Likleyhood of hitting 3.5ghz on air with Q6600?

[blazed]

I don't think you need a high end board when the chips are hitting walls at 425-450 FSB area, a nice P35 board can do that easily. DFI Blood Iron if your serious bout your OC, P5K Pro has 8 phase for the CPU and only costs $140. Also remember the Q6600 is a hot chip, at 3.6ghz its barely coolable on high end air, the Q9450 will do much better.

 

[OBLIVIONLORD]

That wall is due to the chipset being incapable of exceeding the mark for quads compared to duals.

 

[blazed]

Are you sure? How come the same motherboard and chipsets can handle the higher Q6600 overclock then? After all the Q9450 has less power and heat moving through the motherboard, it should be able to do at least equal. This would cause me to believe the FSB walls lie within the processors themselves, that and you can see the wide range of different FSB walls, using the same motherboards. If it were really the motherboard then they should all be roughly the same.

 

[OBLIVIONLORD]

The chipset controlls the FSB since there is no memory controller built on the processors. Therefore the chipset will have a wall for whatever it can do which is why some mobos are able to goto higher FSB than others.

Q6600 has a 9x multiplier with a buss speed of 266 which means that 400mhz ddr2 will get you 3.6ghz. It's very little stress to the chipset since alot of them these days are nativly able to do 1600mhz which is 400x4.

The Q9450 has an 8x multi but, stock buss of 333mhz which doesn't give you a whole lot of room to work with since 400mhz is just 3.2ghz vs the Q6600's 3.6ghz. The majority of mobo's out can nativly just do 400mhz fsb and some can be pushed higher.

Also keep in mind that the bandwith difference from a dual to a quad is different which is why some mobos running duals can hit 500fsb whereas with a quad it seems like 425fsb is the avg rate that I'm seeing unless you buy a really expencive board which will most likley reach high FSB while using a quad.

 

[freeagent]

i think thats why the higher boards cost a lil more.. the extra features they give you is meh, but i think they make you pay for overclocking ability. especially for quads.. they are voltage and current pigs once you get em up to speed. just fo instance, in this thread, on this page and not to brand bash.. there was one guy running a q6600 on an abit. hes running his q6600 at 3600mhz, with 1.5v set in bios, and his board droops to 1.39 on load, thats alot! for me whether im running 400x9, or 450x8 with 1.48 set in bios, it droops to 1.46, that to me seems worth the extra money.. now for fsb walls.. i cant say im a pro, ive only had 2 conroes, and 1 kentsfield. my 6300 made it to 535fsb on this board, my e6600 walled at 444 fsb, and this is stable at 475fsb, it might do more, but im too chicken to find out. long story short.. how fast do you want to go, and how much do you want to spend?

because like everything else, you always get what you pay for..

 

[Sarsbaby]

well this post got me curious, I was doing fine at 3.2Ghz, but I pushed it to 3.4Ghz, load temps are about 66C, a little high for my liking, but I'll let it hang there until I buy a TRUE120. definitely get good cooling for 3.6Ghz.

terran 2k, If your getting up to 66c on 3.2, you NEED better cooling for sure!!!

With a Zalman 9700 and my CPU at 3.6 at 1.55v, I get 60c at the most.

OBLIVIONLORD, You should be able to get to 3.6, cooling makes a big difference, also RAM and mobo, and it looks like your set in those areas so you should be good to go.

sorry, shoulda read more, I don't really know much about quads.

 

[KillrBuckeye]

terran 2k, If your getting up to 66c on 3.2, you NEED better cooling for sure!!!

With a Zalman 9700 and my CPU at 3.6 at 1.55v, I get 60c at the most.

There is way too much confusion about C2D/C2Q temps to make comparisons like this unless everyone is very specific about how load temps are measured (what stress test, what monitoring program, what readings?) and the conditions under which the temps are observed (actual load Vcore--not idle or what you set in BIOS). Sarsbaby, your load temps seem way too good to be true if you are measuring everything in the same way as me. I have 1.265VCore under load, and I'm running 389x9 = 3.5 GHz with a Xigmatek HDT-S1283. According to CoreTemp, 2 of my cores hit 70C after 10 minutes of 4-threaded small FFTs in Prime95. If you measure in the same way, I would be shocked if you come back and say you're only hitting 60C. I'd expect you to be hitting 70C or higher easily.

 

[freeagent]

Sarsbaby, your load temps seem way too good to be true if you are measuring everything in the same way as me. I have 1.265VCore under load, and I'm running 389x9 = 3.5 GHz with a Xigmatek HDT-S1283. According to CoreTemp, 2 of my cores hit 70C after 10 minutes of 4-threaded small FFTs in Prime95. If you measure in the same way, I would be shocked if you come back and say you're only hitting 60C. I'd expect you to be hitting 70C or higher easily.

sounds like you need some louder fans