Initial Overclock Results: E6600 @ 3.6 Ghz

[theYipster]

Hi everyone,

First, thanks for the help and advice given with building my new PC. Things have been excellent so far.

After building last week, I'm now over-clocking the system. My setup is:

Intel Core 2 Duo E6600 w/ Tuniq Tower
evga nForce 680i A1 motherboard
4 Gigs of PC-6400 RAM / g.skill
2x GeForce 8800 GTX graphic cards in sli
2 raid setups in RAID-0: 2x 500 Gig, 2x 250 gig
Case is a Silverstone TJ-09 with two TriCool fans mounted at the top.
OS is Windows Vista x64 Ultimate

At stock speeds, the processor cores idle in the mid 20s and load in the mid 30s. I first overclocked to 3Ghz following the nVidia guide. When all was stable, I followed the screen shot posting to get to 3.6 Ghz.

Note: The last time I over-clocked a system, it was by setting a jumper to increase a Pentium Pro from 200Mhz to 233Mhz. Today I'm overclocking with a good amount of hand holding, seeing how much the process has changed in 13 years.

Anyway, I haven't run any 8-hour tests yet, but I did run Orthos for twenty minutes, and a Super Pi 8M calculation. At idle the core temps are 29/30, with cpu surface around 33-34. At load using Orthos, the core temps max out at 55C, with cpu surface at 60. Load using Super PI was much lower, with the surface temp staying below 53C. Cores were in the mid 40s. Measurements were taken with Speed-Fan, which I find to be pretty reliable. Ambient air temp is about 19-20C.

So the big question is: how does this rate? Are the temps in line with good results considering the air cooling setup (Tuniq Tower?) Note that I set vCore in the BIOS to 1.50625 according to the AnandTech screenshot post. Speedfan reports vCore at 1.44. I assume the next thing to do is lower until I can't boot windows, then run tests at the lowest bootable setting. Does that make sense or is 1.44 a reasonable setting?

Thanks,

Mark.

 

[swtethan]

speedfan is off by 10c for me, now intel thermal analysis tool is way off also +13c, asus probe seems to be the one in the middle with a idle of about 35c

 

[Shimmishim]

1.44 with droop at 3.6 is a very good and reasonable amount of voltage for 3.6

your temps look good for idle/load.

i'd say just go ahead with the stability testing!

just leave orthos on overnight and see if it's still running in the morning.

 

[iviustang50h]

I just built a similar system but am using the Asus Commando. Either my cooling isn't as good inside the case as yours or my chip is a dud because I need to run a vcore of 1.5 (1.45 according to speedfan, probably due to vdroop) to maintain stability at 3.2 (lame). I am also using the Tuniq Tower.

So let me ask you something, which way did you mount your Tuniq tower 120 (which way is the fan blowing the air)? How tight did you tighten down the screws? I used artic silver 5 as well, so I am at a loss.

My ram passes 2 hours of memtest at rated speeds (G.Skill DDR1000 at 4-4-4-5) so I don't know if I got a dud E6600 or perhaps the Commando is messsed up...or maybe the Tuniq is mounted wrong? My CPU temps idle in the high 20's without the side of my case being off. With it off it idles at 24. I wonder why I need so much voltage, but you can be orthos stable @ 3.6 using only 1.5 Vcore!? Oh, my Tuniq is mounted so the fan blows toward to back of the case, acting as an exhaust.

 

[theYipster]

Hmm... My setting was 1.50625 wit a vdroop at 1.4, not 1.4 with a droop at 3.6. I tried that, and while Windows booted, ORTHOS failed after a minute.

Mark.

 

[theYipster]

Iviustang: The tuniq tower is mounted so that if you look at the case standing up, with the motherboard seated normally, the ends of the waves point down and away, as in ^. Let me know if that's not clear.

The screws are mounted so that the actual screw portion is completely seated, with the solid portion sticking out and through the heat sink bar. This provides a very secure fit, but the screws are not overly tight. I use AS5 too btw.

It looks like your idle temps are good at 3.2, especially with the voltage at 1.5. My temps at 3.6 idle in the low 30's, but I suppose that's to be expected. At stock, my idle temps were in the mid to low 20s. Given that, I think your tower is working fine. You might want to check the commando...

Mark.

 

[iviustang50h]

Originally posted by: theYipster
Iviustang: The tuniq tower is mounted so that if you look at the case standing up, with the motherboard seated normally, the ends of the waves point down and away, as in ^. Let me know if that's not clear.

The screws are mounted so that the actual screw portion is completely seated, with the solid portion sticking out and through the heat sink bar. This provides a very secure fit, but the screws are not overly tight. I use AS5 too btw.

It looks like your idle temps are good at 3.2, especially with the voltage at 1.5. My temps at 3.6 idle in the low 30's, but I suppose that's to be expected. At stock, my idle temps were in the mid to low 20s. Given that, I think your tower is working fine. You might want to check the commando...

Mark.

I have my Tuniq mounted the exact same way, so the air from the fan should be pushed out the back then. I think the mounting is good and my screws aren't overly tight, but not quite all the way down till I can't turn them anymore - I don't want to bend the board.

I dunno if my Commando is the issue, or maybe the E6600. I may RAM the commando anyway because one of the white secure clips at the edges of the ram slot was missing (on the first blue slot). I think that is enough to warrant an RMA.

Anyway, I am fully stable right now at 3.2GHz with a vcore of 1.5 (1.45 with vdroop), 1.55 MCH voltage and 1.4 FSB voltage. I want to hit 3.6, but I will take it slow and will probably try with a new Commando. Thanks for the quick response!

 

[Shimmishim]

@iviustang50h - do you have the gskill 4-4-4-5 kit? my guess is that you do.

be warned that the sub timings are pretty tight with that mem and you may need to loosen them up a bit to gain stability.

 

[iviustang50h]

Originally posted by: Shimmishim
@iviustang50h - do you have the gskill 4-4-4-5 kit? my guess is that you do.

be warned that the sub timings are pretty tight with that mem and you may need to loosen them up a bit to gain stability.

I do have the ram, yes. I know it runs stable at ddr1000, but you never know I guess. I may loosen to 4-4-4-12 or maybe even 5-5-5-12 if necessary. Thanks for the tip!

 

[Shimmishim]

4-4-4-X refers to the main timings. The subtimings are hidden on most 965/975 boards. so when you get into windows, you need to adjust them with a program like memset 3.2 and then try to see if you can go farther or if you do have access to the subtimings, trC, trfc, twtr, trrd, etc. then change those to bigger numbers and try again.

 

[theYipster]

Hey everyone,

I'm very pleased to report that Orthos ran without fail for 9.5 hours over the night, with core temps remaining within the mid 50s throughout the test.

I tried reducing the voltage to see if I could get a stable 3.6Ghz on anything less than a 1.5 bios vCore setting. While Windows would boot up at 1.38, Orthos would fail after a minute or two on anything below 1.48, and then it would fail after 10-15 minutes on anything below 1.5. However, 1.5 works flawlessly with a vdroop of 1.44, and my system shows no indication that it'll be a problem. Besides, I plan to replace the processor with a Peryn in a year.

Perhaps when I get some time I'll see if I can push things further, as I don't think I've quite found the wall. How feasible is an E6600 at 3.8Ghz (heh, but now I'm probably kidding myself.)

Anyway, thanks for all the help.

Mark.

 

[Shimmishim]

3.8 is possible but you'll have to watch your temps. Try to keep the voltage below 1.55 on air.

 

[iviustang50h]

Originally posted by: theYipster
Hey everyone,

I'm very pleased to report that Orthos ran without fail for 9.5 hours over the night, with core temps remaining within the mid 50s throughout the test.

I tried reducing the voltage to see if I could get a stable 3.6Ghz on anything less than a 1.5 bios vCore setting. While Windows would boot up at 1.38, Orthos would fail after a minute or two on anything below 1.48, and then it would fail after 10-15 minutes on anything below 1.5. However, 1.5 works flawlessly with a vdroop of 1.44, and my system shows no indication that it'll be a problem. Besides, I plan to replace the processor with a Peryn in a year.

Perhaps when I get some time I'll see if I can push things further, as I don't think I've quite found the wall. How feasible is an E6600 at 3.8Ghz (heh, but now I'm probably kidding myself.)

Anyway, thanks for all the help.

Mark.

Wow, I am jealous! I cannot even maintain stability without using a 1.6 vcore at 3.4GHz. I am grossly disappointed in the E6600 and the Asus Commando MB, especially after reading how others have EASILY hit higher speeds at voltages with the Tuniq Tower. Just my luck... I may be returning this lame MB!

 

[gramboh]

What steppings are your E6600's, both the 3.6 and 3.2 in this thread?

 

[theYipster]

6-B2

Mark (@ 3.6Ghz.)

 

[equetefue]

where you buy from and when ?

 

[theYipster]

I bought on Wednesday, March 7th from Newegg.com

Mark.

 

[dannybin1742]

i got my E4300 on march 4th and its sitting at 3ghz (so far) SL9TB 1.35V

 

[idiotekniQues]

Originally posted by: swtethan
speedfan is off by 10c for me, now intel thermal analysis tool is way off also +13c, asus probe seems to be the one in the middle with a idle of about 35c

TAT should be right workin on a 965 chipset for an e6400. asusprobe reads the old school sensor which is in the middle of the cores.

TAT actually reads each core more accurately. you should pay attention to TAT more than asusprobe.