Printable View
I'm on latest I think (1101?) but i'm about to go look for anything else or beta, if not I may even go back to an earlier bios, this is unbarable....
Edit: nah I'm on latest :(
Looking foward to another.... or something to fix my terrible latency problem.
I wondered if anyone else noticed this. On my evga 680i I ran an e6850 at 450x9 with cheap ram set linked and synced (900mhz at 5-5-5-15) with no tweaking involved, and my everest read scores were over 11k with latency of about 48ns. On my evga 780i, with an e8400 at similar speeds, I got read speeds in the mid 9k's with latencies in the very high 60's at best. With the q6600 in the evga 780i (at 465x8), I got read scores in the hi 8k's to low 9k's and latencies in the low to mid 70's. I even used good ram and tried unlinking and running at 1100+mhz, and in some settings, the results were even worse. I think I was using the P03 bios though in the 780i. I was so disgusted with the scores that I tabled the board and went back to the p35 mobo.Quote:
Originally Posted by The0men
NBF, are you running the 266Mhz strap or the 333Mhz strap? When I ran my Q6600 at 450x8 my Super Pi times were slower by more than a minute at 32M. I'm going to try and see what happens with performance using the 333Mhz strap. I'm runnning 400x9 with 1.425 VTT and 1.55v on the NB. I lowered my Vcore to 1.4375 which is 1.42v with a vdroop as low as 1.38v in CPU-Z.Quote:
Originally Posted by NBF
I've witnessed this first hand. At the same speed all three of my quads ran at different temps(after multiple remounts). The hottest being the vid of 1.2125 and the coldest being vid of 1.325. It wasn't a small change either, we are talkin about a 10'c change in temps.Quote:
Originally Posted by Leeghoofd
Got a SLACR L739A976 with 1.25 VID
So far OCCT stable at 3.6GHz with 1.443v in BIOS (loaded 1.344v)
Motherboard is a Gigabyte DS3R, with F8 BIOS. Watercooled with Fuzion, PA 120.3 and an 18w DDC.
Think it can do more but the motherboard droops a whole 0.1v or more if you go higher, and I dont feel like setting 1.6v just to get 1.48 / 1.5v real. Maybe I shouldnt have gotten this budget board, with the enormous droop. For the rest however it does this clock at stock fsb and mch voltages.
how well do these 6600's do at 3.6 with good air like the true?? i see a lot of people doing watercooling is that something that should be done at 3.6?? or is air okay??
All depends on the CPU you get mate, if it's hothead and pumps out loads of heat then it's best to be with a watercooling setup to keep temps down... if you get a good one things can be nicely cooled with a thermalright Ultra 120 or sortlike...
should i look for a general vid range for good air oc or mainly just as low as i can find??
im thinking about getting both e8x00 and a q6600 and see what i like better
@hank, what motherboard are you on?
My L737B505 on an IP35 Pro and only 1.42v -
http://img183.imageshack.us/img183/2...prime95cu8.jpg
Abit P35 pro baby! I need to make a dry ice tube or something and I think MUCH higher clocks are not out of the question.Quote:
Originally Posted by Bail_w
also need to do a Vmod to the mobo and I think even on water higher clocks should come!
Would people recommend the Abit IP35 Pro or MSI P35 Neo2? Both seem very good and they were almost tied in Quad FSB performance in the Anandtech review...
I'm leaning towards the MSI as its cheaper and has a prettier NB / VRM cooling setup :p
Do the exact batch for several chips correspond to the same VID? I mean for example, do two L750B500 (imaginary batch) will have 1.2500V VID or can one have 1.2500 and another 1.3000 for instance?
Do you have some batches of 1.2V VID to give me to look for in the shops tomorrow?
The VID can vary within identical batches, I was looking into good batches recently and the VID varied a lot in the same batch :(
Then again, there can be a good overclocking batch with a relatively high VID etc
Its all guesstimation and gambling tbh
Hey guys, I got my G0 to 3.6ghz at 1.48V and there it’s rock stable, one day primed.
For some reason, I cannot get this damned thing to run stable at 3.8ghz. I’m running it under water, so my temps never exceed 60 at 3.6 (highest core gets 55C as reported by Everest with a 25C ambient). When I push it to 3.8, 1.52V and prime it, I don’t exceed 60C, however it doesn’t run for more than a few minutes. I’ve tried to address this issue before, but now it’s really ticking me off.
I’m thinking it is something to do with my motherboard settings. Any tweaks that you guys do to reach those high overclocks, or do you just increase the FSB and Vcore and you’re done?
Just for your info, I have it on a P5k-E wifi motherboard.
Things I’ve changed:
PCIE at 100mhz
PCI at 33mhz
Vcore at 1.53 read in windows
Load Line calibration Enabled
CPU GTL Voltage Refference at 0.63
NB GTL Voltage Refference 0.61
CPU and PCIE Spread Spectrum disabled.
Running it at a 423FSB with a 9x multiplier.
Edit: I just flashed my mobo bios to the newest version, and it seems to have been priming well - for 21 minutes when it finally failed.... At least it didn't BSOD this time. Any imput very much appreciated.
Quote:
Originally Posted by adpr_02
Well, if it takes 1.48v ( I assume you mean BIOS or do you mean real?) just to do 3.6ghz then the chip will most likely alot more than 1.52v to do 3.8 stable.
Voltages don't scale linearly, so it's likely that thats as far as your chip goes, maybe a little more, but with needing alot more voltage.
My own Q6600 needs just 1.35v real (1.42v BIOS) to do 3.62ghz stable. However after that I need a substancial increase for every few mhz after about 3.65ghz to make it prime stable.
Not even all g0's are created equally. Srry to burst anyones bubble.
I would go with the MSI P35 NEO-2 FR if the price is the same as in the states. Zipzoomfly.com has the NEO-2 FR for $97 vs. $140 for the ABIT after rebate from newegg.com The Abit is $170 before rebate. The NEO-2 FR is identical to the Platinum P35 minus exotic heatpipes.
Quote:
Originally Posted by conzymaher
Agreed, if you look up reviews for the NEO-2 FR you will see that it is an amazing board and for $97, it is a steal.Quote:
Originally Posted by Hans Gruber
I agree with TheOmen. I'm currently in the same situation as you are- trying to get my q6600 to 3.8Ghz prime stable. I am watercooling too, and temps under load are not an issue yet (I'm using the DFI P35 board though). I am up to 1.55v vcore (measured by everest) and I'll let you know if this works for me- I left it priming while I'm at work. Any vcore lower than this will result in a crash and reboot some time after an hour or so of priming. I think if I can't succeed on 1.55v (real), then perhaps I will abandon my desire to run mine at 3.8ghz 24/7.Quote:
Originally Posted by adpr_02
By the way, what batch and VID is your q6600? (mine is batch L743A and VID 1.300)
My Q6600 with Thermalright ultra 120 Extreme, to 3.825Mhz at 1.448 with temps 34-33-28-28 in Antec P182!!
http://www.pctunerup.com/up/results/...timobien10.JPG
Asus 8800GTS 512 at 36º with stock cooler!!
Nice!!! Do you know the batch# or VID of your star performer?Quote:
Originally Posted by manyga
Just curious- anyone have good luck so far with the new L8xx batches?
Quote:
Originally Posted by drnick79
Thanks bro!!
The batch is: L730B616 and vid is 1.2625v
Regardss from Spain!
Well, with my new BIOS, I tried priming it last night at 1.54v at 3.8ghz and it SEEMED to work. I had it running for 1 hour and 20 minutes, and then decided to call it a day. All the voltages I say are as specified by windows.
I REALLY don't think that it's a good thing running 1.54V through this thing 24/7, especially since when it's not in idle, that jumps up to 1.56V. Anybody used this proc. for a while at that high vcore, is it adviseable or not considering my temps never reach 60.
Either way, If 3.8 works, I'll just go for the big 4.0 and try to get it somewhat stable there just for benchmarks....
Just an interresting thing - with this new bios, I cannot seem to be able to run 3.6ghz stable anymore...... even though 3.8 is now stable.... I'm thinking it may be the multiplier or something as when I go for 3.6 I tone it down to 8x.
I've tried to get a generally accepted ceiling vcore value for 24/7 use (if there is such a thing) on the q6600. I thought I've seen 1.55v listed once or twice, but I'm not sure what most people think.Quote:
Originally Posted by adpr_02