Finally overclocking my i5 4670k

[blazerclaw]

What's up people, I'm finally getting to overclocking my CPU after having it for roughly 3 years. It's my first time OC so I researched a bit and decided to give a whack at it. Anyways, right now I'm running my CPU at 4.3 GHz  Core Voltage 1.26V. I've been running MSI Kombuster's CPU Burner and Core temp to stress test and keep track of my temperatures for past 45 mins. Oh yeah I forgot one thing, I'm running an EVO 212 for my CPU Cooler. Here's my results so far in the picture. Looks like I'm getting around an average of 68c Max Temperature. What do you guys think? Should I tone my voltage down? Or is it fine at where it is? Also is the CPU stress test I'm using any good? Any feedback is appreciated.

[Narnash]

68°C under synthetic max. load is perfectly fine (you even have much room up, I usually target 75°C but temps till 80°C-85°C are OK to, keep in mind it is a synthetic max workload) what is your ambient temperature during the test? Since a cooler is much less effective how closer the gab between ambient temperature and cooler temperature is.

 

And if you want to test for a 24/7 setting, ramp up your GPU (especial if you have a GPU with an open cooler design) a bit, because it can effect the temperature in your case.

[blazerclaw]

10 minutes ago, starcoaster said:

I use intel burn test to stress mine, if you can run IBT stable you can run ANYTHING stable

here's some usual voltages for clocks (quite conservative, you WILL be able to tune these down a bit)

 

1.26@4.4GHz

1.275@4.5GHz
1.3@4.6GHz

I run 1.35@4.7GHz

 

You won't be able to get much higher than 4.5/4.6 on that cooler if you're lucky. You want to monitor temps and keep them under 85C

To begin with, use the stress testing tool in CPU-Z, if that passes, then fire up Intel Burn Test on 1024Mbit of RAM (standard seting) and set the trials to 10 passes, then hit run. Monitor the temps, if it exceeds 85C, shut the test down. You WILL get higher temps on IBT as it is an incredibly stressful Linpack test. My watercooler exceeds 85C (do as i say, not as i do) on my clocks and voltages.

Unfortunately, I lost the silicon lottery since my build crashes booting to windows when I try 4.5 GHz at 1.3V, so 1.275V doesn't seem like it would be able to hold up. Similar thing with 4.4 GHz, it won't boot at 1.26V, but it will at 1.30V, and for that I ran tests for 20 minutes it was stable for that amount of time, the temps were at low-mid 70s the max temp being 78c. 

[blazerclaw]

33 minutes ago, Narnash said:

68°C under synthetic max. load is perfectly fine (you even have much room up, I usually target 75°C but temps till 80°C-85°C are OK to, keep in mind it is a synthetic max workload) what is your ambient temperature during the test? Since a cooler is much less effective how closer the gab between ambient temperature and cooler temperature is.

 

And if you want to test for a 24/7 setting, ramp up your GPU (especial if you have a GPU with an open cooler design) a bit, because it can effect the temperature in your case.

I'm not able to check my ambient temperature atm, but it's very cool since I'm in a downstairs room. I'm running a open cooler design GPU the MSI R9 280X Twin Frozn, does really make a big impact or is it negligible? 

[Narnash]

34 minutes ago, starcoaster said:

jesus you really did loose out. I can run 4.8 mostly stable at 1.4. not as high as it could be but it's not bad

might also be due to the fact it's haswell 1.0? idk, the temps were higher on pre-devil's canyon haswell

yeah the haswll pre-devil's canyon lack a bit in OC potential my 4670K had stability issues on > 4,6 GHz 1,375v  with 1,425V it run kinda stable on 4,7GHz (under synthetic test it might crash after >2,5h testing under high gamig workload no issues there)

 

 

blazerclaw try to play with your RAM (a bit higher timings) to eliminate the chance that this cause some stability problems and like a said maybe there are some MB settings that are ON on default that can cause stability issues, too

 

[Narnash]

56 minutes ago, blazerclaw said:

I'm not able to check my ambient temperature atm, but it's very cool since I'm in a downstairs room. I'm running a open cooler design GPU the MSI R9 280X Twin Frozn, does really make a big impact or is it negligible? 

it depends on your region (high temperatures in the summer?) if you have a GPU that spread heat in your case it CAN damage badly, espacial if you go with a high OC'd profile from winter season into higher temperatures. In a case with much room in it and a decent airflow with lower ambient temperatures the effect is insignificant.