Haswell overclocking discussion.

[crashtech]

I'd like to give Balla and the other contributors to this thread kudos for helping blaze the trail to successful Haswell OCing. Even though I don't have one (yet), I'm following with interest.

 

[BallaTheFeared]

Uncore is matching my CPU clock... so 4.4 Ghz at the moment. Memory is untouched from stock (10-10-10-27 1600 @ 1.5V). Vin is 1.2V, Vcore is .896V.

Drop your uncore back to 38, set it manually.

Work your core first, otherwise try bumping your uncore voltage but it's a hassle to try to figure out which is causing the problem so focus on one aspect (core) then move to the next (imo).

 

[Eureka]

Drop your uncore back to 38, set it manually.

Work your core first, otherwise try bumping your uncore voltage but it's a hassle to try to figure out which is causing the problem so focus on one aspect (core) then move to the next (imo).

I have no "uncore" setting, but there is a CPU cache ratio. I set it to 38 manually, it still crashes at 4.5 GHz. This is with 1.32V on CPU voltage, 1.9V voltage input, highest load-line calibration.

Is this just the limit of the silicon??

 

[BallaTheFeared]

Try backing down the voltage, to say 1.25v also turn off LLC it's pretty much worthless now with the voltage regulators on the die.

 

[Eureka]

Try backing down the voltage, to say 1.25v also turn off LLC it's pretty much worthless now with the voltage regulators on the die.

Thanks, I think turning off LLC did the trick (I had it at 1.25V before). It cleared normal Intel Burn Test at 4.5 GHz.

It's getting late so I'm going to save tweaking it for tomorrow, but Balla you da man.

 

[BallaTheFeared]

Here to help, hopefully it stays true to that when next you pick up tweaking it!

Be careful of avx if you are using offset, it will increase your voltage needlessly!

 

[Rvenger]

Kind of dismayed how immature the bios's are, and that's from all the vendors.

You can say that again. My Asrock Z87 Extreme 4 is a paperweight at this time.

 

[Rvenger]

I have no "uncore" setting, but there is a CPU cache ratio. I set it to 38 manually, it still crashes at 4.5 GHz. This is with 1.32V on CPU voltage, 1.9V voltage input, highest load-line calibration.

Is this just the limit of the silicon??

Cache ratio should match CPU multiplier. I think Uncore is cache ratio. Some bios' state it differently.

 

[BallaTheFeared]

Cache ratio should match CPU multiplier. I think Uncore is cache ratio. Some bios' state it differently.

Actually the cache is so fast with Haswell it makes almost no difference in performance at low core speeds.

Besides it's counter productive to overclock two or more aspects of the processor at the same time.

Nothing wrong with bringing it up later, but when you're just starting out it adds unneeded complexity.

Edit: Just look at the speed difference between a 5.1GHz i7-2600k, and my chip at 4.7GHz Uncore (cache ratio).

I'm not even sure if it's reading it right, there is just a stupid amount of bandwidth there.

 

[notty22]

It looks like the 4770k has all the highest ram frequency records already. So that improvement mentioned in Anands review is correct.
http://valid.canardpc.com/records.php

Highest RAM Frequency Reached :

• Nº 1 : Nick + John @ HKEPC OC Lab reached 2172.7 MHz with 4096 MB of Team Group Inc. Memory [Details]
  MB : ASRock Z87M OC Formula (Intel Haswell rev 06) - CPU : Intel Core i7 4770K @ 1241.55 MHz
• Nº 2 : Lin222 + Planet + Nickshih @ Computex reached 2145.2 MHz with 4096 MB of Corsair Memory [Details]
  MB : ASRock Z87M OC Formula (Intel Haswell rev 06) - CPU : Intel Core i7 4770K @ 1170.04 MHz
• Nº 3 : G.Skill OC Show @ Computex 2013 reached 2144.6 MHz with 4096 MB of G.Skill Memory [Details]
  MB : Asus MAXIMUS VI IMPACT (Intel Haswell rev 06) - CPU : Intel Core i7 4770K @ 1608.34 MHz
• Nº 4 : G.SKILL COMPUTEX 2013 reached 2141.6 MHz with 4096 MB of G.Skill Memory [Details]
  MB : Asus MAXIMUS VI IMPACT (Intel Haswell rev 06) - CPU : Intel Core i7 4770K @ 1606.09 MHz
• Nº 5 : Andre Yang reached 2105.1 MHz with 4096 MB of Avexir Memory [Details]
  MB : Asus MAXIMUS VI IMPACT (Intel Haswell rev 06) - CPU : Intel Core i7 4770K @ 2152.85 MHz

 

[NFarnzy]

I have my 4770k @ 4.4 and 1.26v, I would like to get it a bit more lower on v core, I have it set manual vcore.. I think that adaptive would be better but I cant get it stable on it. Is there a setting that I can try ?

 

[Eureka]

I have my 4770k @ 4.4 and 1.26v, I would like to get it a bit more lower on v core, I have it set manual vcore.. I think that adaptive would be better but I cant get it stable on it. Is there a setting that I can try ?

Make sure current and power settings are set high. Set the cache/uncore speed at 3.8. Play with LLC. That's how I got mine stable at 4.5.

 

[SPBHM]

Edit: Just look at the speed difference between a 5.1GHz i7-2600k, and my chip at 4.7GHz Uncore (cache ratio).

http://farm8.staticflickr.com/7454/9080057869_74beb2fac0_o.png

I'm not even sure if it's reading it right, there is just a stupid amount of bandwidth there.

this difference is software related, you need to use the same version of Aida.

 

[AkumaX]

Dark Knight 2 with two static pressure corsair fans in push/pull.


Yeah I saw that

Can you use http://www.mediafire.com/?zp28fy8ru48x7r6 (cpuz 1.64.3) so I can see your actual voltage, unless that is what you set in bios.

What were your temps at stock? And is your board running 3.4 stock or 3.8?

nice temps on that then!

i d/led the new cpu-z 1.64.3. vcore bounces between 1.12v and 1.21v.

did another load test, top temp is now 92-94°C. idle's around 34-36°C.

 

[BallaTheFeared]

If it's bouncing like that it's because you're using offset voltage and running an avx instruction.

If you use fixed while doing your testing it won't bounce, even 1.12v could be excessive for 4GHz. You should try toning that back, while using fixed voltage. Also make sure you're setting your Uncore to x38 as to not infer any sort of added chance of failure from different aspects of the overclock you aren't aware of.

Eureka has a ASRock board and his is automatically keeping his uncore (cache ratio) in a 1:1 ratio with his core.

Also something else to consider, if you aren't going to delid is using a more practical method of testing for your needs and cooling solution. Haswell will throttle so there is no risk of instability in non avx programs if you're getting too hot, assuming of course you don't currently plan to run avx based programs in your daily tasks.

this difference is software related, you need to use the same version of Aida.

Makes sense!

I have my 4770k @ 4.4 and 1.26v, I would like to get it a bit more lower on v core, I have it set manual vcore.. I think that adaptive would be better but I cant get it stable on it. Is there a setting that I can try ?

The difference between adaptive and fixed doesn't occur under actual load, only if your chip is downclocking to save power will it have any effect. Also the difference in power is pretty negligible. Adding LLC really only adds voltage, the chips don't really suffer from vdroop so using LLC is just basically the same thing as adding more vcore.

If you haven't already I would set your uncore back to x38 and then while using fixed start trying to dial back the voltage, if you can't get it any lower with the uncore turned back and your ram at 1600/1866 than I don't think you have much choice. You're at the mercy of your silicon, and I haven't used a setting yet that can help with that aspect, some of them do help with memory stability but that's a horse of a different color.

 

[24601]

The reason people are seeing "more stable" at lower uncore is because Prime95 is bottlenecked by cache speed when using AVX instructions (The whole reason Haswell has a huge cache throughput to start with).

Not overclocking the cache with the core means you are just gimping Prime95 performance, which is why you have lower temps. You get the same effect if you just let the chip thermally throttle.

It's not magic folks.

Read the Prime95 forum from the developers to get an in-depth technical analysis on this whole issue.
They have a nice thread on this.

 

[BallaTheFeared]

The reason people are seeing "more stable" at lower uncore is because Prime95 is bottlenecked by cache speed when using AVX instructions (The whole reason Haswell has a huge cache throughput to start with).

Huh? Nobody is saying don't overclock the uncore, though a case can easily be made to just ignore it.

Can you link this thread you're referencing, I couldn't find it with google.

Not overclocking the cache with the core means you are just gimping Prime95 performance, which is why you have lower temps. You get the same effect if you just let the chip thermally throttle.

Who cares about the performance of prime95? Lower temps for me were due to cooler ambient and have nothing to do with cache ratio or bottlenecking AVX from what I have seen. Same effect as what? The only thing shown in here with prime95 was 4.8GHz core 4.7GHz UNcore and the temps were no different than AIDA64.

It's not magic folks.

Evidently Prime95 is since it's so fast other linpack avx code doesn't suffer the same amazing performance crippling effect. I can't even begin imagine the amount of heat and power Prime95 would use if unthering it's performance was from the uncore bottleneck, considering it isn't any hotter than ADIA64 or IBT even though both run AVX code paths!

Read the Prime95 forum from the developers to get an in-depth technical analysis on this whole issue.
They have a nice thread on this.

I'd love to read it, though I'm not sure the relevance. I couldn't find it with google though, could you link it?


Here is what I have:

4.8GHz core 3.8GHz Uncore - Hottest core 88C - Fastest time 129 GFLOPs

4.8GHz core 4.6GHz Uncore - Hottest core 88C - Fastest time 130 GFLOPs

 

[24601]

Huh? Nobody is saying don't overclock the uncore, though a case can easily be made to just ignore it.

Can you link this thread you're referencing, I couldn't find it with google.



Who cares about the performance of prime95? Lower temps for me were due to cooler ambient and have nothing to do with cache ratio or bottlenecking AVX from what I have seen. Same effect as what? The only thing shown in here with prime95 was 4.8GHz core 4.7GHz UNcore and the temps were no different than AIDA64.



Evidently Prime95 is since it's so fast other linpack avx code doesn't suffer the same amazing performance crippling effect. I can't even begin imagine the amount of heat and power Prime95 would use if unthering it's performance was from the uncore bottleneck, considering it isn't any hotter than ADIA64 or IBT even though both run AVX code paths!



I'd love to read it, though I'm not sure the relevance. I couldn't find it with google though, could you link it?


Here is what I have:

4.8GHz core 3.8GHz Uncore - Hottest core 88C - Fastest time 129 GFLOPs

4.8GHz core 4.6GHz Uncore - Hottest core 88C - Fastest time 130 GFLOPs

Here is the gist of it for small length ones (specifically the ones that don't directly benefit from AVX2's increased bit length processing ability)

+40% on legacy (SSE2) code due entirely to increased bandwidth.

http://www.mersenneforum.org/showpost.php?p=343718&postcount=131

If you make it exactly 256-bit obviously you get the hilarious huge yields that a purely synthetic (Non-dependent) benchmark would give you.

Thats a huge increase purely from increased bandwidth.

 

[BallaTheFeared]

Here is the gist of it for small length ones (specifically the ones that don't directly benefit from AVX2's increased bit length processing ability)

+40% on legacy (SSE2) code due entirely to increased bandwidth.

http://www.mersenneforum.org/showpost.php?p=343718&postcount=131

If you make it exactly 256-bit obviously you get the hilarious huge yields that a purely synthetic (Non-dependent) benchmark would give you.

Thats a huge increase purely from increased bandwidth.

I'm not sure your point as it pertains to this thread though. Nobody is suggesting not to overclock your uncore, only to decouple it from your core clock testing at first to get an idea of where your core is stable before you move on to the next variable (uncore or ram).

Prime95 isn't producing any less heat than the other avx driven synthetic stress tests.

Though it's testing is very limited and most people are advising against using it until there is an update. Past that there are tests like AIDA64 which are running more instructions than just avx based, sure it won't constantly warm your chip as well. However are you stress testing your cpu or are you stress testing your cooling solution I guess would be my question?

 

[BallaTheFeared]

Maybe I'm too tired, but I'm not even really sure what we're discussing!

Anyways, I'm not even sure what to think. It's pretty clear the IBT linpack isn't hitting Haswell the way it could be, neither is anything else really. So what I'm guessing is that you are saying these stress tests we're using which were designed for older systems aren't optimized for Haswell, or perhaps even older uarchs and that we're kidding ourselves?

At least that's what I'm thinking, I ran the latest Intel linpack.. Jesus H Christ.

Stock: (Kept opening the file to make sure it was doing something, ops)

=================== Timing linear equation system solver ===================

Size   LDA    Align. Time(s)    GFlops   Residual     Residual(norm) Check
1000   1000   4      0.026      26.0931  1.002198e-012 3.417754e-002   pass
1000   1000   4      0.008      83.1892  1.002198e-012 3.417754e-002   pass
1000   1000   4      0.008      85.3023  1.002198e-012 3.417754e-002   pass
1000   1000   4      0.008      86.6677  1.002198e-012 3.417754e-002   pass
2000   2000   4      0.048      112.1212 4.327261e-012 3.764187e-002   pass
2000   2000   4      0.050      106.2897 4.327261e-012 3.764187e-002   pass
2000   2000   4      0.061      87.4023  4.327261e-012 3.764187e-002   pass
2000   2000   4      0.049      109.7069 4.327261e-012 3.764187e-002   pass
3000   3000   4      0.140      128.7982 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.176      102.2257 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.140      128.6445 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.140      129.1520 9.653528e-012 3.717342e-002   pass
4000   4000   4      0.314      135.8601 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.313      136.2905 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.312      136.9157 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.311      137.1751 1.565487e-011 3.412126e-002   pass
5000   5000   4      0.590      141.2435 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.592      140.8687 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.590      141.3670 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.592      140.9005 2.512907e-011 3.504046e-002   pass
10000  10000  4      4.280      155.7971 1.012627e-010 3.570624e-002   pass
10000  10000  4      4.275      155.9875 1.012627e-010 3.570624e-002   pass
15000  15000  4      15.932     141.2575 1.979849e-010 3.118294e-002   pass
15000  15000  4      16.082     139.9325 1.979849e-010 3.118294e-002   pass
20000  20000  4      32.094     166.2039 3.564028e-010 3.154945e-002   pass
20000  20000  4      32.116     166.0878 3.564028e-010 3.154945e-002   pass
25000  25000  4      63.792     163.3114 6.220813e-010 3.537553e-002   pass
25000  25000  4      64.971     160.3481 6.220813e-010 3.537553e-002   pass
30000  30000  4      117.158    153.6544 9.270371e-010 3.654390e-002   pass

4.8GHz 3.8GHz Uncore - shut my motherboard down, the heat was insane. There is no way my cpu cooler can cope with this...

=================== Timing linear equation system solver ===================

Size   LDA    Align. Time(s)    GFlops   Residual     Residual(norm) Check
1000   1000   4      0.027      24.4817  1.002198e-012 3.417754e-002   pass
1000   1000   4      0.006      116.9661 1.002198e-012 3.417754e-002   pass
1000   1000   4      0.006      113.6987 1.002198e-012 3.417754e-002   pass
1000   1000   4      0.006      115.8336 1.002198e-012 3.417754e-002   pass
2000   2000   4      0.035      150.8695 4.327261e-012 3.764187e-002   pass
2000   2000   4      0.035      151.3860 4.327261e-012 3.764187e-002   pass
2000   2000   4      0.034      155.2040 4.327261e-012 3.764187e-002   pass
2000   2000   4      0.036      149.4019 4.327261e-012 3.764187e-002   pass
3000   3000   4      0.104      173.0388 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.107      167.7035 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.104      173.6963 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.103      174.1454 9.653528e-012 3.717342e-002   pass
4000   4000   4      0.229      186.4405 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.231      184.9304 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.229      186.1771 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.229      186.1232 1.565487e-011 3.412126e-002   pass
5000   5000   4      0.433      192.4411 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.434      192.2064 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.433      192.6871 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.434      192.0620 2.512907e-011 3.504046e-002   pass
10000  10000  4      3.169      210.4104 2.977802e-008 1.050003e+001   FAIL
10000  10000  4      3.158      211.1710 9.805032e-011 3.457353e-002   pass

211 flipping GFLOPs? Are you kidding me?!?! This is with it throttling and failing!

4.6GHz core 4.6GHz Uncore

=================== Timing linear equation system solver ===================

Size   LDA    Align. Time(s)    GFlops   Residual     Residual(norm) Check
1000   1000   4      0.032      21.0860  1.002198e-012 3.417754e-002   pass
1000   1000   4      0.006      112.8623 1.002198e-012 3.417754e-002   pass
1000   1000   4      0.006      114.3003 1.002198e-012 3.417754e-002   pass
1000   1000   4      0.006      114.0276 1.002198e-012 3.417754e-002   pass
2000   2000   4      0.042      128.4695 4.327261e-012 3.764187e-002   pass
2000   2000   4      0.036      149.8790 4.327261e-012 3.764187e-002   pass
2000   2000   4      0.036      148.4749 4.327261e-012 3.764187e-002   pass
2000   2000   4      0.050      106.6145 4.327261e-012 3.764187e-002   pass
3000   3000   4      0.107      167.7921 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.108      167.2586 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.107      167.7909 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.109      165.8319 9.653528e-012 3.717342e-002   pass
4000   4000   4      0.236      181.2239 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.235      181.8316 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.238      179.4995 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.237      179.8140 1.565487e-011 3.412126e-002   pass
5000   5000   4      0.446      186.8200 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.445      187.3511 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.445      187.2766 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.444      187.6106 2.512907e-011 3.504046e-002   pass
10000  10000  4      3.313      201.2619 1.012627e-010 3.570624e-002   pass
10000  10000  4      3.233      206.2836 1.012627e-010 3.570624e-002   pass
15000  15000  4      11.923     188.7423 1.979849e-010 3.118294e-002   pass
15000  15000  4      11.838     190.1115 1.979849e-010 3.118294e-002   pass
20000  20000  4      24.219     220.2491 3.564028e-010 3.154945e-002   pass

220 GFLOPS D: D: D: D: D:

Chip hit 98C and I closed it, I'm not that crazy!


I don't even know what to think. On the one hand, clearly old tools are ineffectual (all of them), on the other hand what in the heck do I need to test this for? I had a problem with the old tests, they weren't representative of actual loads I used so they made little sense outside of some people wanting to tout their cooling solution... But this?

lolz...

Anyways I'm going to bed, maybe I'll dream happy thoughts where I wake up and every game is optimized like this and I can downclock to 3GHz and get more performance than I was getting at 5.3GHz three hours ago.

 

[24601]

Maybe I'm too tired, but I'm not even really sure what we're discussing!

Anyways, I'm not even sure what to think. It's pretty clear the IBT linpack isn't hitting Haswell the way it could be, neither is anything else really. So what I'm guessing is that you are saying these stress tests we're using which were designed for older systems aren't optimized for Haswell, or perhaps even older uarchs and that we're kidding ourselves?

At least that's what I'm thinking, I ran the latest Intel linpack.. Jesus H Christ.

Stock: (Kept opening the file to make sure it was doing something, ops)

=================== Timing linear equation system solver ===================

Size   LDA    Align. Time(s)    GFlops   Residual     Residual(norm) Check
1000   1000   4      0.026      26.0931  1.002198e-012 3.417754e-002   pass
1000   1000   4      0.008      83.1892  1.002198e-012 3.417754e-002   pass
1000   1000   4      0.008      85.3023  1.002198e-012 3.417754e-002   pass
1000   1000   4      0.008      86.6677  1.002198e-012 3.417754e-002   pass
2000   2000   4      0.048      112.1212 4.327261e-012 3.764187e-002   pass
2000   2000   4      0.050      106.2897 4.327261e-012 3.764187e-002   pass
2000   2000   4      0.061      87.4023  4.327261e-012 3.764187e-002   pass
2000   2000   4      0.049      109.7069 4.327261e-012 3.764187e-002   pass
3000   3000   4      0.140      128.7982 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.176      102.2257 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.140      128.6445 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.140      129.1520 9.653528e-012 3.717342e-002   pass
4000   4000   4      0.314      135.8601 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.313      136.2905 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.312      136.9157 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.311      137.1751 1.565487e-011 3.412126e-002   pass
5000   5000   4      0.590      141.2435 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.592      140.8687 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.590      141.3670 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.592      140.9005 2.512907e-011 3.504046e-002   pass
10000  10000  4      4.280      155.7971 1.012627e-010 3.570624e-002   pass
10000  10000  4      4.275      155.9875 1.012627e-010 3.570624e-002   pass
15000  15000  4      15.932     141.2575 1.979849e-010 3.118294e-002   pass
15000  15000  4      16.082     139.9325 1.979849e-010 3.118294e-002   pass
20000  20000  4      32.094     166.2039 3.564028e-010 3.154945e-002   pass
20000  20000  4      32.116     166.0878 3.564028e-010 3.154945e-002   pass
25000  25000  4      63.792     163.3114 6.220813e-010 3.537553e-002   pass
25000  25000  4      64.971     160.3481 6.220813e-010 3.537553e-002   pass
30000  30000  4      117.158    153.6544 9.270371e-010 3.654390e-002   pass

4.8GHz 3.8GHz Uncore - shut my motherboard down, the heat was insane. There is no way my cpu cooler can cope with this...

=================== Timing linear equation system solver ===================

Size   LDA    Align. Time(s)    GFlops   Residual     Residual(norm) Check
1000   1000   4      0.027      24.4817  1.002198e-012 3.417754e-002   pass
1000   1000   4      0.006      116.9661 1.002198e-012 3.417754e-002   pass
1000   1000   4      0.006      113.6987 1.002198e-012 3.417754e-002   pass
1000   1000   4      0.006      115.8336 1.002198e-012 3.417754e-002   pass
2000   2000   4      0.035      150.8695 4.327261e-012 3.764187e-002   pass
2000   2000   4      0.035      151.3860 4.327261e-012 3.764187e-002   pass
2000   2000   4      0.034      155.2040 4.327261e-012 3.764187e-002   pass
2000   2000   4      0.036      149.4019 4.327261e-012 3.764187e-002   pass
3000   3000   4      0.104      173.0388 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.107      167.7035 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.104      173.6963 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.103      174.1454 9.653528e-012 3.717342e-002   pass
4000   4000   4      0.229      186.4405 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.231      184.9304 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.229      186.1771 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.229      186.1232 1.565487e-011 3.412126e-002   pass
5000   5000   4      0.433      192.4411 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.434      192.2064 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.433      192.6871 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.434      192.0620 2.512907e-011 3.504046e-002   pass
10000  10000  4      3.169      210.4104 2.977802e-008 1.050003e+001   FAIL
10000  10000  4      3.158      211.1710 9.805032e-011 3.457353e-002   pass

211 flipping GFLOPs? Are you kidding me?!?! This is with it throttling and failing!

4.6GHz core 4.6GHz Uncore

=================== Timing linear equation system solver ===================

Size   LDA    Align. Time(s)    GFlops   Residual     Residual(norm) Check
1000   1000   4      0.032      21.0860  1.002198e-012 3.417754e-002   pass
1000   1000   4      0.006      112.8623 1.002198e-012 3.417754e-002   pass
1000   1000   4      0.006      114.3003 1.002198e-012 3.417754e-002   pass
1000   1000   4      0.006      114.0276 1.002198e-012 3.417754e-002   pass
2000   2000   4      0.042      128.4695 4.327261e-012 3.764187e-002   pass
2000   2000   4      0.036      149.8790 4.327261e-012 3.764187e-002   pass
2000   2000   4      0.036      148.4749 4.327261e-012 3.764187e-002   pass
2000   2000   4      0.050      106.6145 4.327261e-012 3.764187e-002   pass
3000   3000   4      0.107      167.7921 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.108      167.2586 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.107      167.7909 9.653528e-012 3.717342e-002   pass
3000   3000   4      0.109      165.8319 9.653528e-012 3.717342e-002   pass
4000   4000   4      0.236      181.2239 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.235      181.8316 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.238      179.4995 1.565487e-011 3.412126e-002   pass
4000   4000   4      0.237      179.8140 1.565487e-011 3.412126e-002   pass
5000   5000   4      0.446      186.8200 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.445      187.3511 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.445      187.2766 2.512907e-011 3.504046e-002   pass
5000   5000   4      0.444      187.6106 2.512907e-011 3.504046e-002   pass
10000  10000  4      3.313      201.2619 1.012627e-010 3.570624e-002   pass
10000  10000  4      3.233      206.2836 1.012627e-010 3.570624e-002   pass
15000  15000  4      11.923     188.7423 1.979849e-010 3.118294e-002   pass
15000  15000  4      11.838     190.1115 1.979849e-010 3.118294e-002   pass
20000  20000  4      24.219     220.2491 3.564028e-010 3.154945e-002   pass

220 GFLOPS D: D: D: D: D:

Chip hit 98C and I closed it, I'm not that crazy!


I don't even know what to think. On the one hand, clearly old tools are ineffectual (all of them), on the other hand what in the heck do I need to test this for? I had a problem with the old tests, they weren't representative of actual loads I used so they made little sense outside of some people wanting to tout their cooling solution... But this?

lolz...

Anyways I'm going to bed, maybe I'll dream happy thoughts where I wake up and every game is optimized like this and I can downclock to 3GHz and get more performance than I was getting at 5.3GHz three hours ago.

I just wanted to make sure people weren't kidding themselves when they pick and choose which "stress tests" they are using by how much >>Cooler<< they can show their processor as being by running an >>Inferior<< stress tester.



The whole point of running stress tests is so that you don't have to run every single possible combination of factors in every single one of your software in every scenario yourself.
If your stress testing application doesn't seem to actually be stressing your hardware, its probably because it's not, especially if there are other stress test applications that stress it more.

 

[Franzi]

Hey guys I have set both the turbo multiplier and the cache ratio to x44 to keep it in sync. My 4770k is stable like that at 1.170v. In the long run I prefer to use the adaptive mode and i'm not sure how to set the min/max settings for the cache ratio when the CPU will use power saving modes.

Am I supposed to set min cache ratio all way down to x8 because 800Mhz is the idle clock speed and max cache ratio to x44 or just leave min at AUTO or both at x44?

And btw, would you consider 4.4Ghz stable at 1.170v a good chip? I could boot into windows 8 at 4.6Ghz with 1.200v but it BSOD'd under full load.

Kind regards

 

[wand3r3r]

Don't forget to check for WHEA errors, they don't (always) cause BSODs so they can be occurring even if you think you are stable. (See sig for simple app to quickly check)

 

[Franzi]

Don't forget to check for WHEA errors, they don't (always) cause BSODs so they can be occurring even if you think you are stable. (See sig for simple app to quickly check)

Thanks, already checked for WHEA errors through the Event Viewer. That tool is nice though

I ran both stress tests and real world tasks so far things are looking good. I'm still unsure about adaptive vs. manual voltage and how to set the cache ratio. x8 min and x44 max would be logically correct when using adaptive mode and C states. I'm trying to find more infos about that.

 

[BullHorn]

Hey guys, I'm new here. Most of the terminology you're using is like a foreign language to me.

I followed one of JJ's Asus guides, changed the Multiplier to 42 and CPU voltage to Adaptive with a 0.150V Turbo Boost - the voltage seemed very unstable during stress testing, ranging from 1V to 1.21V under full load and eventually BSODing.

This could also be related to my cooling (Hyper 212 Evo) which, I think, is not designed for extreme over-clocking. On the other hand, 4.2GHZ is not extreme so...

p.s. the Asus Ai Suite managed to OC the CPU to 4.6GHZ, it must be something I'm doing wrong. I wish there was a way to stop the auto-OC in the middle at like 4.3 or 4.4 but there isn't.