Overclocking a 9600k

[Colin Edwards]

I have a hyper 212 on my i5 and was curious as to what hertz I should try out to be safe, this is my first build so any tips will help out a lot thanks!

[KodiakWithAK]

Put the power limit the highest below 180W (the TDP of the Hyper 212, try the lowest power limit without stability issues).

 

Overclocking isn't the same for everyone (silicon lottery) so try minimal increments (25 to 50 per iteration) and confirm stability with benchmarking.

 

Don't worry about voltages.

[Fasauceome]

2 minutes ago, KodiakWithAK said:

Don't worry about voltages

That's about the worst advice you ca give for overclocking. Why wouldn't you worry about voltage, the main contributor to heat?

[dgsddfgdfhgs]

8 minutes ago, Colin Edwards said:

I have a hyper 212 on my

try add a fan in push & pull, helped me alot on that cooler, felt just warm on fins while cpu @70C+

[KodiakWithAK]

Just now, fasauceome said:

That's about the worst advice you ca give for overclocking. Why wouldn't you worry about voltage, the main contributor to heat?

I meant modifying voltages. The Hyper 212 is a cooler suitable for modest overclocking. Significant modifications to voltages would most likely provide no significant difference with the cooler we are working with. Also, voltages don't define the TDP of a processor. Though they may contribute, it's rash to base how much heat something produces by the intensity of the current is received. My 9 volt battery (a voltage 7-8 times more drastic than that of what a CPU receives idle) hooked up to a light bulb doesn't noticeably heat up to 7-8 times that of a 95W TDP CPU at idle. For OP, modifications to voltages provide more risk than reward. You should be able to recognize that.

[Fasauceome]

Just now, KodiakWithAK said:

I meant modifying voltages. The Hyper 212 is a cooler suitable for modest overclocking. Significant modifications to voltages would most likely provide no significant difference with the cooler we are working with. Also, voltages don't define the TDP of a processor. Though they may contribute, it's rash to base how much heat something produces by the intensity of the current is received. My 9 volt battery (a voltage 7-8 times more drastic than that of what a CPU receives idle) hooked up to a light bulb doesn't noticeably heat up to 7-8 times that of a 95W TDP CPU at idle. For OP, modifications to voltages provide more risk than reward. You should be able to recognize that.

As an owner of a hyper 212 and a 6 core CPU I can confirm that all of this is wrong, sorry.

[KodiakWithAK]

5 minutes ago, fasauceome said:

As an owner of a hyper 212 and a 6 core CPU I can confirm that all of this is wrong, sorry.

Congratulations, you trivialized my incredibly valid statement out of angst in a single sentence, I salute you. I'm pretty concerned for that rig if you've been modifying voltages do significantly? Sure, higher voltages = higher TDP usually. But for OP, his focus should be on finding what works for him, an owner of a non-binned processor, unlike a 8086k.

[Fasauceome]

Just now, KodiakWithAK said:

Congratulations, you trivialized my incredibly valid statement out of angst in a single sentence, I salute you. I'm pretty concerned for that rig if you've been modifying voltages do significantly? Sure, higher voltages = higher TDP usually. But for OP, his focus should be on finding what works for him, an owner of a non-binned processor, unlike a 8086k.

here's some legitimate advice for OP then:

Temperature is very sensitive to voltage. The change from 1.17 to 1.18 volts can be as much as a couple degrees, and that's rather significant, especially considering the context of auto vs manual voltage, and getting as much performance as possible. Leaving auto voltage is usually a big no-no if you can be bothered to use any kind of offset.

Reduced voltage is no harm to a CPU whatsoever, it's far from harmful to tune your voltage if it's below the auto voltage set by a given BIOS.

Naturally a light bulb and CPU are producing different amounts of heat, they're processes of different efficiencies and scale.

 

I'm in a ragged mood and I seem to be looking to throw insults and start fights. Definitely isn't doing me any favors.

[KodiakWithAK]

2 minutes ago, fasauceome said:

here's some legitimate advice for OP then:

Temperature is very sensitive to voltage. The change from 1.17 to 1.18 volts can be as much as a couple degrees, and that's rather significant, especially considering the context of auto vs manual voltage, and getting as much performance as possible. Leaving auto voltage is usually a big no-no if you can be bothered to use any kind of offset.

Reduced voltage is no harm to a CPU whatsoever, it's far from harmful to tune your voltage if it's below the auto voltage set by a given BIOS.

Naturally a light bulb and CPU are producing different amounts of heat, they're processes of different efficiencies and scale.

 

I'm in a ragged mood and I seem to be looking to throw insults and start fights. Definitely isn't doing me any favors.

I feel we stray from the essence of this conversation, is the modification of voltages significant to OP for obtaining a modest overclock.

 

I agree, this conversation, particularly my side, has lacked professional connotation and I apologize.

[xg32]

24 minutes ago, KodiakWithAK said:

Put the power limit the highest below 180W (the TDP of the Hyper 212, try the lowest power limit without stability issues).

 

Overclocking isn't the same for everyone (silicon lottery) so try minimal increments (25 to 50 per iteration) and confirm stability with benchmarking.

 

Don't worry about voltages.

i WOULD NOT pump 180w into a 212, the actual tdp is somewhere between 120-140w without melting the cpu at 100C+

 

30 minutes ago, Colin Edwards said:

I have a hyper 212 on my i5 and was curious as to what hertz I should try out to be safe, this is my first build so any tips will help out a lot thanks!

That said, 1.3v 4.7 avx is a good starting part, make sure the temperature is under 90C on full load and the power usage per hwmonitor is below 100w

 

iirc 1.3v 4.9 avx was barely 90w on my 8600k, whether or not u hit 4.9 depends on silicon lottery, most of them hits 4.7

[KodiakWithAK]

Just now, xg32 said:

i WOULD NOT pump 180w into a 212, the actual tdp is somewhere between 120-140w without melting the cpu at 100C+

 

That said, 1.3v 4.7 avx is a good starting part, make sure the temperature is under 90C on full load and the power usage per hwmonitor is below 100w

 

iirc 1.3v 4.9 avx was barely 90w on my 8600k.

Yeah, I actually don't remember what I was thinking when I wrote that. I believe I was thinking the top of spectrum allowed in a BIOS and my mind defaulted to 180.

[Fasauceome]

2 minutes ago, KodiakWithAK said:

is the modification of voltages significant to OP for obtaining a modest overclock.

In terms of the 9600k, the Hyper 212 is a case where sensitive voltages would be rather useful to control. A BIOS setting something like a 4.7GHz overclock with 1.3 volts would be rather hot for it, when the chip may well get away with 1.25, or even close to 1.2 if you're lucky.