• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Pentium E5800 review...

/dev/null

/dev/null

[H]F Junkie
Joined
Mar 31, 2001
Messages
15,181
My google-fu has failed me.

Anyone know where I can find a review of this? I'd especially like to see it compared to athlon ii's x2s in Linux &/or gaming/computational crunching stuff. Not too interested in synthetic benchmarks.

I was pondering a celeron E3300 or E3400 (can't remember) at microcenter for $39.99 when I found this (after coupon) for $52 shipped @ ewiz. I figured the extra L2 & bump to 3.2ghz couldn't do anything but help. Considering I'd have to spend $10 in gas to get to microcenter & don't need any other parts, it was pretty much a no-brainer.

I picked this up for a P35 board I have sitting here empty except for 4G of ddr3. I'd like to use it as a linux desktop & possibly a dual boot to some version of Windows (for testing).

Also: Is overclocking as simply as bumping up FSB to 1066 assuming I have good cooling? I usually don't OClock (1090T, phenom 2 x4's...) but on a $50 cpu I'm not too concerned if it blows up.
 
don't know about the E5800, but i bought a E5400 with a cheap Gigabyte EP41-UD3L for shits & giggles a while ago to have as a backup / test platform.

spent and afternoon oc'ing to see how far i could push it (thus by no means an exhaustive test).

E5400%20(cpu-z).PNG


E5400%20(coretemp%20idle).PNG


essentially the same chip except the E5800 has a 16 multiplier.

at 13.5 x 282 (3807MHz) it was prime95 stable 9.5hrs (in place large FFT's). the gigabyte board i used (from the reviews i read about it) is by no means a great oc'er board. i didn't drop the multiplier on my chip b/c i knew from reviews it wasn't capable of high FSB. also i think i was only using the stock HS from my Q6600.

only made 3 changes to get the chip @3.8GHz;
-drop the FSB:MEM divider to get cheap 667 RAM is was using below its default speed.
-increased the vcore from 1.2125v to 1.280v
-increased NB volts to 1.2v (not sure what the default voltage was)
 
Last edited:
aphasia said:
don't know about the E5800, but i bought a E5400 with a cheap Gigabyte EP41-UD3L for shits & giggles a while ago to have as a backup / test platform.

spent and afternoon oc'ing to see how far i could push it (thus by no means an exhaustive test).

E5400%20(cpu-z).PNG


E5400%20(coretemp%20idle).PNG


essentially the same chip except the E5800 has a 20 multiplier.

at 13.5 x 282 (3807MHz) it was prime95 stable 9.5hrs (in place large FFT's). the gigabyte board i used (from the reviews i read about it) is by no means a great oc'er board. i didn't drop the multiplier on my chip b/c i knew from reviews it wasn't capable of high FSB. also i think i was only using the stock HS from my Q6600.

only made 3 changes to get the chip @3.8GHz;
-drop the FSB:MEM divider to get cheap 667 RAM is was using below its default speed.
-increased the vcore from 1.2125v to 1.280v
-increased NB volts to 1.2v (not sure what the default voltage was)
Click to expand...

I was thinking of something more like:

Set FSB to 1333
Set Multi to 10 = 3.33ghz (hopefully @ stock volts.
 
quick note:
was an error in my original post, quoted incorrect multiplier for E5800. should be x16 instead of x20.​

Robstar;
I was thinking of something more like:

Set FSB to 1333
Set Multi to 10 = 3.33ghz (hopefully @ stock volts.
Click to expand...

its doable, depends on the motherboard mostly i guess. i don't see why the CPU couldn't do it with little to no bump in vcore.

if the MB is rated for FSB 333 (1333 quad pumped), shouldn't be an issue. if the MB is only rated to FSB 266 (1066 quad pumped), the northbrige might need a bump in voltage.

course you'd have to drop the memory strap to NB if your RAM isn't rated for 1333 operation.
 
aphasia said:
quick note:
was an error in my original post, quoted incorrect multiplier for E5800. should be x16 instead of x20.​



its doable, depends on the motherboard mostly i guess. i don't see why the CPU couldn't do it with little to no bump in vcore.

if the MB is rated for FSB 333 (1333 quad pumped), shouldn't be an issue. if the MB is only rated to FSB 266 (1066 quad pumped), the northbrige might need a bump in voltage.

course you'd have to drop the memory strap to NB if your RAM isn't rated for 1333 operation.
Click to expand...

It's an EP35C-DS3R and it's DDR1333...
http://www.gigabyte.us/products/product-page.aspx?pid=2740#sp

Under specs it has official support for DDR3-1333
 
i don't see you having an issue doing what you want then.

you could probably push the chip a lot further than 3.3GHz if you wanted on stock cooling.
 
Don't expect to set your fsb to 333 and drop the multiplier. The E5xxx series has an FSB ceiling typically around 300-315, regardless of frequency.
 
ok for some more shits & giggles i attempted to do just as Robstar was wanting to do from his 1st post.
i booted up my E5400 and set my FSB to 333 and dropped the multiplier to 10 (3.33GHz). had to drop the NB strap to memory to accommodate my crappy 667 rated RAM.

Cpuz%20-%20E5400%20@%203333MHz%20(333x10).PNG

Coretemp%20-%20E5400%20@%203333MHz%20(333x10).PNG


NOTE: i didn't stress test my chip at this clock. i did however load up prime95 to kick the cores out of power saving mode (all those features are still enabled in the bios).

anyway Robstar, i contend that what you're after is possible. that's not to say that all E5xxx chips can run at 333FSB. i really haven't done my homework here. do some googling and decide for yourself. to get those screenshot above took me literally 5mins.
 
I might as well pick this CPU up. (e5800)

I have a Celeron 420 in my computer right now... Double the GHz and double cores.
 
An Update:

I've had this at stock a few days w/ no issues.

This morning I was toying with it before coming to work:

I went from 200 * 16 to 333 * 12 doing nothing other than changing the bus & multiplier (no voltage increases anywhere).

At stock a 2.6.37 kernel compile was about 11:26. At 333*12 it rang up a 9:32.

For comparison, doing a kernel compile on my work machine (Dual Xeon 5520) with 2 threads is at 10:30....(although moving that to 16 threads makes it 2:30 :) )

Another Update: It appears the board was choosing 1.55v when set to auto!! Eating 100% cpu @ 4.33ghz with a small cooler & stock case sensors showed my processors hitting 91c.

I've knocked it back down to 10 x 333 @ 1.175v and under 100% load I'm hitting 47c or so.
 
Last edited:
You must log in or register to reply here.
Back
Top