Erfahrungen bei Tempauslese?

scadell

Lieutenant
Registriert
Sep. 2002
Beiträge
712
Hallo

Hab mal ne Frage bezüglich der Genauigkeit der Temperaturauslese an einem Abit NF7-S mit neustem BIOS. Wie genau ist es? Und muss ich etwas dazurechnen? Ich frage weil mein Barton 2500 da laut auslese eine Temp von 32° hat. Zwar läuft da nix (Programm oder Game) aber ich bvenutz auch keine Tools um den Prozzi in den unbenutzten Phasen in den Ruhemodus zu bringen. Ich hatte vorher einen Xp2200 und der wurde schnell selbst im Ruhemodus 50-55°. Dieser Unterschied ist ja heftig. Der einzige unterschied zu den beiden Prozzis in meiner Konfig ist das ich allein andere Wärmeleitpaste benutze sonst hab ich nix am System verändert.

MfG Scadell
 
sogut ich weiss is abit da etwas ungenau beim nf7 aber das mit dem 2000+ kann ich dir erklären:
das war ein palomino kern und die dinger hatten 55 also standart.

falls kein palo hattest du ne scheiß kühlung
 
danke für deine antwort.

Mein Vorgänger Prozzi war ein Throubred Kern und zwar ein 2200. Und ich hatte auch keine andere Kühlung als ich sie jetzt bei dem Barton verwende. Das einzige was anders ist im gegen satz zu dem Throubred ist das ich die Premium Wärmeleitpaste von Coolermaster benutze. Sonst ist alles gleich. Nagut die Aussentemperaturen damals waren ja auch etwas höher aber denoch ist es ein gewaltiger unterschied von Throubred auf Barton was die Wärmeentwicklung angeht.
Gerade steht bei meinem Temps das meine systemtemp bei 18° liegt und meine CPU temp bei 29°. Gut es ist nun auch recht kühl heute morgen aber solcher werte waren bei meinem Throubred unmöglich. Also liegt es nun echt an dem Barton und der neuen Wärmeleitpaste? Und wenn ja wieviel grad muss ich beim Abit dazuzählen wenn es nicht genau ausliest?


MfG Und einen guten Morgen

Scadell
 
mit mbm5 kannst du die temperatur der prozessorintegrierten temperaturdiode auslesen, dann haste ein 100%ig genaues ergebnis!
das tool gibts hier.
 
hmm ich hab zwar kein abit board, aber bei mir ist es ähnlich, seit meinem 2400+ liegt die temp im ruhezustand bei ~29 Grad
Ich traue dem ganzen aber noch nicht so, da ich auch nur einen arctic cooling copper silent 1 habe.....
mbm 5 sagt übrigens die gleichen werte wie das bios, evtl. sogar etwas weniger
 
@ mbm

hi also ich habe nen 3200+ und nen revoltec avalance TMD drauf, mein epox 8rda+ (neustes bios) bios und mein epox bios toll sagen so bei idle 42° aber wenn ich beim motherboard monitor die dioden von winbond 1 , 2 und 3 auslesen lasse, dann sagt mir mbm das die cpu 127° hat!

wie bekomme ich den nun die richtige"real" temp raus?
 
Die Temperatur stimmt, doch sie ist nicht Grad Celsius sondern Fahrenheit.....
musste im mbm umstellen, direkt unter Generals
 
Hi,

welches OS hast du? XP versetzt die CPU in Ruhephasen selbstständig in den HALT-Modus. Ich glaub der heisst so. Allerdrings funktioniert das bei AMD-Cpus nicht immer 100%. Bei meinem alten Palo klappte es immer nur sporadisch. Vielleicht klappts ja beim Barton besser. Ich benutzt jedenfalls CPUIdle:

34°C (idle) @ Boxed Kühler (nimmer lang) @ 9V

Friday
 
Zuletzt bearbeitet:
Temperatur

@ FRIDAY
mal eine Frage,wie versetzt Win XP die CPU in den
IDLE Zustand?
Muss man das einstellen,oder erledigt das
Win XP selbstänig.
Habe schon viele Tools ausprobiert,
aber immer Probleme damit.
Meine CPU ist ein Athlon 2000+ Palomino.
zur Zeit :Board 28°, CPU 35°.
mfG bennue
 
How does CpuIdle work?

Why CPUs Produce Heat

Due to CMOS technology modern microprocessors are manufactured not only with transistors but also capacitors and resistors. The resistors create heat that must be conducted away from the processor.

Classic Cooling Methods

Removing the heat from the CPU core is easier with higher temperature gradients. For this reason microprocessors are not only equipped with large heat sinks but also actively cooled with fans or Peltier cooling elements. Higher temperatures worsen parasitic effects like electromigration or oxide breakdown that might eventually lead to failure. A temperature increase of 10 degrees centigrade means a halfed lifetime for chips. This holds true for the microprocessor and for the voltage regulation module (VRM) that supplies the I/O and core voltages. Many outdated or cheap motherboard designs employ passive VRMs that generate additional heat in the PC.

The CpuIdle Approach

Under normal circumstances the CPU isn't always active but spends much time waiting for the keyboard, harddisk or CD-ROM. What would be more logical than to turn off the CPU for that period? That's exactly what the HLT machine instruction (Opcode F4) does. Whenever the CPU encounters a HLT instruction the clock is halted and the CPU enters suspend mode until an interrupt, NMI, or reset happens. With the advent of power saving microprocessors like the Cyrix Cx486S the HLT instruction elicits an additional benefit. When "Suspend on HLT" is enabled in the configuration register the processor not only stops on HLT but also enters the power saving suspend mode.

Why Doesn't Windows Do That?

Modern operating systems like Linux execute the HLT instruction in an idle priority thread. This thread is always executed when the CPU is otherwise idle. No additional execution time for HLTing is needed, the CPU will not run slower.

While other operating systems like Linux always used this mechanism, Windows only learned it with NT. But even with NT and following versions it is only enabled when the BIOS and ACPI implementation is recognized by the OS.

The Chipset Part

With modern chipsets like VIA KX or SiS 7x5 an additional variable enters the equation. These chipsets need to be specifically programmed to allow the CPU entering the HLT state. Windows itself does not perform this programming (chipset optimization in CpuIdle), the BIOS must handle it- which many, even modern, BIOS implementations don't do.

Für ne deutsche Version müsst ihrs halt über altavista / Babelfish übersetzten ;)

FRIDAY
 
Zurück
Oben