Printable View
Looks like I'm having better luck with x9 multiplier. Currently set FSB411 x 9 to get 3699mhz or 3.7ghz ;)
1.500v BIOS set = 1.464v idle and 1.4480v full load. I can try lower the Vcore one or two interval lower at current clock speed.
My goal is to find the sweetspot that's under Intel thermal spec of 71c for Q6600 G0. :)
http://img338.imageshack.us/img338/9573/3700vr5.jpg
That's great to see. 3.7 GHz at 1.45v and good temps is awesome.
looks damn good, what speed do you expect to stay on? sure looks to go up more and more all the time :D
XtremeTiramisu, with AI Suite showing a cool System/Mobo temp of 24C, what was your ambient during these tests?
Yup and I've just surpassed the previous 3.7ghz threshhold...have a look below.Quote:
Originally Posted by Xilikon
Seriously, I really don't know..:DQuote:
Originally Posted by lv_88
But one thing for sure is that I'll never go back to dual cores. At this kind of clock speed : volt : temp on air and times 4 cores, I don't see what's the reason for me to stay with my previous golden E6700.:p:
Good question, with a little help of lower temp in the early morning - 20c as of nowQuote:
Originally Posted by Icer
Temps are still great in this current OC stage.
SWEET LORD!
I've just obtained a new high record!
Forgot to load Asus AI Suite in the last screenie...
Details reserved, a new screenie will be uploaded very soon !:D
nice going there man
:clap: :clap: :clap: :clap: :clap:
Agreed, better than then E6700 at same voltage, with 2 more cores is icing on the awesome cake :clap: :clap: :clap:Quote:
Originally Posted by dinos22
Quote:
Originally Posted by dinos22
I concur x3 ! :DQuote:
Originally Posted by Xilikon
Same voltage, same clock speed but 2x more cores. What's the math? 3737mhz x 4 = 14948mhz (Did I mentioned 3737?...it's not a typo..;) )
I really fail to see any reasons of going back to dual cores again...
THANK YOU GOD !!!
This kind of thrills is far better than winning the big jackpot....or maybe not...LOL:rofl: :ROTF:
is that stable for windows...like, you can surf the internet, send receive emails and just peruse your hard drive?
what was your memory ratio set to? 1:1?
Oh it can do a lot more than that...prime all 4 cores stable.Quote:
Originally Posted by Ace-a-Rue
The Ballistix 8500 tracers kit is actually capable of going up to ~DDR2-1200 @ 2.2v on the same P5K-Dlx mobo I've now. The divider has always being set to 4:5 divider and becouse I'm using x9 multilier and FSB415 right now so it appears in lower than its original stock speed. I don't need to worry about the rams just yet as I know what it's capable of at given low 2.2v ;)
Have a look at the following post:)
Ladies and Gentleman: 3737mhz per each core x 4 = 14948mhz - The future is here now!
So far, it's prime stable 4+ loop and 1 hour 20min+ and counting.
I can totally forget about dual cores now
http://img338.imageshack.us/img338/5...stable1rz1.jpg
NICE!..this is a"G0" stepping; right?
V Nice :clap:
Now we just need a few more people to post results...I'll be happy if I can run mine 3.6GHz 24/7. I'll try to post something later today or tomorrow....Come on FedEx truck!
so that's 1.5 vcore in bios for the 3.737 Ghz right?
YES! believe what your eyes can see. CPUZ doesn't lie:p: G0 it is indeed!Quote:
Originally Posted by Ace-a-Rue
I'd love to see what other G0 Q6600 owner's OC results so we can come up with a good avg OC as the norm for new G0 chips :cool:Quote:
Originally Posted by Icer
That's correct!Quote:
Originally Posted by jeffyjaixx
And I'm about to increase the clock speed once again, I feel that I can strech the clock further at the current given Vcore.
P5K-Dlx is once again without a doubt the best mobo for quad.
Very nice - Congrats on the new chip!
BTW, I know you've read up on the Tjunction issue, but it's possible (probable) that was also increased by ~10-11C with the new stepping, which would make it 111C now, instead of 100C. I think at this point we just don't know yet what the value truely is supposed to be.
Could you do your next screen-shot with CoreTemp set to display the distance to Tjunction? - that's the only important/meaningfull number - how far you have to go before the cpu throttles itself.
http://skinprops.com/images/ct095.jpg
...of course the only reak meaningful information is whether or not the system is stable at some setting, but if you use that option above, it will help you determine 'why' it might not be stable at some setting :).
Very nice. :clap:Quote:
Originally Posted by XtremeTiramisu
Have you tried 400x9 at the same 1.425 vcore in bios to see if it's PRIME stable?
Just want to know the fluctuation in temps. :)
Quote:
Originally Posted by Spanki
His Core Temp 0.95 pic shows 100C Tjunction.
That's a hard-coded value (based on the determination that it's a quad), not something CoreTemp can 'read' from the cpu. My guess is that it's wrong (again).Quote:
Originally Posted by Icer
HOLY Sweet Jesus ! ! !
At the current clock speed, it's already the same clock per core as my golden E6700 can do at the same voltage but 2x the core ! ! !
I wonder what's the max clock speed at the current given voltage, can anyone care to take a guess ?! :D :D :D
If 3780mhz is going to be prime stable for hours at the current given Vcore, I'll mark today as my second BD to celebrate...LOL!
A one full round just passed...:)
http://img338.imageshack.us/img338/5...testoc1pd3.jpg
Quote:
Originally Posted by Spanki
Quote:
Originally Posted by Spanki
As per requested:
Tjunction Remaining.
My brain is too tired to work out the calculations right now, care to do that for me? :D
Edit: Nevermind, I'm too eager to find out myself so I just did the calculation and it's indeed Tjunction at 100c.
100 - 65 Tjunction temp remaiing = 35c Idle Coretemp.
Take 35c Idele coretemp's value and - 15c = Speedfan's idle temp readings (It reads 85c Tjunction instead of the actual 100c Tjunction from the Quad6600 G0)
http://img255.imageshack.us/img255/5...aining1wp4.jpg
holy moly :yepp:
:shocked:
ok look out someone get you beat they have it at 9x445 on air
http://www.coolaler.com/~coolaler/fo...t=157070&pp=10
http://www.coolaler.com.tw/coolalercbb/Q6600G0/1.gif
Thats an enginieering sample, it can't be fairly compaired to a retail CPU.Quote:
Originally Posted by ruff97
ruff97, this is worthless as it's a ES sample. We are talking retail here ;)
Actually, that just confirms that CoreTemp is using 100C for Tjunction :). The issue is that it might supposed to use 110C or 111C (or some other value between 100C and 127C - Intel hasn't supplied this info yet). So unfortunately, it doesn't tell us what the actual Tjunction value should be for this cpu/stepping.Quote:
Originally Posted by XtremeTiramisu
So, what I'm suggesting is that you use that option in CoreTemp, so you can monitor the DTS value, as it counts down towards zero when you're overclocking, under Load. If it's getting down in the 5-7C remaining range, you'll know that you're about to get throttled.
BAH, another ES.
Always nice to see those cherry picked ES only serves as a kick to see but not well enough to be considered as a reference for the general public.
1. He's on open bench.
2. I don't see anywhere he has posted prime stable or related...
3. He's on XP OS
4. 1.57v idle @ 4ghz does not impress me as being an ES sample even. This is 1.6v+++ Vcore set in BIOS. With that amount of Vcore I think I can acheive possible the same if not, very close by my speculation at x9 multiplier. But I'm not interested in taking my beloved chip to the beat...not just yet or at least for quite a while.
Totally not comparable but definetely fun to see nonetheless being our G0 ancester ES week 712A.
Just got my CPU! I'm pressed for time and got my G0 Q6600 to easily run 2 orthos @ 3.6GHz. I thought I'd try to quickly see what it can do...this is all i could boot into windows at...FOR NOW;) Room ambient is 24C...unsure what it is under the desk where the puter is.
Where are you guys getting all these G0's?
Quote:
Originally Posted by Icer
Can you post batch/steeping code ?
Why is your VID 1.3000v otherwise very nice.Quote:
Originally Posted by Icer
Is that vcore accurate?
The temps and o/c vales are truly amazing speaking as a B3 stepping Q6600 owner. There are a few wildcards to me though... tell me what you guys think:
1) Vista 64-bit OS... can you use XP 32bit and confirm the temps under load?
2) Stepping of the chip could have a higher tjunction -- others have mentioned it but I don't know if it's 100 °C or not.
I did find the g0 stepping on Intel's processor finder, but I can't find any mention of the tjunction.
3) The P5K-Deluxe could somehow be reporting temps that are different.
L723A930Quote:
Originally Posted by hausner
VID say's 1.3, what does that mean?Quote:
Originally Posted by jeffyjaixx
vcore set in bios about 1.55
To be 3.6 stable vcore was 1.45
I thought VID was supposed to be like the stock voltage.Quote:
Originally Posted by Icer
Mine says 1.2625 in coretemp but I'm using 32bit XP, not sure if that makes a diff.
Okay, so 1.425 didn't work for your 3.6ghz? I'll have to see when I get home. :D
I maybe shouldn't have posted anything yet cause I haven't had time to fine tune anything.:shrug:
Hmm, that's the same stepping as me!Quote:
Originally Posted by Icer
Haven't gone over 1.4V myself since the 70 degree Small FFTs load temps were worrying me. Need to start thinking about lapping now.Quote:
To be 3.6 stable vcore was 1.45
http://www.ncixus.com/Quote:
Originally Posted by Emerica
Again, I really haven't had time to fine tune anything and I'm no longer home to be able to mess with it.Quote:
Originally Posted by jeffyjaixx
Nice thread XT, grats on a great chip too. My friend ordered a Q6600 G0 from ncix the same day you did, maybe he will get the same stepping! Good for you and him, bad for me, cuz then I have no choice but to make the jump too! :shakes:
edit: I will let you know what his results/settings are like for sure if it is the same stepping.
Told you you'd have better luck with the 9x multi :D
3.74 GHz on air is nothing short of amazing! :up:
Wonder what it'd do on better cooling...
If you are still in a room that is in the 25 - 29C ambient range, if I was you I would not trust these temps. Your "system" temp and your "HD0" temp are at or below ambient.Quote:
Originally Posted by XtremeTiramisu
If it was me I would install XP and see if the temp readings stay where they are. I am not trying to say you are cheating, just that maybe the 64 bit OS is not compatible with the temp monitoring proggies. If I am right you may well be burning up your CPU without even knowing it. Of course there is always the 99% chance that I am wrong, in which case...NICE!!! :)
Quote:
Originally Posted by jimmsch
He said his room was about 20C http://www.xtremesystems.org/forums/...&postcount=106
hey XtremeTiramisu is then any way you can show some 3DMark06 benchmark
There you go: http://forums.erodov.com/showthread.php?t=950Quote:
Originally Posted by Eightballrj
follow the steps and you will be good to go. The signed driver requirement stops Coretemp from working.
almost 3.8ghz prime stable :up:
Okay, just received a second Q6600 from NCIX (but not for myself unfortunately), it's also a L723A930. Probably won't be much better than mine so 3.3GHz on Air. Maybe I'll test it discreetly later.
Quote:
Originally Posted by Icer
I still don't trust those temps. Like I said if it was me I would check it out in XP just to be sure. I realize this screenie was taken in morning, but he has others in afternoon with same low HD and system temp.
This post
http://www.xtremesystems.org/forums/...9&postcount=11
says his ambient temps go from 21 - 30++. Even 24C HD and 25C system at 21C ambient is kinda weird. Just hoping he is not burning his chip up due to a temp monitor on 64 bit OS that isn't proper, but maybe I am just paranoid.
XtremeTiramisu's CPU is on par with the ES CPUs WOW
I'm getting around 47C on all 4 cores at idle with my Big Typhoon @ stock voltage and 8 X 400 = 3.2GHZ. You think the Typhoon is too small for the Q6600 G0 stepping I got from NCIX?
Thanks,
Y2J
Just ignore them, I got the same buggy temps (system is -31C) with Vista Business. However, the core temps is accurate.Quote:
Originally Posted by jimmsch
Yeah I was thinking the same thing. He should try to get coretemp working.Quote:
Originally Posted by jimmsch
still, very nice!....3.8+ is nothing to sneeze at with a quad core on air!:)Quote:
Originally Posted by Icer
I'm finally back.
Alright, very nice to know that a few of you can also bring the Q6600 G0 baby up to 3.7~3.8ghz range. It's definetely something out of the norm and amazing. For those of you who also got yours up to ~3.7ghz what was your cooling method, ambient temp and relative Vcore? It's always nice to make some comparason :)
Also my VID is 1.2625 so this generally means the chip has better overclock-ability @ lower given volts for the same clock as the same chip running at same clock speed. For example, someone need 1.45v to get 3.6~3.7 stable but I only need about 1.38v load to get 3.6+ stable.
VID doesn't determine your overclock-ability in terms of your chips' max FSB or clock speed but it has to do with the voltage that's needed to acheive those speed. At stock speed 2.4ghz, we each start off with different voltage being different VID from chip to chip. I started of at 1.23v (1.26v VID) at 2.4ghz and some may start off at 1.28v (1.30v VID) at the same 2.4ghz for our Q6600 G0s.
An update on the new high OC progress:
You guys all know I'm testing my xtreme quad OC on AIR right so ambient def plays a very significant role here. I was having terrific luck running 3780mhz at only 1.45v load with ambient temp being 19c~20c in the early morning.
Now the ambient has rise up to 26~28c all afternoon and it's considerable warm at the current time - 7:30pm PST so I need to back down on the clock : volts a little bit. So far I know for sure of is that this chip is capable of getting prime stable at 3700mhz ++ that's for sure. 3.8ghz may even be possible if we're on a different season :p
But you gotta love summer time and the sun shine and we're all hiding inside OC-ing like bunch of geeks...lol no offense, saying that including myself :p
So I'd say Max stable OC for this season will be in the low-mid 3700mhz and given that, it's still an outstanding OC on air. Expect 3.8ghz+ on air towards the end of the year ;)
This screen shot is just to show the effect of increased ambient temp and its relation to the core temps.
50min+ Small fft full load on all 4 cores. (This isn't the max stable OC if you haven't seen my previous screen shots)
http://img255.imageshack.us/img255/1...mbient2ke9.jpg
Anyhow, I'll update the next screen shot shortly.
Thanks to everyone who has participated in the thread :)
this is great
and it's what i was hoping they would be able to do
~3.6GHz :)
anything above that is bonus :):):):):):)
Quote:
Originally Posted by dinos22
Yep!:D
I'm already showing you proven solid 100mhz bonus at 3.7ghz on air :)
It's a RETAIL chip afterall and not one of those ES too good to be true BS.
At 3.7ghz with low volts on air, it's still PRIME STABLE during the season we OC'er hate the most - SUMMER!
Even at 3700mhz and a little higher is still consider as safe and stable for running it daily as long as you strictly keeping your 4 x 100% load 4 cores prime temp down below Intel Spec's 71c. :)
I'm sure none of the applications I do will ever get all 4 cores to run like prime..lol hence much lower core temp in the end. But I just want to use Prime's 100% x 4 cores temp as my standard.
let's hope they are ALL like your chip now :D
Not quite unfortunately.Quote:
let's hope they are ALL like your chip now
Thought I would grab this mid prime to show the vdroop.
Set at 1.56v in bios. Idle at 1.51-53. As you can see 1.45v at load.
Didn't realise your tests were small FFTs. All my previous tests were blending. Seems blending is just a tad more difficult to get stable, but runs cooler than ffts. My PWM temps are looking good:)
These temps are cpu underwater. Aircon fighting a warm Thai day.
RLM
http://img511.imageshack.us/img511/5...0bench1mb4.jpg
i bet you if you used it on the Asus mobo you would get the same result as XtremeTiramisu
agreed.Quote:
Originally Posted by dinos22
Similar result but not quite the *same* unless his VID is exactly as the same as mine at 1.2625v ;)
Also it seems like RTL is on high-end water cooling setup + Priming in Blend instead of small fft which will produce even more heat.
His ambient temp will also vary his overall core temp but not as much as air cooled quads.
I will not doubt his Q6600 OC-ability but he may ended up using a tad more volts than mine clock for clock.
But all in all, it's great to see the G0 Q6600 all have very good results :)
The future is here now! essentially, I didn't expect this day to come so soon, especially for air cooled quad core OC'ers :)
abit will use more volts and droop harder than Asus
hence why i said what i did as i've had experience with both on dual cores
quad core droop will be even worse
Quote:
Originally Posted by dinos22
True, the new P35chipset Asus P5K series has the "voltage damper" feature that's very helpful in later stages OC. When voltage damper is set to enabled, it's always 0.008v droop from idle, same for C2D and C2Q. If the voltage damper is set to dissabled, you'll get more vdroop, hence lower core temp in the end and this may or may not be a good thing.
Heh, stable or not (not), I always wanted to post a 4GHz screenshot...here it is.:D
hey man congratz on your Q6600 G0 :)Quote:
Originally Posted by XtremeTiramisu
hehehe is luihed convince.. :) I think he is in for the lookout of a good chip.. :D
hey dinos22 when is your Q6600 G0.. :D
nice.. :D hehehe.. what batch?? Q6600 is really flying.. :DQuote:
Originally Posted by Icer
I think you have me sold on the P5k Dlx, that abit is very cheap but waaay to much vdroop. Hopefully my NCIX g0 (on the way and confirmed SLACR) will be comparable to yours. Off topic but do you guys have anything to say against getting open box P5K Dlx from Newegg?
nice results guys... :up:
anyone at 4.5ghz air stable yet?
Not a great fan of Asus boards, but given the vdroop, dampeners and the like, I might of given it a bit more consideration. Still...........Quote:
abit will use more volts and droop harder than Asus
Just a few more tests. 1.63v bios 1.57 idle in guru etc.
8800 gtx 655/1055
http://img518.imageshack.us/img518/4650/3dmarkon9.jpg
http://img518.imageshack.us/img518/3...nebenchyq2.jpg
http://img260.imageshack.us/img260/4687/superpigw2.jpg
Nice:clap:
Kinda makes me wish I had watercooling.
is that orthos stable?Quote:
Originally Posted by RLM
When we seek Orthos or Prime stable, this usally means the OC'ed setting will be used as a 24/7 machine. Taking 24/7 machine into account, his given high voltage 1.63v in BIOS and 1.57v Idle to acheive 3.9ghz may already defeat the purpose of running it on a daily basis unless he doesn't give a damn about longevity of the nice new G0 chip. So it's pointless at this stage whether his G0 Q6600 is prime stable or not.Quote:
Originally Posted by easyrider
I can clearly tell by the way his screenshot was captured is that he's seeking for the max possible "bench-stable" OC'ed settings at this stage.
I know this thread is meant for Q6600 G0 Overclock-ability on air but don't missed that his nicely done 3.9ghz clock is on high-end water. :)
I wonder what's his temp are like at this OC stage on water and what could be the max OC-able speed with water cooling a G0 Q6600. It would be nice to compare whether hundreds of dollars invested in a watercooling is worth taking up the OC up to the next high level from proven 3.7ghz+ on air configuration. (Saying this becouse I'm thinking about watercooling, but short on fund :p: )
Well it would still last a few years at that voltage, so hardly a big deal considering the turn over of chips. If its prime its prime, temps and running costs aside. Its under water so I doubt the temps are as high as air.Quote:
Originally Posted by XtremeTiramisu
All that from a screenshot? This is XS you know, that's the point.Quote:
Originally Posted by XtremeTiramisu
Debatable if a swiftech kit is high end, nah that's a disservice just about I suppose.Quote:
Originally Posted by XtremeTiramisu
In the real world 200/300MHz from a decent watercooling setup would be hardly noticeable. Doesn't seem to stop anyone, always want that little bit extra. Prolongs the chips lifespan and its nice and silent.Quote:
Originally Posted by XtremeTiramisu
Quote:
Originally Posted by jeffyjaixx
Regarding chip's VID:Quote:
Originally Posted by Icer
This is what I've answered in another forum for someone who asked the same questions on VID:
VID is also one thing that we can take into account to determine the chip's overclocking-ability in the later stages. Lower VID almost always means that the overclocking room will be more. Bcos you started off with a lower Vcore at stock clock, your chip may ended up requiring much less volts to acheive the desire clock than another chip that has a higher VID than yours at the same clock. Lower volts required to obtain your desired clock speed may also mean lower temp. :)
However, keep this in mind though, although I'm stating this all based on facts and statistics but this still doesn't guarantee the OC outcome in every case, just so you know:)
Gotta say thanks for enlighten me ;)Quote:
Originally Posted by K.I.T.T.
I totally forgot about the x9 multiplier route as I was so used to x8 multiplier route with my E6700.
Quote:
Originally Posted by jimmsch
OK, here is another explanation I'm going to make again just so you're clear about the core temp reported from Coretemp and Speedfan4.32.Quote:
Originally Posted by jimmsch
Obviously you've not being reading and following my every single post carefully throughout the 7 pages, and I understand the thread is long but the infos are all there actually.
1. Forget about Vista 32bit or 64bit vs XP OS enviroment for reading the core temp correct, let me tell you simply that the OS isn't not related to the accuracy of the core temp reporting. I think I've answered about the accuracy part to someone in one of the post reply I made from the previous pages. Tjunction - DTS = coretemp and I even had a screen per the demand asking to show the Tjunction remaining from the coretemp just to show that the core temp is actually picking up the 100c Tjunction properly.
2. Pay close attention again to all my screen captures that has Asus AI suite and Speedfan 4.32 in it and look for the time for the screen that was captured. You can see that the system temp varies throughout the day. My room temp is actually 4~5c lower than the system temp reported. Do the calculations yourself and you'll find out my approx ambient temp. It's very similar to any others who also has Asus AI suite + speedfan 4.32 in the screen shots. However everyone's room conditions are different.
And obviously again you've missed the post I've made regarding the relationship btw ambient temp and OC'ed core temps in my previous page. Go dig it up and read it again.
People please don't make me explain things again for something it's already being covered. I have every motivation to show and explain all OC related subjects as clear and legit as possible so I'd hope you call can get it in the first place so no more time will be wasted from both party.
Of course, this is also one of my primary goal after I'm fully settled on a max stable clock speed : voltage : core temp strictly below Intel thermal spec of 71c on air cooling setup.Quote:
Originally Posted by ruff97
Obviously you've not being following my posts closely enough and have missed the fact that I even made a post reply to help all others who need to get Coretemp 0.95 to work under Vista x64 OS in the previous pages.Quote:
Originally Posted by Emerica
tiramisu
i agree with you there about "orthos stable" requests
there are two overclocks i have on my systems
one 24/7
the other bench stable OC
thanks for the thread it would seem that ill be getting a q6600 soon i just need to think of whether the g92 is worth waiting for
Thanks Dino for understanding what I was trying to get across there :)Quote:
Originally Posted by dinos22
Thanks for reading and best of luck with your upcomming new quad core:)Quote:
Originally Posted by Yakyb
Ladies and Gentlemen: Allow me to present you Prime Stable @ 3744mhz results !!
The system has being running nonstop for two days. So far Prime stable for 4 hours and 20mins and counting!
Overall this is one FANTASTIC chip that's prime stable on air at 1.4560v and still under Intel Thermal Spec's 71c :D I can't get any happier than this.
(Notice that my Ballistix Tracers 8500 are running below its stock clock? nevermind that bcos it's capable of going up to DDR2 1200 or 600mhz at the current given volt of 2.2 and timings ;)
My next screen shot in the near future will show much tighter latency at the current clock and voltage, hopefully)
http://img338.imageshack.us/img338/9...0q66001hr3.jpg
Tjunction Remaining: Tjunc Rem + Core temp = MAX Tjunction temp for the G0 Q6600. (100c)
Is this clear enough Spanki ??
http://img338.imageshack.us/img338/8...0q6600trj9.jpg
WOW, just WOW! You're slowly creeping up to 4.0ghz on AIR. Just crazy.
I purchased one of these bad boys (still waiting for it like many) and I'm glad to see a higher multiplier w/ lower FSB. You're assured me that having ddr2 800 that can ONLY go to about 500 mhz will more than suffice. :D
3850 is getting close, then onward to 4000+!:up:
3.7 w0w. ima hate you now :( Cant believe that is on air
Quote:
Originally Posted by XtremeTiramisu
I think you must have missed my earlier response to this, but it's worth repeating...
Note that what I highlighted above may be misleading. The only thing your math proved was that CoreTemp was in fact using 100C as it's Tjunction reference value - not that 100C is the value it is supposed to be using for this stepping.
But we already knew that it was using 100C, because it was listed on the window :). The question at hand is whether or not 100C is the proper value. Note that Tjunction is not a register on the chip that CoreTemp can simply read to find out. It's basically looking to see which cpu is being used and using a hard-coded value, depending on whether it's a mobile, conroe-duo, allendale, quad, etc.
So, until Intel verifies what value should be used, we just can't be certain that 100C is correct.
What makes me think it's changed for the G0 stepping? Here's my reasoning...
1. Temperatures being reported for the new E6x50s are way off - obviously 85C is no longer the proper Tjunction value to be used for those.
2. If you look at the Intel alert/notice about the stepping change for this particular cpu (Q6600), you'll see the following:
...also note that 'Tjunction' is also (and perhaps more properly) known as 'maximum Tcase value'. So, what Intel did with this stepping (in addition to some optimizations that allow lower overall voltages) was to "let the cpu get 11C hotter, before cranking up the cpu fan (and/or throttling itself)". This leads me to believe that the Tjunction value was also increased - maybe by 11C, but possibly by some other number.Quote:
Tcase for the Intel® Core™2 Quad processor Q6600 and Intel® Xeon® processors X3220 and X3210 on G-0 stepping has been increased by 11 oC. Tcontrol offset will remain the same relative to increase in Tcase which will help reduce acoustics
I'd also note that (at least speaking for myself), there is some confusion/ambiguity about how all of these values relate to each other... Tcase, for example, is (as far as I know) the temperature probe in the center of all cores and is the one that the BIOS and most apps report as 'cpu' temperature, but the way the above is worded, they seem to be specifically talking about 'Tjunction' (or TcaseMax)... they've obviously increased the top-end value of the allowable temperature range.
Given all of the above, I think it's a fairly safe assumption that the 'proper' Tjunction value to use for a G0 stepping Q6600 is 111C - but we really just won't know until Intel shares this information one way or the other (the only thing they've said for certain so far is that the mobile cpus use 100C - at least at the time they said that).
So basically, I don't think we can say for sure what the core temperatures truely are, but for practical/discussion purposes, using CoreTemp's numbers is as good as anything else until/unless some clarification comes from someone who really knows. As mentioned earlier, the only truley meaningfull measurement is whether or not your system is stable at some given speed. If it is (and the voltage you're feeding it is not dangerously high), then it doesn't really matter what the temperature is.
That being the case, the more meaningful number that CoreTemp can give you is the actual DTS value (enable the "Show Delta to Tjunction" option). This tells you how much headroom you have left - regardless of what all the various temperatures are reading. XT showed this @idle earlier, but idle temps never have interested me :)... I'd like to see some DTS values for some of these overclocks while under load.
Quote:
Originally Posted by Spanki
Please look at my post above ;)
I just edited it.
I hope this will clear out things
:)
I appreciate the enormous text (well, not really), but I don't think you're getting my point. Your quote above would more properly read:Quote:
Tjunction Remaining: Tjunc Rem + Core temp = MAX Tjunction temp for the G0 Q6600.
In other words, the fact that the numbers add up to equal 100C, just means that that's the number CoreTemp subtracted the DTS value from - not that that was the value it was supposed to subtract it from.Quote:
Tjunction Remaining: Tjunc Rem + Core temp = 100C (which is what CoreTemp is currently hard-coded to use as the Tjunction value for the G0 Q6600).
If you add the DTS value to SpeedFan's numbers, you'll see that it's using 85C as the Tjunction value, but that doesn't make it the right number :).
..in other words, let's assume the Tjunction value is really 110C. If that is in fact the case, you're core temperatures will be 'reported' 10C higher than they are now, but the DTS value would read exactly the same.
If the Tjunction value was really 90C, then you're core temperatures will be 'reported' 10C lower, but the DTS value would read exactly the same.
In any case, thanks for posting the "Delta to Tjunction" shots... it looks like you've still got plenty (~32C) of headroom left - which is great!
Like i said over at NCIX forums, get cracking on the game benchmarks :up:
Once again, amazing overclock dude! I hope to get myself a Q6600 or something when I finally get working.
Thanks for the thorough explanations Spanki, I think you may have the right assumptions regarding what's the actual max Tjunction temp for the new G0 Q6600 chip.
We'll just wait for Intel spec info release on this
Ok? let's move on, much appreciated for your kindly concerns :)
Agreed.. and sorry for the side-track... I don't/didn't really mean to argue about it, just mentioning it for acedemic/discussion puposes. :up:
I might be full of :banana::banana::banana::banana: :).
Don't worry, this needed to be discussed anyway and it's perfectly reasonable for me. What matters the most here is to keep the delta big enough under max load for stability.Quote:
Originally Posted by Spanki
I just gotta say, darn good chip you have there! Holding together on air w/4 cores...wow!
@XtremeTiramisu....Where did you get that chip from NCIX?
NP, thanks for trying to bring this concern to the surface :)Quote:
Originally Posted by Spanki
Yep, this is a major concern for all G0 OC'ersQuote:
Originally Posted by Xilikon
Thanks ! :DQuote:
Originally Posted by AndyM
Yes NCIX's Richmond Retail Store in BC, Canada.Quote:
Originally Posted by Brother Esau
Where exactly from NCIX doesn't matter, but I know most of NCIX Q6600 stock are G0s :)
But it's never a guarantee though, said the NCIX admins.
Ok, one last update and I'm going to have some rest......7hrs Prime Stable at the moment :)
http://img338.imageshack.us/img338/3...0q66007pf3.jpg
Congrats XT .. :)
US
Show off!! :p:
j/k ;)
Fantastic chip mate!
This whole tjunction issue really gets at the heart of the G0/B3 debate. In other words, if the tjunction from which you subtracting the DTS from to arrive at the core temp is really 111 instead of 100, the tests speedfan is reportings are off by 26 (26 too low); likewise, the coretemp temps are off by 11 (11 too low). I'd really wanna know the tjunction for sure before trading my B3 in for a G0.
First off congrats on your chip there:up: I'm struggling to get close to your results and as you know I'm on water.Quote:
When we seek Orthos or Prime stable, this usally means the OC'ed setting will be used as a 24/7 machine. Taking 24/7 machine into account, his given high voltage 1.63v in BIOS and 1.57v Idle to acheive 3.9ghz may already defeat the purpose of running it on a daily basis unless he doesn't give a damn about longevity of the nice new G0 chip. So it's pointless at this stage whether his G0 Q6600 is prime stable or not.
I can clearly tell by the way his screenshot was captured is that he's seeking for the max possible "bench-stable" OC'ed settings at this stage.
I know this thread is meant for Q6600 G0 Overclock-ability on air but don't missed that his nicely done 3.9ghz clock is on high-end water.
I wonder what's his temp are like at this OC stage on water and what could be the max OC-able speed with water cooling a G0 Q6600. It would be nice to compare whether hundreds of dollars invested in a watercooling is worth taking up the OC up to the next high level from proven 3.7ghz+ on air configuration. (Saying this becouse I'm thinking about watercooling, but short on fund )
I see your results which are at full load, tell me albeit the voltage dampening how close are those vcore readings to the ones set in bios? This thread is certainly selling Asus.
Spot on about the bench and 24/7 things. This is actually my work kit used for 3D, so all 4 cores are maxed out pretty regularly. Albeit I'm not quite as precious with this chip as I was with the QX6700, I'll be looking for something closer to prime stable 24/7. 3.7ghz is probably a good figure to consider. Make that an excellent figure:) Either way I'm still pretty smitten at being able to run those benches at 3.9.
As for the watercooling. I'm in Thailand here, so temps are evenly hot all year round. This makes watercooling a pretty easy choice. In addition I know that the bulk of the kit, rad, pump etc will get me through a few more upgrades.
That's the justifying out the way, but basically buying all these bits of kit are just part of the addiction. In no way helped by this forum:)
RLM
Quote:
Originally Posted by graysky
But... here's the thing - if the new chips are actually 11C hotter than what's being shown currently - it doesn't matter - because they have 11C more headroom to work with (everything's relative).
- Since the G0 stepping 'generally' needs less voltage for a given overclock (YMMV), then they will 'generally' run cooler than B3 stepping. This means you can either run them cooler at the same overclock, or potentially higher overclock for the same temperature (by adding more voltage). Either way is a win (again, assuming you get a good chip).
- In addition to running cooler, they might (likely) also have that extra 11C of headroom to play with before overheating, along with a scaled fan setting (for anyone who uses PWM) to keep the system quieter at higher temps (the fans don't ramp up as early).
...having said that, if you already have a B3 that you're happy with, it might be difficult to justify swapping out for a G0 (I'd probably personally wait for 45nm quad-core parts). But if you just bought it after the price drop and can still return it... that's a different story :).
ugh my G0 is has been sitting in limbo in Richmond . BC for 2 days now being sorted.