e5200 @ 3.4ghz

xxvv

xxvv

xxvv

Associate
Joined
7 Mar 2009
Posts
52
e520034.png


e520034load.png


Furthest i've pushed my e5200 so far, had it running at 3.33 for ages but fancied a change. Unfortunately my board - g31-es2l has limited memory options (+0.1,.2,.3,.4) so i'm not entirely sure on the next step.

I want to get 3.5ghz stable do i just have the ram at .4 and let it rise with the fsb? (ram is corsair 1066mhz). in terms of volts how far should i go, i've tried 1.36v and that failed to boot, how much higher is safe? Temps are good so no worries there.

Edit - 3.4 - 1.3v
 
Last edited:
Your RAM timings need setting in manual to stop them going all over the place once you pass your RAMs rated speed of 533MHz.
 
unfortunately my mb doesn't allow you to set manual timings, there are only 4 pre determined options. Is there any software or utility that allows me to control the ram timings?

3.4ghz @ 1.3v stable -

e520034load9.png
 
I'm not sure there is as you would have to have the option in the BIOS for the utilty to work - and that type of utility is never advisible when clocking, ideally.

Have you thought of sellling your MB and buying a second hand Asus p43/45 board as your cpu looks like it has the potetntial to be good clocker? It, obviously, wouldn't make financial sense but i get the impression it would be worth it just for your own experimental enjoyment? Plus, you would have the added bonus that you would have no problem clocking your future upgrades.
 
Last edited:
micro atx case = matx board which limits me to expensive g43,q/g45 boards which may or may not have manual memory settings.

what volts are you putting through your e5200 to get 3.5?
 
xxvv said:
micro atx case = matx board which limits me to expensive g43,q/g45 boards which may or may not have manual memory settings.
Click to expand...

Whoops - missed that.

Then, I personally think you're not doing too badly for a confined space and limited BIOS values. You have a rig that potentially could get very hot and limits you with BIOS settings yet you've balanced it so you can run at 3.33Ghz - i think you've played well with the hand you have.

xxvv said:
what volts are you putting through your e5200 to get 3.5?
Click to expand...

I'm on my laptop at the mo but if i remember correctly i'm not running it much over stock - probably in the region of 1.18xxV (i'll post back with results when i'm at my desktop later.)
 
Plec said:
Whoops - missed that.

Then, I personally think you're not doing too badly for a confined space and limited BIOS values. You have a rig that potentially could get very hot and limits you with BIOS settings yet you've balanced it so you can run at 3.33Ghz - i think you've played well with the hand you have.



I'm on my laptop at the mo but if i remember correctly i'm not running it much over stock - probably in the region of 1.18xxV (i'll post back with results when i'm at my desktop later.)
Click to expand...

yea my case the coolermaster 340 is very good for modding, at only £30 it has great potential, if you remove the hard drive cage and mount the hard drive in a 5.25 bay with heatsinks (reduce heat and noise) you can fit two 120mm fans at the front -

img0016ed8.jpg


this isn't mine but you get the idea, i've also cut out all the grills for pure airflow using a dremmel, replaced the northbridge with a zalman heatsink and fitted an accelero tt on the hd4850 (the two main heat issues in the case).

back to the point :) i wouldn't mind getting 3.5ghz but after hiking the volts up from 1.3 to 1.36 i can't get it to post.

will try a bit higher and see if i can get it stable.
 
Impressive airflow mod - you've certainly made the most of what you've got.

With that in mind, is 0.17Ghz worth all that extra voltage and potential stability problems when you already have a stable and significanylty clocked system?

I'm guessing 3.5Ghz has become one of those targets that seems attainable but for one, potentially, small setting? - yet you seem to have tried all that you can with your memory restrictions.

I'll try d/ling your manual later (if it's online) but from reading previous posts in other theads i think you may have hit your barrier - and it's your memroy settings.

Good luck with it - i admire your determination, i hope you find a work around.
 
memset works on your chipset, problem you are going to run into is that the settings are not applied till your in windows. if your memory is stopping you from getting into windows at higher overclocks then it wont really help you.

it is very handy for tweaking the timings on a stable overclock though ;)

http://www.tweakers.fr/memset.html
 
I think you've hit the nail on the head with that one, 3.5ghz is just a milestone so is it really worth sacrificing stability and temps for the 0.1ghz?

I'll have a go with the memory timings, although i'm not entirely sure what i should be doing with the timings -

memset.png


any ideas?
 
your top timings need to be 5-5-5-5-15, with the lower ones its just trial and error tbh.

lower them 1 at a time and only in single steps, run everest memory tests for a quick glance at any improvement and stability. some settings have a greater effect than others so just play around. your pc may lock up using this, if it does just reboot and reapply the settings that were working before it locked up.

tip: write down what you change and what to.
 
spookywillow said:
your top timings need to be 5-5-5-5-15, with the lower ones its just trial and error tbh.

lower them 1 at a time and only in single steps, run everest memory tests for a quick glance at any improvement and stability. some settings have a greater effect than others so just play around. your pc may lock up using this, if it does just reboot and reapply the settings that were working before it locked up.

tip: write down what you change and what to.
Click to expand...

Orthos didn't like those timings,

mem1.png
 
Last edited:
you can but you aren't going to see what improvement each setting does, some will make you lose performance so you do really need to test each setting.

everest has a trial version which lets you run the tests and see the results, i do it this way as you can quickly see what settings are best and then once happy i run memtest for a few hours and see what its like.

to make the settings run at startup you need to click save, to cancel the settings you need to click del. save (appears once you click save) but dont click ok on the next window or it will put the setting back at boot, you need to click cancel.
 
Last edited:
memset.png


I edited the post after you had answered my questions, unfortunately orthos didn't like those settings, i've clocked back down to 3.33 so the memory runs at 1066 and applied the timings you mentioned, i presume the spd is the what the memory settings should be?


Edit - quick look at the corsair website indicates the timings should be 7-7-7-20 but even after 1 minute stress this fails :(
 
Last edited:
according to ocuk site you ram runs at 5-5-5-15, not sure why corsair says its 7-7-7-20.

EPP#2 is what your timings should be at 1066mhz, you have set the vdimm to 2.1 haven't you?

on top of that your other timing may be affecting the stability, i was going to say that your performance level is pretty low. i can't get anywhere near that low before the pc crashes.
 
interesting one this -

35ghz.png


couldn't get it to post at 3.5ghz, so i turned the voltage control and memory control on auto and it's booted straight into 3.5ghz with core at 1.18v :confused:

is it fine to use the auto function or will this be detrimental? running orthos to check stability.
 
seems stable, current problem is the ram timings, can't get it stable in orthos, don't know if it's just the program being funny, tried that memory test in everest worked fine -

went back down to 3.4 and changed all the settings back to stock, the only thing i can't change is the refresh period tREF (see below) i've just selecting the closest number. even this doesn't run stable in orthos but runs fine in everest :confused:

memeve.png


any idea..
 
You must log in or register to reply here.
Back
Top Bottom