Sign in with
Sign up | Sign in
Your question
Solved

How does my Q6600 Compare?

Last response: in CPUs
Share
December 13, 2009 1:19:43 AM

I was wondering where on the spectrum my chip sits?

With stock HSF and a cheap ass basic 2 fan hp mid-tower case,

(Using Core Temp 0.99.5 for the readings)

Revision: B3
Processor: Intel Core 2 Quad Q6600 (Kentsfeild)
Frequency: 1595.94MHz (265.99x6.0) (What the heck, I just noticed this-is my CPU underclocked?)
VID: 1.1625v
Tj Max: 85c
Core #0 low: 20c High: 29c
Core #1 low: 24c High: 35c
Core #2 low: 17c High: 27c
Core #3 low: 17c High: 24c

The reason I want to know is that I might be buying a brand new motherboard that will allow me to overclock my CPU...and just curious to know whether mine is any good! Responses from those who know what they are talking about would be greatly appreciated

More about : q6600 compare

a b à CPUs
December 13, 2009 1:27:33 AM

Speedstep is on. That is what is lowering the cpu to 6x266. Through a load at it and it should go up 9x266.

Those temps are false. Not unless your ambient temp is VERY cold those temps are wrong.


What are you doing with your machine that is making you think about overclocking the cpu?
m
0
l
December 13, 2009 1:39:01 AM

I'm not sure if human-cold is computer-cold but those readings are from ten minutes ago and the current temperature outside is 57F. No heater or anything in my room to alter that.

To be honest, the reason I want to overclock the CPU is mostly just-because and for overclocking experience.
I want to say I'm overclocking because of performance gains but TBH I don't want to sound ignorant and have no idea if I'll feel the difference (That's why I want to try - to find out. Plus I've read enough to understand the Q6600 is over-clocker friendly).
Also, I want to overclock to be able to sit at 1:1 synchro with higher frequency ram (either 3.0 for 667 /or/ 3.6 for 800, depending how this CPU will perform with better case/HSF)



P.S. after about two week's worth of reading this is my first time trying to "speak overclock" so let me know if anything I say sounds clearly wrong/illogical.
m
0
l
Related resources
a b à CPUs
December 13, 2009 2:23:51 AM

Sounds ok. Just those temps are not correct. Hardware and human temp as you stated are the same. 17c would be 62.6F In order for a core to actually be that cold the temp in the case would prob have to be in the mid 50s F. Which would mean your room temp would be lower....

Try some other temp reading apps.

m
0
l
December 13, 2009 2:33:10 AM

someguy7 said:

Try some other temp reading apps.



Speedfan readings:

Core #0--27c
Core #1--33c
Core #2--25c
Core #3--24c
GPU:56c
Temp#1--37c
Temp#2--21c
Temp#3--23c


Also why is core #1/2nd core always running considerably hotter?
m
0
l
a b à CPUs
December 13, 2009 2:41:43 AM

That is normal to have different temps.

Those temps readings look more realistic then the other ones. Keep in mind that they are still very cool for 65nm B3 quad core on stock cooling in a oem case. The cpu cooling in your system is very effective and you probably have one of the better binned B3 Q6600s.
m
0
l
December 13, 2009 2:48:03 AM

someguy7 said:
That is normal to have different temps.

Those temps readings look more realistic then the other ones. Keep in mind that they are still very cool for 65nm B3 quad core on stock cooling in a oem case. The cpu cooling in your system is very effective and you probably have one of the better binned B3 Q6600s.



Can anyone else confirm this before I start jumping from joy and run to get a p45 motherboard/antec 1200 case/ low latency 800 mhz ram? =]]
m
0
l
a b à CPUs
December 13, 2009 3:08:27 AM

What are we confirming? that temps will be different with different cores, or that those are "normal" temps.
I think those would be normal temps for a pc in a cold environment. I live in Colorado and work in my separate garage often( not heated). So those temps look about right for a house in the 60's
I would confirm by checking what temps the BIOS is reporting.

Yes, it is normal for each core to have different temps. Some cores are working harder than others, but even at 100% load on each core you are likely to have some variation between cores.
m
0
l
December 13, 2009 3:33:18 AM

buzznut said:
What are we confirming?



Is my Q6600 considered average, better, or worse? If I need to supply additional information to help you come to a decision please let me what bits are missing and how I can obtain them =].

buzznut said:
I would confirm by checking what temps the BIOS is reporting.


I'm not sure whether my BIOS displays that information, but I'm rebooting into setup to find out!


Edit: My BIOS doesn't supply any temperature info.
m
0
l
a c 131 à CPUs
December 13, 2009 4:10:26 AM

All I can really do is agree with everyone.
Recommend HWmonitor for temperature monitoring.
And ask what bios and/or motherboard you have to see if maybe you just missed where to find the temperatures or where you can overclock.
You can find your motherboard info using CPU-z in the mainboard tab. Is this an OEM (ie made by HP, Dell, Acer etc) system or did you build it yourself or have someone build it?
If you give me your system specs and your purpose in overclocking, I might also be able to see if it would be worth investing in a brand new motherboard or not just so you can overclock your cpu.

I can't really say if your Q6600 is "better normal or worse" because I don't have one to compare or know your cooling setup (other than the CPU fan& heatsink) or ambient temperature of the room. But for a CPU those temperatures are within normal no matter what the setup. When I overclock, I like to keep temperatures under load of the CPU no more than 55C. Prime95 is an excellent program for loading the CPU and testing for a stable overclock.

If you need anything else or more on something let me know.
m
0
l
December 13, 2009 4:40:19 AM

Here is the information regarding the chipset and ram

Chipset
-------------------------------------------------------------------------

Northbridge Intel P35/G33/G31 rev. A2
Southbridge Intel 82801IR (ICH9R) rev. 02
Graphic Interface PCI-Express
PCI-E Link Width x16
PCI-E Max Link Width x16
Memory Type DDR2
Memory Size 3072 MBytes
Channels Dual, (Symmetric)
Memory Frequency 332.5 MHz (4:5)
CAS# latency (CL) 5.0
RAS# to CAS# delay (tRCD) 5
RAS# Precharge (tRP) 5
Cycle Time (tRAS) 15
Row Refresh Cycle Time (tRFC) 36
Command Rate (CR) 2T
MCHBAR I/O Base address 0x0FED14000
MCHBAR I/O Size 4096
m
0
l
a c 131 à CPUs
December 13, 2009 4:42:20 AM

What about the bios info and/ or motherboard model?
m
0
l
December 13, 2009 4:47:23 AM

DMI BIOS
vendor American Megatrends Inc.
version 5.07
date 08/08/2007

DMI Baseboard
vendor ASUSTeK Computer INC.
model Berkeley
revision 1.xx
serial X312345678


Sorry had to break it into two...and computer is OEM/HP
m
0
l
December 13, 2009 7:04:40 AM

First of all it's load temps that matter not idle. Download Prime95 and run stress test for couple of minutes then report temps back here. For temp monitoring I recommend CoreTemp.

It is really hard to compare Your CPU because that would require exact same PC(m/b, case/ heatsink) and ambient temps for fair comparison.

Even if the Temps are good/low it does not guarantee good overclock.

Most of the Q6600 are good overclockers and reach at least 3.2GHz but You never now about Yours before You try.

New motherboard will probably require new HSF anyway because OEM's very often use their own design which does not fit another m/b or does not work very well on different m/b or case. This means You will need new HSF. Therefor there is not much point comparing Your temps now. If You get good new hsf there should be no problems to keep Your CPU cool enough for 3.2-3.6GHz overclock no matter how good/bad it is. That is assuming that it will hit 3.2GHz as like I said before there is no guarantee but most CPU's should get 3.2GHz easily.

3.2 GHz is sweetspot for Q6600, You can lower multiplier to 8 and run it 400MHz x 8 with RAM 1:1 at 800MHz speed. Going higher than that will get power usage up quite a bit and run hotter whith little improvement to performance.
m
0
l
December 13, 2009 7:11:49 AM

Just had a second look through Your posts, it says TjMax 85C, it Should be 100C for Q6600. This is a problem with a lot of monitoring software specially older versions as there was little information from Intel about temp sensors.

So Your CPU temps are actually 15C higher then those reported.
m
0
l
December 13, 2009 8:11:22 AM

I went ahead and did what ainarssems said to do, downloaded and ran prime95, and experienced mini-heart attacks along the way as the temperature began to increase. Looking at the temperature charts climb, I started apologizing in my head to everyone who I have ever sinned towards.
The results:


(Note: I don't know the correct procedure to getting maximum temperature, so I stared at the speedfan readings every second of the 15 minutes. the highest recorded temperatures are

Core #0 -- 59c
Core #1 -- 61c
Core #2 -- 58c
Core #3 -- 57c

temperature #1 -- 41c
temperature #2 -- 47c
temperature #3 -- 51c

Also I just ran Prime95 for 15 minutes BUT used >Core Temp 0.99.5< with the following results, assuming that adding an offset of 15 degrees makes up for the default Tjmax being 85:

Core #0 -- 69c
Core #1 -- 71c
Core #2 -- 67c
Core #3 -- 67c

ainarssems said:

So Your CPU temps are actually 15C higher then those reported.


Ok. For some reason I cant adjust the actual Tjmax value.Since you said the Tjmax should be 100 I went ahead and opened coretemp>options>adjust offset and added 15 degrees to all cores. All temperatures do appear 15 degrees higher, however, so is this sufficient to compensate?
m
0
l
December 13, 2009 8:33:18 AM

Those temps are good for OEM PC assuming the readings are correct. Do not get heart attacks, CPUs are designet to prevent killing themselves from heat. Q6600 will reduce it's speed and voltage when it hits 85-90C and shut off PC at 100C. Anything below 70C at 100% load is OK if You do not want to use CPU for more then 10 Years.

Can You check temps with CoreTemp, try version0.99.4 that should report temps correctly for Q6600. TjMax is 100C for Q6600 check that is correct.

Temp sensors on CPU are reporting distance to Tj Max. So if the value is 20, then software reports 80C ( 100-20) but if software has wrong TjMax for Your CPU it will report wrong temp. If it says TjMax 85C it will show lower temp when value from the CPU temp is the same (85-20) 65C when in fact it is 80C.
m
0
l
December 13, 2009 8:35:17 AM

I really wouldn't worry about the temps too much. As long as they don't exceed 70 under 100% load (that's my preference anyway), if they do, decrease the overclock and voltage a touch and run another prime 95 test and adjust accordingly.

I have had experience with 4 Q6600's (G0), all of which could hit 4.0Ghz, and 3.6Ghz for 24/7 use. A decent cooler is needed mind. Since yours is a B3 stepping, 3.2Ghz would be easily attainable, and higher if you're willing to upgrade the heatsink and push it a bit.
m
0
l
December 13, 2009 8:35:40 AM

Quote:
Ok. For some reason I cant adjust the actual Tjmax value.Since you said the Tjmax should be 100 I went ahead and opened coretemp>options>adjust offset and added 15 degrees to all cores. All temperatures do appear 15 degrees higher, however, so is this sufficient to compensate?



Yes that is sufficient.
m
0
l
a b à CPUs
December 13, 2009 8:53:19 AM

B3 Q6600s likely have a Tjunction Target of 90C, not 100C. They are essentially two B2 E6600 in the same package, and the B2 E6600 is 85-90C. Tjunction Max is not the same for each CPU or even each core anyway, so it's not worthwhile to worry about 5-10C here and there. If the CPU isn't throttling and isn't unstable, it's fine. Everything else is just guesswork.
m
0
l
December 13, 2009 9:04:16 AM

After reading a little more into it I checked and noticed that my VID says something about overclocking too? CPU-Z says 1.163v so is that a good idle VID? (if idle VID is what you call it) (Also it switches to 1.300v when something makes the CPU kick up from idle to normal)


P.S. I do apologize for pouring my information in segments but I'm learning as I go, and I hope that you very helpful members of Tom's Hardware can accommodate to idiots such as myself =]

m
0
l
December 13, 2009 9:21:43 AM

I think idle VID is1.1625V for all Q6600 and full speed is 1.2500-1.3750V with most being 1.3250V so 1.3000V for full speed considering You have B3 is pretty good.

And I would not call Yourself an idiot. You know some stuff and want to learn some more. Idiots do not know a s**t, don't want to learn and ignore any advice.
m
0
l
December 13, 2009 1:00:40 PM

I agree. If CoreTemp is showing your VID at 1.1625, you should be able to overclock like crazy with a good heat sink, decent case with good airflow and a decent mobo. I have a Q6600 with a VID of 1.3250 in a Lancool K-62 case. I have to set my core voltage at 1.45 to achieve 3.2Ghz but you should be able to do so with much less voltage. I have a Prolimatech Megahalems heat sink and my temps are 33C idle and 60C under full load during Prime 95 at an ambient (room) temperature of 21C. VID isn't the only factor but a low VID is a great start for you.
m
0
l
a b à CPUs
December 13, 2009 10:03:25 PM

VID does not indicate overclocking capability.
m
0
l
December 13, 2009 11:24:03 PM

Do you say that because he could purchase water cooling if his voltage and temps get too high or because VID does not indicate higher voltage requirements for overclocking? I am asking this for my own knowledge, not to be argumentative. I have had less success overclocking higher VID cpu's at voltages and temperatures that were within the limits I prefer. Maybe I just have terrible luck.
m
0
l
a b à CPUs
December 13, 2009 11:32:21 PM

I'm talking about the VID. VID alone does not indicate a good or bad overclocker. There are people who can get high VID chips very highly clocked. There has been some correlation drawn between the two but it's pretty weak and yet still taken as gospel.
m
0
l
December 14, 2009 4:43:47 AM

randomizer said:
VID does not indicate overclocking capability.



Do you mean to say that it is not the sole determinant, or do you mean to say that it is completely irrelevant? Please clarify.
m
0
l
a b à CPUs
December 14, 2009 4:49:27 AM

It's not the sole determinant.
m
0
l
December 14, 2009 6:57:09 AM

Based off of the information provided in this thread, can anyone wrap it all up and answer my initial question? Not to be rude, but please only give a response to this question if you know what you are talking about, the responses I get here will weight heavily on whether i build a new system around this cpu or just sell and move on. Thank you!
m
0
l

Best solution

a c 172 à CPUs
December 15, 2009 1:42:30 PM

You have a pretty nice Q6600 with a pretty low VID. Like randomizer says, a low VID is not the only determinant of overclockability, but it is a good indication.

You can build a pretty capable, inexpensive system around this chip. I have a G0 Q6600 with an average VID - 1.2625 volts. It's running at 3.6 GHz. in a Gigabyte GA-EP45-UD3L motherboard (about $90 from newegg). Of course, I have a better than average cooler (Xig Dark Knight) in a better than average case (Antec 900).
Share
a c 172 à CPUs
December 17, 2009 2:18:05 PM

Quote:
... because my VID on my GO is 1.325. Very high.

That's what I call losing the the CPU lottery. You can tell what stepping you have by the S-Spec code on the box, but there's just no way to determine VID.
m
0
l
a b à CPUs
December 17, 2009 11:14:51 PM

Quote:
Not the sole, but a huge indicator of overclocking ability. The reason I cant get mine over 3.2ghz is because my VID on my GO is 1.325. Very high.

No, you can't overclock well because your chip is simply not the same grade as other chips, and requires more voltage to run at a given clock speed. If it had a lower VID it wouldn't automagically be a better overclocker. Intel could have the same high VID for every chip and the same overclocking differences would exist.
m
0
l
!