Intel Overclocking Club

[4Ryan6]

Does this mean I have a good chip then?

Yes you do!

Cooling has everything to do with how far you can safely overclock your CPU my 3770K with hyper threading disabled is comparable to your 3570K, but the cooling I am running allows much further overclocks and tolerable temperatures.

You need to learn where your limits are and that does not mean you don't have a great CPU, it just means you've taken it as far as the cooling you have will allow you to safely go.

My CPU has never been delidded, I had no need to do so as my cooling is more than capable of supporting the 5ghz overclock and the voltage it takes to get there, the test below was run at an 8c water temperature which was 15c below my ambient room temperature.

 

[neograndizer]

Okay, call me lazy, but here's a repost of my OC from a year ago. Still running the same setup and never touched it since. Well, besides replacing the SSD and a couple HDDs in the process.

P95 blend still running close to 9 hours

P95 blend stopped at 9 hours

IBT 10 runs Max settings

This was original post: http://www.tomshardware.com/forum/283638-29-3570k-thoughts

 

[AdmiralDonut]

CPU: i5 3570K
Motherboard: MSI Z77A-G45 Gaming
CPU Voltage: 1.1500v set in bios.
CPU Bus Speed/Multiplier: 101Mhz/ x45 (MSI Control misreads BCLK as 101 for some reason)
Clock Speed: 4498.95Mhz
RAM: 16GB Corsair Vengeance LP 1600Mhz
Cooling: Cooler Master Seidon 120V
OS: Windows 7

My latest run now that I've moved the cooler to the front of the case for better airflow:

Now to see about a 100% stable 4.6GHz.

 

[AdmiralDonut]

CPU: i5 3570K
Motherboard: MSI Z77A-G45 Gaming
CPU Voltage: 1.1900v See below for a little confusion on this matter
CPU Bus Speed/Multiplier: 100Mhz/ x46
Clock Speed: 4599.1Mhz
RAM: 16GB Corsair Vengeance LP 1600Mhz
Cooling: Cooler Master Seidon 120V
OS: Windows 7

I have a question here for those able to answer; Why is it that when setting the voltage to 1.1900v in control centre, with Vdroop Offset Control to +100% and Digital Compensation Level set to high, do both CPU-Z, and the live voltage readout in Control Centre both say between 1.1440 and 1.1550 during IBT runs? Is this motherboard just not capable of keeping a sustained voltage? Does this mean that I have actually managed to hit 4.6GHz on just 1.1550v? I am a bit baffled.

 

[AdmiralDonut]

Okay, call me lazy, but here's a repost of my OC from a year ago. Still running the same setup and never touched it since. Well, besides replacing the SSD and a couple HDDs in the process.


This was original post: http://www.tomshardware.com/forum/283638-29-3570k-thoughts

Why did you underclock to 1.7 and 2.6GHz for those last two tests? What does it prove?

 

[neograndizer]

Okay, call me lazy, but here's a repost of my OC from a year ago. Still running the same setup and never touched it since. Well, besides replacing the SSD and a couple HDDs in the process.


This was original post: http://www.tomshardware.com/forum/283638-29-3570k-thoughts

Why did you underclock to 1.7 and 2.6GHz for those last two tests? What does it prove?

Not proving anything as I didn't underclock. That's Intel SpeedStep working the moment I stopped the tests and the system went 'idle' for that brief moment.

 

[AdmiralDonut]

Okay, call me lazy, but here's a repost of my OC from a year ago. Still running the same setup and never touched it since. Well, besides replacing the SSD and a couple HDDs in the process.


This was original post: http://www.tomshardware.com/forum/283638-29-3570k-thoughts

Why did you underclock to 1.7 and 2.6GHz for those last two tests? What does it prove?

Not proving anything as I didn't underclock. That's Intel SpeedStep working the moment I stopped the tests and the system went 'idle' for that brief moment.

Correct me if I'm wrong here, but doesn't CPU-Z still display the max and min multiplier, as it does in every other CPU-Z screenshot and validation webpage? Yours appears to be a fixed multiplier, or is that an artifact of CPU-Z?

 

[neograndizer]

Okay, call me lazy, but here's a repost of my OC from a year ago. Still running the same setup and never touched it since. Well, besides replacing the SSD and a couple HDDs in the process.


This was original post: http://www.tomshardware.com/forum/283638-29-3570k-thoughts

Why did you underclock to 1.7 and 2.6GHz for those last two tests? What does it prove?

Not proving anything as I didn't underclock. That's Intel SpeedStep working the moment I stopped the tests and the system went 'idle' for that brief moment.

Correct me if I'm wrong here, but doesn't CPU-Z still display the max and min multiplier, as it does in every other CPU-Z screenshot and validation webpage? Yours appears to be a fixed multiplier, or is that an artifact of CPU-Z?

Could be because it was an older version of CPU-Z at the time I did it (over a year ago). I noticed it now displays the min/max multiplier with the latest version (but not with the old version I had at that time). The OC was over a year ago but the same one I'm still using today.

 

[AdmiralDonut]

Okay, call me lazy, but here's a repost of my OC from a year ago. Still running the same setup and never touched it since. Well, besides replacing the SSD and a couple HDDs in the process.


This was original post: http://www.tomshardware.com/forum/283638-29-3570k-thoughts

Why did you underclock to 1.7 and 2.6GHz for those last two tests? What does it prove?

Not proving anything as I didn't underclock. That's Intel SpeedStep working the moment I stopped the tests and the system went 'idle' for that brief moment.

Correct me if I'm wrong here, but doesn't CPU-Z still display the max and min multiplier, as it does in every other CPU-Z screenshot and validation webpage? Yours appears to be a fixed multiplier, or is that an artifact of CPU-Z?

Could be because it was an older version of CPU-Z at the time I did it (over a year ago). I noticed it now displays the min/max multiplier with the latest version (but not with the old version I had at that time). The OC was over a year ago but the same one I'm still using today.

I figured that might be the case. I see so many different measurements and pieces of software for doing this with that it gets really confusing :/ I'm new to overclocking, in case you couldn't tell

 

[neograndizer]

I figured that might be the case. I see so many different measurements and pieces of software for doing this with that it gets really confusing :/ I'm new to overclocking, in case you couldn't tell

LOL. Yeah, I remember when I first OC'd my i5-2500k, I killed it in a year's time (memory controller fried due to my RAM voltage being a little high @1.55v and vcore was sitting around 1.2-1.3v). Didn't know any better back then. Why I was a little more cautious OC'ng the IB I have now. I undervolted and increased the timings on the RAM a bit this time around (8-8-8-24-2T @ 1.45v) as well as keeping the vcore below 1.2v. I don't want another fried chip.

 

[AdmiralDonut]

I figured that might be the case. I see so many different measurements and pieces of software for doing this with that it gets really confusing :/ I'm new to overclocking, in case you couldn't tell

LOL. Yeah, I remember when I first OC'd my i5-2500k, I killed it in a year's time (memory controller fried due to my RAM voltage being a little high @1.55v and vcore was sitting around 1.2-1.3v). Didn't know any better back then. Why I was a little more cautious OC'ng the IB I have now. I undervolted and increased the timings on the RAM a bit this time around (8-8-8-24-2T @ 1.45v) as well as keeping the vcore below 1.2v. I don't want another fried chip.

I hear you on that one. I've been insanely careful overclocking this thing, mostly because there isn't a chance in hell I could afford another one for a year or more. Still, I think hitting 4.6GHz at 1.1900 (or 1.1550 if CPU-Z and the live voltage is to believed) is pretty good. Still going to try for a stable 4.7GHz though, since I couldn't get it 100% stable for more than 4 hours on P95 last time.

 

[AdmiralDonut]

Last run of the day, I promise.

As always, my specs are:

CPU: i5 3570K
Motherboard: MSI Z77A-G45 Gaming
CPU Voltage: 1.2400v set in bios.
CPU Bus Speed/Multiplier: 100Mhz/ x47
Clock Speed: 4698.90Mhz
RAM: 16GB Corsair Vengeance LP 1600Mhz
Cooling: Cooler Master Seidon 120V
OS: Windows 7

Successfully completed a couple of runs at these voltages on high stress settings in IBT and she's rock solid, if a little hotter than I'd like. I do appear to have an issue with getting the voltages tight enough though, as according to both CPU-Z and the live voltage feed in Control Center I'm using only 1.2080v at most, but I have to set it to 1.2400. I'm thinking this is down to the motherboard but have no way to check other than ask in here if anyone else has the same setup. Still, at the end of the day I think 4.7GHz@1.2400v is OK and I reckon I'll leave it at that for now


EDIT: Just as an addendum to this post I thought I would add that this 4.7GHz appears to be stable for gaming with temps staying in the low to mid 70's, so I'm fairly satisfied after an hour and a half that it isn't going to blue screen on me.

 

[Borge80]

4.5ghz with only 1.115v?????????????????????????????????????
but still 77 degrees Celsius ????
then you have a golden chip

Not sure how he's even at 77 deg celsius with water cooling and that low voltage. My 4.6ghz at 1.312vcore gets up to the low 70's with prime 95.

P95 is terrible for testing though imo as it induces more power than it would actually use in ideal circumstances at 100% load.
I find AIDA64 is probably the best cpu stressor program. I get mid 60's under 100% load stress using AIDA.

 

[AdmiralDonut]

4.5ghz with only 1.115v?????????????????????????????????????
but still 77 degrees Celsius ????
then you have a golden chip

Not sure how he's even at 77 deg celsius with water cooling and that low voltage. My 4.6ghz at 1.312vcore gets up to the low 70's with prime 95.

P95 is terrible for testing though imo as it induces more power than it would actually use in ideal circumstances at 100% load.
I find AIDA64 is probably the best cpu stressor program. I get mid 60's under 100% load stress using AIDA.

Because it isn't 1.115v, it's 1.1550v. I don't know if that makes much difference, but I though I should correct you both.

 

[mastergup]

even 1.155v is very low for a 3570k at 4.5ghz, and even then the temps are high, I get that temps when I use 1.275v for 4.7ghz.
like set before post a good screenie when you are under load, not when itb is finished but when it is running.

 

[AdmiralDonut]

even 1.155v is very low for a 3570k at 4.5ghz, and even then the temps are high, I get that temps when I use 1.275v for 4.7ghz.
like set before post a good screenie when you are under load, not when itb is finished but when it is running.

Admittedly the temps are too high, but I think that may be something to do with the 23.4-23.8°c ambient in the living room, coupled with the fact that I don't think I did a good job with the thermal paste. I'll do some more testing tomorrow as it's late now and I'll be wanting to hit the sack soon.

EDIT: Also, just a side note here, but you have your i5 cooled by a 240mm cooler, mine is cooled only by a 120mm, and not a particularly expensive one. That will undoubtedly have a fairly big impact on my temps.

 

[Borge80]

/delete/delete/delete

 

[mastergup]

even 1.155v is very low for a 3570k at 4.5ghz, and even then the temps are high, I get that temps when I use 1.275v for 4.7ghz.
like set before post a good screenie when you are under load, not when itb is finished but when it is running.

Admittedly the temps are too high, but I think that may be something to do with the 23.4-23.8°c ambient in the living room, coupled with the fact that I don't think I did a good job with the thermal paste. I'll do some more testing tomorrow as it's late now and I'll be wanting to hit the sack soon.

EDIT: Also, just a side note here, but you have your i5 cooled by a 240mm cooler, mine is cooled only by a 120mm, and not a particularly expensive one. That will undoubtedly have a fairly big impact on my temps.

Yes I do have 240mm cooler but only using fans with 37cfm, and my vcore is much more then yours, so that would make up for the difference.

 

[AdmiralDonut]

even 1.155v is very low for a 3570k at 4.5ghz, and even then the temps are high, I get that temps when I use 1.275v for 4.7ghz.
like set before post a good screenie when you are under load, not when itb is finished but when it is running.

Admittedly the temps are too high, but I think that may be something to do with the 23.4-23.8°c ambient in the living room, coupled with the fact that I don't think I did a good job with the thermal paste. I'll do some more testing tomorrow as it's late now and I'll be wanting to hit the sack soon.

EDIT: Also, just a side note here, but you have your i5 cooled by a 240mm cooler, mine is cooled only by a 120mm, and not a particularly expensive one. That will undoubtedly have a fairly big impact on my temps.

Yes I do have 240mm cooler but only using fans with 37cfm, and my vcore is much more then yours, so that would make up for the difference.

Even with lower CFM fans, you still have double the fin density (if not more) than I do. Also, don't the Haswells produce less heat than the IB's?

 

[mastergup]

even 1.155v is very low for a 3570k at 4.5ghz, and even then the temps are high, I get that temps when I use 1.275v for 4.7ghz.
like set before post a good screenie when you are under load, not when itb is finished but when it is running.

Admittedly the temps are too high, but I think that may be something to do with the 23.4-23.8°c ambient in the living room, coupled with the fact that I don't think I did a good job with the thermal paste. I'll do some more testing tomorrow as it's late now and I'll be wanting to hit the sack soon.

EDIT: Also, just a side note here, but you have your i5 cooled by a 240mm cooler, mine is cooled only by a 120mm, and not a particularly expensive one. That will undoubtedly have a fairly big impact on my temps.

Yes I do have 240mm cooler but only using fans with 37cfm, and my vcore is much more then yours, so that would make up for the difference.

Even with lower CFM fans, you still have double the fin density (if not more) than I do. Also, don't the Haswells produce less heat than the IB's?

No haswell produce more heat then ivy, but I don,t have haswell

 

[AdmiralDonut]

No haswell produce more heat then ivy, but I don,t have haswell

Oh? http://www.tomshardware.co.uk/forum/277215-29-intel-overclocking-club/page-12#14139063

 

[mastergup]

I had haswell, but not any more, in fact I had a good haswell 4670k but not any more

 

[AdmiralDonut]

I had haswell, but not any more, in fact I had a good haswell 4670k but not any more

Why on earth did you downgrade literally your entire system?

 

[mastergup]

I had haswell, but not any more, in fact I had a good haswell 4670k but not any more

Why on earth did you downgrade literally your entire system?

because I sell sometimes computers, so if I can make a good deal I will sell it.
went from a i5 4670k to a i7 2600k, now this 1, had some amd cpu,s .
Also the same with the video cards.
the 1 I have now I upgrade a bit and then sell it again.

 

[AdmiralDonut]

I had haswell, but not any more, in fact I had a good haswell 4670k but not any more

Why on earth did you downgrade literally your entire system?

because I sell sometimes computers, so if I can make a good deal I will sell it.
went from a i5 4670k to a i7 2600k, now this 1, had some amd cpu,s .
Also the same with the video cards.
the 1 I have now I upgrade a bit and then sell it again.

LMAO. That sounds like an awful lot of hassle fella