Intel Core i2600K | Asus Maximus IV Extreme (3.0) | Kingston 9905403-400.A00LF DDR3 @1600MHz | 2x EVGA GTX 570 SC | OCZ Vertex 3 SSD
Verschil
TJMAX 85 IDEL: 23-24 LOAD: 42-44
TJMAX 80 IDEL: 17-18 LOAD: 38-39
Dit met een E6600 @ 3.4 1.45Vcore LOAD (1.5BIOS) .
Ik vind ook dat deze tempratuur erg laag is..
Intel Core i2600K | Asus Maximus IV Extreme (3.0) | Kingston 9905403-400.A00LF DDR3 @1600MHz | 2x EVGA GTX 570 SC | OCZ Vertex 3 SSD
Nu idlen ze op 3,6GHz rond de 46 graden, dat zou dus rond de 36 moeten zijn ?
Als dat echt zo is ben ik onder de indruk (en is de verleiding groot om toch nog wat meer Voltjes door de cores te jagen om nog iétsjes hoger te komen).
Consensus daar is dat de intel waardes toch wel iets aan de lage kant zijn, iemand iheeft een gaatje in zijn heat spreader geboort om de temps te meten etc. wel grappig. Real Temp houd trouwens een tjmax van 95 graden aan voor mijn G0 Q6600 (in de huidige versie).
@KneoK
Ik zou niet meer dan 1.5v real door die cpu jagen...
Intel Core i2600K | Asus Maximus IV Extreme (3.0) | Kingston 9905403-400.A00LF DDR3 @1600MHz | 2x EVGA GTX 570 SC | OCZ Vertex 3 SSD
Koeler: Scythe Kama Cross met 120mm ventilator gemonteerdnarcan schreef op donderdag 30 oktober 2008 @ 15:00:
Ja dat zijn idd erg lage temps, zitten ze boven je kamer temps? Wat is eigenlijk de ambient temp en wat voor koeling gebruiken jullie?
Ambient temperatuur: 21°C
Mijn Core 2 Duo E6420 draait op standaardspanning 3000MHz en doet hierbij:
TjMax=70°C
• Idle: 12°C en 9°C
• Load: 40°C en 41°C
Die idle temperaturen kunnen natuurlijk nooit kloppen want ze liggen ver onder de omgevingstemperatuur. Nu is het wel zo dat de core sensoren niet erg nauwkeurig zijn bij lage temperaturen aangezien ze normaal alleen gebruikt worden om de thermische beveiligingen in te schakelen. Maar 9°C lijkt me toch wel erg laag en de load temperaturen klinken natuurlijk wel leuk maar ik heb er toch sterk mijn twijfels bij
Ja idd, de maker van Core Temp gaat zijn tjmax niet veranderen en de maker van Real Temp gelooft het ook niet zo. Ben wel benieuwd waar het schip strand, ik denk zelf nu dat de oude tjmax waardes toch wat realistischer zijn, wat intel ook zegt.Admiral Freebee schreef op donderdag 30 oktober 2008 @ 19:13:
[...]
Koeler: Scythe Kama Cross met 120mm ventilator gemonteerd
Ambient temperatuur: 21°C
Mijn Core 2 Duo E6420 draait op standaardspanning 3000MHz en doet hierbij:
TjMax=70°C
• Idle: 12°C en 9°C
• Load: 40°C en 41°C
Die idle temperaturen kunnen natuurlijk nooit kloppen want ze liggen ver onder de omgevingstemperatuur. Nu is het wel zo dat de core sensoren niet erg nauwkeurig zijn bij lage temperaturen aangezien ze normaal alleen gebruikt worden om de thermische beveiligingen in te schakelen. Maar 9°C lijkt me toch wel erg laag en de load temperaturen klinken natuurlijk wel leuk maar ik heb er toch sterk mijn twijfels bij
Intel Core i2600K | Asus Maximus IV Extreme (3.0) | Kingston 9905403-400.A00LF DDR3 @1600MHz | 2x EVGA GTX 570 SC | OCZ Vertex 3 SSD
Ach in het artikel staat:Admiral Freebee schreef op donderdag 30 oktober 2008 @ 19:13:
[...]
Koeler: Scythe Kama Cross met 120mm ventilator gemonteerd
Ambient temperatuur: 21°C
Mijn Core 2 Duo E6420 draait op standaardspanning 3000MHz en doet hierbij:
TjMax=70°C
• Idle: 12°C en 9°C
• Load: 40°C en 41°C
Die idle temperaturen kunnen natuurlijk nooit kloppen want ze liggen ver onder de omgevingstemperatuur. Nu is het wel zo dat de core sensoren niet erg nauwkeurig zijn bij lage temperaturen aangezien ze normaal alleen gebruikt worden om de thermische beveiligingen in te schakelen. Maar 9°C lijkt me toch wel erg laag en de load temperaturen klinken natuurlijk wel leuk maar ik heb er toch sterk mijn twijfels bij
Any DTS reading below 50°C should be considered to indicate only a temperature below 50°C and not a specific temperature.
Dat is toch wel een bijzondere uitspraak overigens... maar goed zolang er onder de 50c-60c geen deterioration plaatsvind is het genoeg om de fanspeeds op aan te passen, maar wel lekker goedkoop van intel.
morituri nolumus mori
Ik zou het wel heel leuk vinden dat als ik hem op 80C zet mijn temps rond de 16 graden liggen, maar het lijkt mij iets te mooi.
Ryzen 7 3700x // 32GB Crucial Ballistix Sport LT 3200 // Gigabyte Aorus X570 ULTRA // Gigabyte GTX 1080 Ti Aorus Xtreme 11GB // Corsair Force MP600 1TB // Intel 660p 2TB // Samsung Evo 850 1TB // EVGA SuperNOVA 750 G2
Klopt, ik zou ook wel verder willen klokken danShoarma kip schreef op vrijdag 31 oktober 2008 @ 09:02:
Ik zou het wel heel leuk vinden dat als ik hem op 80C zet mijn temps rond de 16 graden liggen, maar het lijkt mij iets te mooi.
Als ik antwoord heb zal ik het hier posten.
Edit* typo
Hello Frank, Thank you for contacting Intel(R) Customer Support. I understand you want information on the Thermal Junction for the Intel(R) Core(TM) 2 Duo processor E6600. Since there are many variables to determine the Thermal Junction, we do not have a fixed value. Furthermore, you can review the information in the following link for instructions on how to determine the value: http://www.intel.com/supp.../pentium/sb/CS-011039.htm Please do not hesitate to contact us again if you need further assistance. Sincerely, Cindy J.Intel(R) Technical Support
Dus intel zelf weet ook de TJMAX van hun eigen proccesor niet
[ Voor 3% gewijzigd door frank131331 op 31-10-2008 21:13 ]
Heb een beetje het idee na veel lezen, dat het echt per cpu verschillend is, er zijn zowiezo al veel cpu's die de zelfde stepping hebben en een andere VID, wat al veel uitmaakt.frank131331 schreef op vrijdag 31 oktober 2008 @ 21:10:
Ik heb inmiddels antwoord van Intel:
Hello Frank, Thank you for contacting Intel(R) Customer Support. I understand you want information on the Thermal Junction for the Intel(R) Core(TM) 2 Duo processor E6600. Since there are many variables to determine the Thermal Junction, we do not have a fixed value. Furthermore, you can review the information in the following link for instructions on how to determine the value: http://www.intel.com/supp.../pentium/sb/CS-011039.htm Please do not hesitate to contact us again if you need further assistance. Sincerely, Cindy J.Intel(R) Technical Support
Dus intel zelf weet ook de TJMAX van hun eigen proccesor nietIk weet dat elke CPU verschillend is dus ook verschillende clocks/temps zal halen. Maar er moet toch wel een vaste Thermal Junction zijn voor de stepping?
Ben ook aan het PM'en met de auteur van Real Temp, die het volgende zei:
The best test is to run your CPU at low MHz and low voltage and let it idle and see what your core temperatures are with the case open and your CPU fan on high. 266MHz x 6.0 ~ 1600 MHz and 1.10 volts as reported by CPU-z works good for this test. You also will need a reasonably accurate room temperature measurement. Send me a screen shot or two and then we can discuss what is the most likely temperature. For me, during this test, the results are usually pretty obvious.
Ik ga dit van de week even doen en dan laat ik nog wel weten hoe hij zijn conclusies trekt.
Intel Core i2600K | Asus Maximus IV Extreme (3.0) | Kingston 9905403-400.A00LF DDR3 @1600MHz | 2x EVGA GTX 570 SC | OCZ Vertex 3 SSD
Ik heb de hele week al lopen googlen. Maar heb nog weinig bruikbare info gevonden.
Dus door de idle meting te doen en de temperatuur tabel van rge te volgen, kun je bepalen hoeveel graden je boven de kamer temperatuur zit en daar uit kun je afleiden wat je tjmax zou moeten zijn.The important screen shot is the one where you were running at 1600 MHz and low voltage at idle. My testing was done at 1.08 volts so you are close enough to that for comparison purposes.
rge has also been doing some testing and has come up with a table showing what sort of temperatures a user should expect compared to their room temperature for his 45nm E8400. I think for a Q6600 I would add 1C to 2C to his guidelines:
http://www.xtremesystems.org/forums/...postcount=2429
If your room air temperature is about 19C and your CPU is reporting 25C to 27C at idle, that's 6C to 8C above your air temperature. To me that says when you use TjMax=100C, your reported temperatures are right where they should be. Thermometers and testing procedures are never 100% accurate so a degree or two of difference is not unusual.
TjMax=90C for your CPU is just not possible. Quite frankly, Intel's recent IDF news release is not possible for any of the 65nm processors that rge or I have tested. I'm in an indirect conversation with some top engineers at Intel that are going to try and explain why their numbers go completely against the testing that's been done. That should be interesting!
Have you ever noticed what's reported for APIC ID in the upper right information window of RealTemp? To the right of where the MHz is displayed is a small button that will cycle through some additional information. Users tell me I need to put a small label on that toggle button but I haven't gotten around to it yet!
Most temperature monitoring software like Core Temp and I believe Everest, have decided to ignore APIC ID. On a Quad, this means what they display for core0, core1, core2, core3 might not correspond to how the cores are physically laid out within the CPU. They assume that APIC ID is always 0123 but if APIC ID is 0213 then that means they will misreport the two center cores. What they report for the temperature of core1 is actually coming from the sensor on core2 and what they report as the temperature for core2 is coming from the core1 sensor.
If APIC ID is not 0123 then it can be very difficult to compare two different programs. My Q6600 was APIC ID 0123 for 2 or 3 months and then one day it just changed and stayed reversed for a couple of months. I think it is back to 0123 now but I haven't used it recently. A bios update might cause this but some users have reported that their motherboards will change the APIC ID with every re-boot.
Let me know if you need some more info to clarify APIC ID. It's an interesting topic that has been completely ignored by most software.
Overall your Q6600 looks very good and should accurately report your core temperatures from idle to TjMax which for you is 100C.
If you run Prime95 with the small FFT option it is typical for core0 and core1 to report the exact same temperature and for core2 and core3 to report the same temperature. For some reason, core2/core3 are frequently a couple of degrees cooler than core0/core1.
Uit een andere pm:
Ben benieuwd of intel de 65nm core 2 tjmax waardes ooit nog rectificeert, zal wel niet.My contact at Intel was going to head back to the lab to see if he could reproduce my 65nm TjMax concerns. So far I haven't heard anything. With the release of Core i7 maybe they hope this issue will just fade away.
Intel Core i2600K | Asus Maximus IV Extreme (3.0) | Kingston 9905403-400.A00LF DDR3 @1600MHz | 2x EVGA GTX 570 SC | OCZ Vertex 3 SSD