSet and on a lot of boards, is not changeable in the BIOS. It is linked to the FSB / memory divider aka NB Strap. Of course, if you are lucky enough to have the option in the BIOS, none of this is really a problem, but for those of us with NB straps that change when we change the memory divider, it's bit of a pain.

It is with this in mind that I went about trying to find the max FSB to try to attain the lowest tRD.
------------------------------------------------------------
[/Preamble] :D

Basically, I cannot get over 350 Mhz FSB no matter what, and this seems a low wall. I've heard of other E2xxx's getting stuck at 370 etc.

Here's what I've tried:

Running 1:1 (memory 'underclocked')
NB volts (it's watercooled)
CPU VTT volts
GTLREF volts
CPU underclock

Does this mean it's the CPU? I guess I will only find out by trying another. I'm pretty sure the motherboard should be good for 400Mhz+.
 
Damn, 400 and 401 don't work either.

I guess the chip is just a poor one? It really struggles to get to 3.4 needing about 1.6v.

350 does seem a very low FSB though (although, given the chip's default is 200, I suppose it looks pretty high from where the CPU is :D)

EDIT - Reading a review, the reviewer at first had an E4300 max out @ 425Mhz, then an E6320 max out @ 4888 Mhz on the same board, so I guess I've got a poor CPU, which is hopefully the case, rather than a poor board.
 
Last edited:
The board should be fine for near 500mhz, so it's most likely the chip.

Tho with a x10 multi it's not the end of the world.
 
I could't get my q6600 over 3.0ghz with my IP35-e no matter what I tried..... (333x9) I tried suicide volts, changing multis and nothing worked at all.....

One day I decided to change my trfc setting to 42 for my memory and my rig flew up to 3.3ghz without breaking a sweat, 8 hours prime stable.

I'd say its a memory setting holding you back as it was for me.....
 
It could be the board, the -E is the lowest out of the ip35-e ip35 ip35 pro wasn't it? The other two are more stable; the ip35 has solid state capacitors aound the cpu, and the pro everywhere.
 
The IP35-e has solid state caps too.....

I'll bet its ram timings holding him back, they need to be all manually set, including all the trfc, trc settings etc.

The auto timings can be a bit too aggressive when fsb is raised.....
 
 
my attempt

E218036Ghz.jpg


3.7 was a no go.
 
6-6-6-10 is an unusually tight timing on DDR2 RAM, try setting it to 15 or 18 and see if u can get to 3.7 that way.

At 1.65v what cooler are u using?
 
You must log in or register to reply here.
Back
Top Bottom