Thread: Over clocking to 3.4 GHz on stock cooling on i7 920 D0 ok?

You're liability is fine so far, but I wouldn't push it any higher at this time. As you're aware, the stock cooler is unhappily insufficient, so before you get your new cooler, here's what you require expressive.

My objective is to assure know that enthusiastic over clocking are all about stipulation, so it's very important to be specific. If we're not, then the topic makes about as much sense as comparing apples-to-oranges thermal fruit salad in a blender! My objective is to assure that know Intel's stipulation, principles and test method, so they can superior make a decision how to apply and manage their over clocking option.

Intel's second and frequently misunderstands Thermal state, T-junction Max, apply to over temperature protection such as strangle and power cut, so you don't toast your transistors. As such, any Core temperatures which exceed 73c should be considered "over temperature". Further, when specifications are exceeded, then processor poverty becomes a concern, which is explain in the following Anand Tech

"Section 1: Introduction

Core I and Core processors have to different types of temperature Session not computer centered under the Cores and Digital Thermal Session located on each Core. The case events which are CPU high temperature and the Session measure which is Core temperature. Since these sensors calculate to separate thermal levels, there is a 5c temperature difference between them, which is Gradient. Core 2 Quad’s have Session, while Core 2 Duo's have Session.

The monitor utilities provide by motherboard production monitor temperature, while a number of all the rage freeware utilities monitor Core actual temperatures is recommended for users attracted in monitoring Core temperatures only Speed Fan monitor Core temperature The Thermal Specification shown in Intel's Processor Spec Finder is Tcase Max CPU not T-junction which is a very common misconception among most enthusiasts. Since there's a 5c gradient between the computer session and the Core sensors, shown in the following Intel document.

Applications

Prime95 Small FFT's is the normal for process of testing, because it's steady-state workloads which yield steady-state temperatures, while Blend is a reminiscence cyclic workload which yields variable processor temperatures. Small FFT's will reach thermal saturation within minutes, so a minute test is adequate. Thermal testing should be conducted as close as possible to Standard ambient, with case covers removed, the computer clear of any desk enclosures, and all fans at RPM to eliminate cooling variables, and to produce consistent and repeatable results for comparisons. If the Gradient between CPU temperature and "mean" Core temperature is incorrectly coded. CPU temperature and Core temperatures can be individually calibrated in Speed Fan by following the Calibrations Section in the Temperature Guide.

Burn Test use LinPack, which shows signature that look like the ups and down of a shocking day on the stock market, and cycle between light workloads, through test segments which spray all processor registers with all one's, and can push an over clocked with HT enabled at Vcore right on past Tcase Max to ring the T-junction Max bell like a fire alarm!

Since there are extremely few applications or games that will point, let alone continue processor workloads beyond utilities which load all registers with one’s entire are not envoy of real-world compute. While these utilities are certainly very useful for stability testing, they are inappropriate for thermal testing. The 3DMark benches are excellent for stability testing, as are application for rip and programming.

In the direction of creation of sense of computer chip temperature and Core temperature, compare them to a 4 cylinder car with 5 temperature gauges; 4 of the gauges are cylinder head temperatures and the 5th gauge is the on the whole engine heat which is 5c lower than the other gauges, and is the temperature gauge with which we're all familiar. We know the red zone (hot) for the engine temp gauge and 73c on the cylinder head temp gauges, but if we push the engine too hard and peg all the gauges, then the engine will shut down.

To make sense of CPU temperature and Core temperature, evaluate them to a 4 cylinder car with 5 temperature gauges; 4 of the gauges are cylinder top temperatures and the 5th gauge is the overall engine temperature, which is 5c lower than the other gauges, and is the temperature gauge with which we're all familiar. We know the red zone for the starts at 68c on the engine temp gauge and 73c on the cylinder head temp gauges, but if we push the engine too hard and peg all the gauges then the engine will close down.

If you'd like to learn more regarding CPU temperatures, then now tick on the connection in my signature

Wow, very informative post! I did scaled my CPU back since right now I'm only doing it for gaming and my PC no closer to it is@ 3.4. I did note that the PC running that quick made it hotter in my casing making my video card awfully warm got all the way to 89c. So until I add another video card there really inst any gains to be had now. But the new cooling setup should help that out a lot and hopefully make the rest of my components cooler as well.

Also, you are correct about 68c on the specifications from Intel. 67.9c is the max, i will definitely not over clock past the correct clock of 3.1 until the cooling solution comes on this Thursday.

So a quick update, the proli. And noctua fan came today. I have the cooler on a push pull setup with a case fan I had in the case. At stock the system run extremely cool load. But when over clocking watch out! The system runs load. If the case cover is on add another 5c. Testing is being performed with OCCT. System is not stable with over 3.4 GHz on full load never passes of testing without freezing. Right now I'm keeping her @ 3.1. I've been testing for a few hours now and even though I can boot into windows and play the re5 benchmark, act will not get past on anything but 3.

I can my 955 be to 3.83 and run everything stable crisis full settings and all. Leave the pc on for days no problems but when I run Nero it blue screens soon after the burn process begins any ideas? I can lower my clock to 3.7 and it won’t blue screen. Well then its not 100% stable is it, try lowering the memory speed, if not, be content with 3.7.