E6850 L718A overclocks

[jpm]

looks like a bunch of us who got the first batch of shipments in the US ended up with L718A chips. thought it might be useful to see what kind of results we're getting out of this batch.

L718A190, pack date 6/27/07 - have to admit i am fairly underwhelmed so far. first stabilizing at 9*400=3.6Ghz. chip can survive basic usage at voltages as low as 1.35V but won't stand a chance in a long torture test (dual orthos + 3dmark06 loop + hdspeed). needs about 1.425V before torture tests pasts, and i haven't even had a chance to verify that past about 4 hours yet. so far this could pass for a decent E6700.

gets fairly hot too - core reaches some 58C after hours of torturing, of course that is on air and in closed case, as those are my constraints for this system.

oddly enough it looks like it can stand torture at 9*367=3.3Ghz at nice low voltages e.g. 1.275V, which gives some much better temps. for some reason this thing seems to want exponential increases to voltage past 3.3Ghz. but there is a lot of tweaking still to be done so i won't be writing it off yet.

other results?

[OBR]

Why you are starting another thread? We have many G0 (E6850) threads here ...

[jpm]

i saw at least 6 people on this forum mention that they just received E6850 from batch L718A and it would be useful to see the results, if only for future reference of people buying particular batches of E6850.

[Aldy402]

i saw at least 6 people on this forum mention that they just received E6850 from batch L718A and it would be useful to see the results, if only for future reference of people buying particular batches of E6850.

i agree, thanks for sharing
In light of these new fsb 1333 chips , we're all categorizing them as g0 and its already well documented that not ALL of them are monster overclocking chips. reporting the batch helps us differentiate

I probably have 1 of these also, ill post results when my e6850 gets here friday

what cooling are you using jpm and where did u buy from?

[jpm]

i agree, thanks for sharing
In light of these new fsb 1333 chips , we're all categorizing them as g0 and its already well documented that not ALL of them are monster overclocking chips. reporting the batch helps us differentiate

I probably have 1 of these also, ill post results when my e6850 gets here friday

what cooling are you using jpm and where did u buy from?

exactly... honestly what i need from this system is a moderate overclock at low voltage, which is potentially as hard to obtain as an aggressive overclock at high voltage.

cooling is with an ultra-120 extreme. got good contact (both surfaces are fairly flat) with a thin layer of as5, and that is after remounting several times, so i don't think the cooling is holding things back. it's a bigger issue that i've packed this into a lian-li pc-s80 with other silencing equipment, which complicates the air cooling (but did i mention it is very, very quiet?). this is actually a fairly interesting system in terms of design tradeoffs, i'll post photos at one point but first i'd like to get the clocks stabilized... for now, just collecting data.

i bought from priceguidenetwork.com and received it on 6/19. i'm already seeing chips from batches L721A and later, in fact this is the earliest batch i am yet to see codes for on the retail chips.

[scwam]

Do you have different ram to try it on, otherwise try one stick.

[ChaosMinionX]

So the famed G0 isnt the monster we had hoped for.... still waiting to see further batches of this stepping before I buy into it.

[[cTx] Nooc]

So the famed G0 isnt the monster we had hoped for.... still waiting to see further batches of this stepping before I buy into it.

DAMN YOU INTEL

I WAS HOPING FOR A MONSTER CPU AND YOU DIDNT MEET MY STANDARDS

GO TO HELL INTEL

[ChaosMinionX]

Here you go maybe this will help you a bit


[IMG]<a href="http://photobucket.com" target="_blank"><img src="http://i65.photobucket.com/albums/h229/malvare1/untitled5.jpg" border="0" alt="Photo Sharing and Video Hosting at Photobucket"></a>[/IMG]

Well, I took the time to copy/paste the image link, only too see that you have helped me squint

Need better resolution, I cant tell anything from that.

[hlonipha]

Well, I took the time to copy/paste the image link, only too see that you have helped me squint

Need better resolution, I cant tell anything from that.

Sorry about that I had to fix photobucket preference lets see now on air 3 days working with it

[michaLcoughliN]

im hoping to have my new parts this week :P and ill defintaly post what I can do with my Zalman 9700 lawl..

[road-runner]

Heres mine...

CPU: E6850
OC: 3969
Stepping: G0
Week: L718
Code: A190
Vcore: 1.55 volts

Cooling: Water/Fuzion

Motherboard: GA P35 DQ6
Bios Version: F5H
Operating System: XP Pro 32 bit

RAM brand: G Skill HZ
RAM rated speed: PC6400
RAM actual speed: 441
Latency: 5-5-5-15

Suicide 4401

[michaLcoughliN]

my rig in specs.. has been shipped today.. waiting for it to arrive and ill post my clocks

[luihed]

Can somebody please test max fsb even just screenie stable.... Thanks.....

Nice clocks btw guys.....

[jpm]

i will post some more results this weekend, as i've been doing some long torture test burn-ins all week. since i need a stable, quiet, low-heat, closed-case system, i haven't been pushing the higher clocks -- right now it looks like i have a torturable configuration at 8*413=3.3 and 1.28 vcore. i do know that this chip can at least boot into windows at 4 ghz on air, but i'm fairly sure it will need serious volts (>1.5 vcore, 1.55 nb, 1.20 sb, etc) and better cooling (airflow) to survive real torture.

[Grinch]

this is what I have got using just "auto" under the cpu voltage so far...I also added 15C to my temps with speedfan:

[Forsaken1]

Can somebody please test max fsb even just screenie stable.... Thanks.....

Nice clocks btw guys.....

Here you go.

[jpm]

well i've had a week with this E6850 L718A and i'm warming up a bit to it. with a reasonable amount of effort i can push this to 9*425=3.825ghz on air, with 4gb memory hitting 1020mhz / 5-4-4-12, orthos stable:



... but doing so required pumping 1.575V (bios) into the core and 2.15V into the memory, neither of which i am quite excited about doing for a few reasons.

personally i think i will tend to stick with a much cooler configuration at 8*413=3.3ghz @ 1.26V with memory at a 1:1 ratio 825mhz / 5-4-4-12:



low voltage has advantages -- i'm actually very satisfied with those temps, considering that this is in closed case on air. less heat means i can afford to reduce fans and airflow, which is a big part of this case being extremely quiet.

[michaLcoughliN]

not as impressive as i thought these chips would be. If mine clocks the same im selling it for a Q6600 since they are hitting 3,6 on air.

[zsamz_]

not as impressive as i thought these chips would be. If mine clocks the same im selling it for a Q6600 since they are hitting 3,6 on air.

well i'm happy with 4.0 i'm sure i can do 4.2
by the time i'll need a quad i'll get the 8 core cpu

[adamsleath]

well; i reinstalled my 6850 and the temps are now thru the roof (LOL)
1.5v is enuf to push my chip past 85C tjunction and fail dual orthos very quickly.

initial erroneous/low??? temps have all disappeared. I now have ctemp95/94 and speedfan 4.32 and TAT and ASUS probe giving me the SAME readings. (after swapping out my 4300)

currently 1.37V at 32C @ 20C ambient @ IDLE.
now at 450x8 dual orthos ok; tuniq is as tight as it will go.

Dual Orthos takes me to around 68-70C @ LOAD@ 1.4V/1.35droop. - and that is at 20C ambient!!

So my formerly very cool chip is now apparently not so cool - even though the heatsink base does not feel warm at all....all my coretemp type readings are now quite high.

So; this chip still needs very good cooling to achieve its potential.

Otherwise it's obviously got a good FSB potential - I have run it at 500x7 - i seem to reach optimum sandra score at 480x7...@8200MB/s

But i'll need better cooling in summer where ambient is 8-10C higher

[jpm]

not as impressive as i thought these chips would be. If mine clocks the same im selling it for a Q6600 since they are hitting 3,6 on air.

agreed, at least my chip has not panned out to be the low voltage, low temp, high clocking beast that people were touting.

i am minimally satisfied with my chip, because finding an E6x00 chip that does 3.3ghz at 1.26V on air would not have been a good bet. but if my goal was highest possible clocks, i would be disappointed.

but i am still interested to see what the other results from this and other batches are. intel is binning heavily and i'm sure some lucky people will receive some golden chips, even possibly in this same batch.

[lazy]

seems like these chips want some voltage if you wanna go above 3.6GHz

mines a L719B batch and im currently running 9*400 on water, but only had it on for 1day currently running on orthos for few hrs @home

just wish i had a better memory divider on my mb

[bigcat]

Mine is L718A, and same experience, wants too much voltage to be stable beyond 3.6Ghz.

seems like these chips want some voltage if you wanna go above 3.6GHz

mines a L719B batch and im currently running 9*400 on water, but only had it on for 1day currently running on orthos for few hrs @home

just wish i had a better memory divider on my mb

[RPGWiZaRD]

I still believe in G0 and that the US guys got a bad batch initially and that newer ones will do better. I've seen in a swedish forums now 3 users that found max stability on high volts on air at about 4.1~4.2GHz and 4.3~4.4GHz windows clock shots. Stepping L722A186.