Associate
- Joined
- 19 Oct 2008
- Posts
- 856
after about 10 minutes of prime test my processor has maxed out at 78 degrees is this too much for a e8400 its has been overclocked to 3.85
is 78 degrees to hot on prime test
- Thread starter EpicCarnage
- Start date
Yeah 78°C @ 3.85GHz on a dual core is hot!
Having said that remember it's a stress test and your unlikey to load up the chip that much in real world scenarios unless your doing a lot of Video-Ecoding, Folding etc that puts a heavy load on all the cores . . .
What vCore are you using? . . .is it possible to turn it down? . . .is it on [Auto] or?
Having said that remember it's a stress test and your unlikey to load up the chip that much in real world scenarios unless your doing a lot of Video-Ecoding, Folding etc that puts a heavy load on all the cores . . .
What vCore are you using? . . .is it possible to turn it down? . . .is it on [Auto] or?
Yeah . . or it's a Overvolted low VID chip!
I don't think 78C under full load at 3.86GHz is silly-hot, but it is on the high end of what I would expect. I think the thermal maximum on those is 100C so you're well off that, but it might be worth reseating the cooler if it really worries you. I wouldn't bother to be honest.
Every chip is different and E8400's come in at least two steppings (E0 run hotter iirc) so it could be just fine.
Every chip is different and E8400's come in at least two steppings (E0 run hotter iirc) so it could be just fine.
Yeah . . or it's a Overvolted low VID chip!
Pardon? Temperature is very dependent on the voltage being pushed through the chip but independent of the chip VID.
To get 80 degrees out of an e8400 using a TRUE I think you'd have to be pushing the bad side of 1.45V, though all I can say for certain is that my one never came close to that temperature until I deprived it of airflow and used a £9 cooler.
You said to get those temps he must have a ****** cooler and I agreed with you, I also went on to add that another possibility is that in fact he may actually have a reasonably good cooler but is using a low VID over-volted processor
low VID = Processor that has been set to run using low vCore . . .
My theory on this is simply the batch of silicon that these chips are made from do not react well to voltages and are therefore stamped as Low VID.
When an overclocker pumped additional voltages into these particular chips they freak out completely and the chips temp *FLY* through the roof . . .
I was made aware of this by gurusan and can confirm from my own testing this does indeed seem to be the case, in short low VID sucks for an overclocker who wants max frequencies from his chip and is using air cooling!
That's very interesting. I can't say I understand it though, probably because I don't know why some chips have low vid and some higher. Some i7 chips are described as leaky and draw far more current than the norm at a given voltage, perhaps this is related?
Vid doesn't seem to correlate very strongly with maximum overclock. The general opinion is low vid => higher clock, yours is the first argument I've seen for the opposite. I'm inclined to think that there's no relation but it's a shame to throw out one of the few pieces of data available to work with. Do you know of any theory behind this, or is it just a pattern some people have noticed?
Vid doesn't seem to correlate very strongly with maximum overclock. The general opinion is low vid => higher clock, yours is the first argument I've seen for the opposite. I'm inclined to think that there's no relation but it's a shame to throw out one of the few pieces of data available to work with. Do you know of any theory behind this, or is it just a pattern some people have noticed?
If Gurusan is repeating/promulgating a theory, there will be something behind it. He has a practical engineering background and a good understanding of the physics. I'll be honest, I quite often don't follow all the theory, but every time I've challenged what he's said he's always been proven correct by my own testing.
And Big.Wayne's no mug either
The way I understand what Big.Wayne is saying is that some chips run the Intel QA tests at extremely low voltages. They are therefore give low VIDs at the factory. These CPUs therefore run exceptionally cool at stock. When they have big voltage applied to them however, whatever gives them their low VID can't handle the voltage and they heat up out of proportion to the voltage applied. I would have to say that this certainly seems to be the case with many CPUs I've owned which have run very cool at stock, and even run with a moderate overclock when extremely undervolted, but required quite small increases in terms of overvoltages to turn them into little furnaces.
And Big.Wayne's no mug either
The way I understand what Big.Wayne is saying is that some chips run the Intel QA tests at extremely low voltages. They are therefore give low VIDs at the factory. These CPUs therefore run exceptionally cool at stock. When they have big voltage applied to them however, whatever gives them their low VID can't handle the voltage and they heat up out of proportion to the voltage applied. I would have to say that this certainly seems to be the case with many CPUs I've owned which have run very cool at stock, and even run with a moderate overclock when extremely undervolted, but required quite small increases in terms of overvoltages to turn them into little furnaces.
We just have to look at the facts as they present themselves and Theorize ourselves. I believe that silicon from different batches have different properties, that would be advantages and disadvantages. The advantage of Low VID silicon is that it is able to run stably at a lesser voltage than silicon that is stamped at high VID, the disadvantage of low VID silicon is that adding additional vCore is like waving a red flag in front of a bull, it doesn;t like it and the temps *charge* upwards heh!
Conversely silicon that is stamped as HIGH VID is unable to run stably at the given frequency without additional voltages however . . . even though you need to pump the vCore on these HIGH VID chips they do not react badly to the additional voltages so you can raise the vCore extremely high and still keep the processor cool enough using a premium aircooler
In short, a low VID chip is great for
Serious Overclocker/Benchmarker who wants maximum processor frequency and has good custom water-cooling or similar . . . . A low VID chip is also good for a ULV freak who wants to run a modest overclock using sod all voltages . . .my last E8400 was quite happy to run rock solid stable at stock 3.0GHz using 1.00vCore . . .
A HIGH VID chip seems to be better suited to your average overclocker who is using modest air cooling, they will need to raise the vCore a lot higher but the silicon does get get agitated and the temps remain reasonable.
You have to keep things in context as your truncating the information you are reading, I am not disagreeing with the general consensus but the full statement would read like this:
"The general opinion is low vid => higher clock if you can cool the chip sufficiently"
Last edited:
Interesting, cheers. I think I'll try to look into this from a materials perspective, there must be countless papers written on how to better make processors. Some might be in the public domain.
Learnt something new here. Much appreciated
Learnt something new here. Much appreciated
Both of my previous intel quads (q6600 and q9550) were sort of middle of the road chips vid wise, q6600 @1.2750, q9550 @1.200. The q6600 clocked to 3.8 ghz on 1.51 vcore, and hit temps of 81-81-72-72, lapped it and the cooler and changed the fans which brought it down to 68-68-62-62. The q9550 hit 3.8ghz on 1.21 and topped out at 57-52-54-52. In relation to vid and how it effects overclocking, i think a middle ground vid gives the best of both worlds, good overclocks but cool running particularly, for someone on air cooling.
- Joined
- 19 Oct 2008
- Posts
- 856
i'm at 1.36750 and using a zalman cpu cooler the led one its not the best tbh woahhhhhhhh boy was i wrong about what i had the cpu voltage at i had it at 1.51 but don't worry been solved now i have it at 1.38 and im idle @ 50 and 65 on full load so thanks for you help everyone
Last edited:
your cooler may not be seated right since im running a Q9400 at 3.6 at 1.35volts and it never gos over 55c (my cooler is a Arctic Cooling Freezer 7 Pro)
