The Reality of Haswell Overclocking - Results Poll

Page 5 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

rtsurfer

Senior member
Oct 14, 2013
733
15
76
EDIT: Oh, and Linx is simply a GUI for Linpack libraries, you can even replace them with newest if you like to do so.

What is key in Linx/Linpack runs, is making sure that processes have proper affinity on HT enabled systems. You don't really need to disable anything in the BIOS the following is needed:

1) Linx: Before starting run, in task manager, set linx.exe process affinity to 1,3,5,7 cpus in the list. Linx will start linpack binaries and they will inherit affinity set
2) manual linpack run: you need to modify provided bat file and need to use trickery with affinity. START /AFFINITY 0x55 linpack_xeon64.exe blablabla is what you need to restrict it real CPU cores (0x55 for quad or 0x555 for hexa).

4770K on 4.5G clocks should hit 230GFlops in latest Linx or manual (large enough set) linpack run

That is what I used to believe but the GFLOPS difference & Temps difference between the two has lead me to believe that they are somehow different.

Maybe I need to tinker with the problem size in Linx to get the same results as the Intel one.

You can run your own tests to see the difference if you do not believe me.

Also, I am gonna do rerun in an unrealistic but not thermal limited condition later today to see if I can get Higher GFLOPS if I am not hitting 100 C.

Wow, I'm surprised it stays stable at those temperatures. You actually have a pretty cherry chip, because it's typically voltage limited rather than thermally limited for me currently, which is unfortunate. I can do many things to improve thermals, but nothing to improve the voltage situation. I'd be curious if it started throttling though, My understanding is that tjmax is 100C and it starts throttling at 95C, you peaked 99C... that's... high.

You really should delid, your temps would be significantly better and you may even be able to move up your overclock. There's no real reason not to, unless you've bought the enthusiast warranty and are concerned about physically modifying the CPU. Direct-to-die is an option, but a hassle, so you can do the same as I did and lap the IHS and put it back on top so you don't have to modify the socket retention mechanism. With an H100i, you should see a pretty significant temperature drop.

That makes the two of us.
I was suprised too that I didnt reboot at those temps. Usually my CPU at 4.6 Ghz reboots no matter what volts when temps go above 85 C. Maybe it has to do with the leakage at higher temps & high volts. Or maybe I am not truely stable at 4.6 .

I am planning of getting a custom loop to take care of my cooling down the road. Maybe. First I have to sell my 2 months old H100i. Anybody want one..??

Also, I wouldnt dare to run my CPU at such high temps if I didnt have the Intel OCers warranty. But that doesnt deter me from delidding, fear of damaging the CPU does.
 
Last edited:

PliotronX

Diamond Member
Oct 17, 1999
8,883
107
106
I gather from the discussion, the results of the poll will be muddled because some of the respondents are using Linpack, some are using Prime95, and some are like "f*** it, it runs BF4." The reality is that these clock results will be different depending on how stable is defined.
 

JoeRambo

Golden Member
Jun 13, 2013
1,814
2,105
136
Maybe I need to tinker with the problem size in Linx to get the same results as the Intel one.

You can run your own tests to see the difference if you do not believe me.

If you go to LinX-0.6.5\64-bit directory, you will see Linpack executable. In the past i even updated them. So as long as Linpack version is the same and problem sizing the same as well, you will get same results. Linx is just convenient for affinity setting on HT enabled systems.
 

Tweakin

Platinum Member
Feb 7, 2000
2,532
0
71
I gather from the discussion, the results of the poll will be muddled because some of the respondents are using Linpack, some are using Prime95, and some are like "f*** it, it runs BF4." The reality is that these clock results will be different depending on how stable is defined.

And to think I thought I had made it clear by stating "Please add screen shots if you post with your 24/7 settings and thermals (air/water) to validate please.

IBT voltage with CPU-z and AIDA64 Thermals"
 

Termie

Diamond Member
Aug 17, 2005
7,949
48
91
www.techbuyersguru.com
And to think I thought I had made it clear by stating "Please add screen shots if you post with your 24/7 settings and thermals (air/water) to validate please.

IBT voltage with CPU-z and AIDA64 Thermals"

It was a noble effort, but this is why we still need professional reviewers who can standardize the method of test.

I got shot down despite posting screenshots because I'd run IBT on standard using 8 threads, and not maximum using 4 threads. And frankly, that was an incorrect criticism - choosing four threads isn't sufficient to lock it to four threads on an i7.

So, the moral of the story is that it's very hard to get comparable numbers.
 

Tweakin

Platinum Member
Feb 7, 2000
2,532
0
71
...It was a noble effort, but this is why we still need professional reviewers who can standardize the method of test...

Thanks...I should have used core voltage as the reference...that might have added some reliable metrics to it!
 

PliotronX

Diamond Member
Oct 17, 1999
8,883
107
106
And to think I thought I had made it clear by stating "Please add screen shots if you post with your 24/7 settings and thermals (air/water) to validate please.

IBT voltage with CPU-z and AIDA64 Thermals"
Unfortunately yeah that's the real reality, different strokes of reliability for different folks. I think it'd have to be phrased as a yes/no poll, set the bar at like 4.4GHz and ask if it'll do 10x Linpack loops at x voltage. The results would look a lot different to these
 

rtsurfer

Senior member
Oct 14, 2013
733
15
76
Unfortunately yeah that's the real reality, different strokes of reliability for different folks. I think it'd have to be phrased as a yes/no poll, set the bar at like 4.4GHz and ask if it'll do 10x Linpack loops at x voltage. The results would look a lot different to these

Well its not like we can stop people from lying.

They would simply click yes.
 

Tristor

Senior member
Jul 25, 2007
314
0
71
Delid..?
Custom loop..?

From his sig:

GA-Z87X-UD5H /4770K delidded / F8 modded BIOS / Kingston HyperX Genesis RAM
CL SM8 r-atx w/pedestals / custom water / acrylic pipe / light panel
SeaSonic Platinum 1000 all custom wiring and sleeving
2xPowerColor AX7950 2DHV4 w/EK blocks

So yes, he delidded and has a custom water setup.

I'm assuming he's doing direct-to-die, which makes me less concerned about my core temperature delta, since he's also got a huge delta (59C to 72C is a HUGE leap). I've been stressing over my mount, possible misapplication of CLU, etc that I plan to check later this week when I mod my case to accept 3 more 120mm fans and to add additional vibration dampening. Seems I may not be as bad off as I thought.
 

Tweakin

Platinum Member
Feb 7, 2000
2,532
0
71
All of this has me playing with my memory. Setting an XMP profile really kicks up my core voltage requirement, so I'm playing with 4.4Ghz and 1600@9.9.9 1T with the same voltage!!!

Maybe I can pull off 4.5 as thermals have not been an issue yet. 20 runs with IBT and thermals in the 70's. Hopefully more to follow.

I'm just plain addicted to this...

edit: 4.5 is just not stable for me using IBT or AIDA64 within my acceptable voltage range. "SIGH"
 
Last edited:

rtsurfer

Senior member
Oct 14, 2013
733
15
76
From his sig:

GA-Z87X-UD5H /4770K delidded / F8 modded BIOS / Kingston HyperX Genesis RAM
CL SM8 r-atx w/pedestals / custom water / acrylic pipe / light panel
SeaSonic Platinum 1000 all custom wiring and sleeving
2xPowerColor AX7950 2DHV4 w/EK blocks

So yes, he delidded and has a custom water setup.

I'm assuming he's doing direct-to-die, which makes me less concerned about my core temperature delta, since he's also got a huge delta (59C to 72C is a HUGE leap). I've been stressing over my mount, possible misapplication of CLU, etc that I plan to check later this week when I mod my case to accept 3 more 120mm fans and to add additional vibration dampening. Seems I may not be as bad off as I thought.

Sorry, I didn't see his sig as I was on my phone.
Looks like he has a pretty good chip.

I suggest if you want to push your chip further, you go custom loop. No matter how much you fans you add you will still hit a wall and the only way to go over that wall is going to be a custom loop.
 

Tristor

Senior member
Jul 25, 2007
314
0
71
Sorry, I didn't see his sig as I was on my phone.
Looks like he has a pretty good chip.

I suggest if you want to push your chip further, you go custom loop. No matter how much you fans you add you will still hit a wall and the only way to go over that wall is going to be a custom loop.

I want to really bad, and I will in the next build, but it's just not in the cards right now. This is just a stop-over until I can build a X99 monster at the beginning of next year, at which point I'll either be getting a Caselabs/MountainMods chassis or fabbing my own and doing a custom loop. In the mean-time, I'm trying to see what sort of wizardry I can work using AIO coolers.
 

boondocks

Member
Mar 24, 2011
84
2
71
From his sig:

GA-Z87X-UD5H /4770K delidded / F8 modded BIOS / Kingston HyperX Genesis RAM
CL SM8 r-atx w/pedestals / custom water / acrylic pipe / light panel
SeaSonic Platinum 1000 all custom wiring and sleeving
2xPowerColor AX7950 2DHV4 w/EK blocks

So yes, he delidded and has a custom water setup.

I'm assuming he's doing direct-to-die, which makes me less concerned about my core temperature delta, since he's also got a huge delta (59C to 72C is a HUGE leap). I've been stressing over my mount, possible misapplication of CLU, etc that I plan to check later this week when I mod my case to accept 3 more 120mm fans and to add additional vibration dampening. Seems I may not be as bad off as I thought.
Not direct to die. I just removed the IHS adhesive and applied CLP to the cpu.
I'm sure it would have benefited if I had lapped the IHS: I know it made quite the difference (less core temp variation) with my delidded 3770K.
 

Tristor

Senior member
Jul 25, 2007
314
0
71
Yeah, I highly recommend lapping the IHS. It's very easy to do post-delid and I can attest that the IHS these days aren't even close to flat (they never really were but they seem to get worse).
 

RadiclDreamer

Diamond Member
Aug 8, 2004
8,622
40
91
Wow, how times have changed. I remember the days where a few % increase were fantastic. Now days people arent happy unless they are getting 1000mhz increases.
 

podspi

Golden Member
Jan 11, 2011
1,982
102
106
Wow, how times have changed. I remember the days where a few % increase were fantastic. Now days people arent happy unless they are getting 1000mhz increases.

Part of that is that we're no longer using cheap slower chips and messing with them to go faster. We're now buying chips that are geared towards using for overclocking. If people pay more for overclocking, they (quite reasonably) will expect that part to actually be able to overclock.

And I realize that is not the case (Intel and AMD do not guarantee any OCs), but the psychology behind people's complaints makes sense to me. This is why I have never purchased an unlocked processor. I refuse to pay extra for something the manufacturer won't guarantee me (at least so far...)
 

rtsurfer

Senior member
Oct 14, 2013
733
15
76
Part of that is that we're no longer using cheap slower chips and messing with them to go faster. We're now buying chips that are geared towards using for overclocking. If people pay more for overclocking, they (quite reasonably) will expect that part to actually be able to overclock.

And I realize that is not the case (Intel and AMD do not guarantee any OCs), but the psychology behind people's complaints makes sense to me. This is why I have never purchased an unlocked processor. I refuse to pay extra for something the manufacturer won't guarantee me (at least so far...)

AMD did do a gauranteed OC chip recently.
Its called Fx 9590 & 9370.
AMD's move wasn't widely applaused because their processors are far behind Intel's in IPC.

So even a 5 Ghz Piledriver wasn't good enough to keep up with Intel. But they did deliever a gaurenteed OC chip.
 

OCGuy

Lifer
Jul 12, 2000
27,227
36
91
Throwing some voltage at that chip aren't you...

Nothing wrong with that voltage with his setup.....unless there is something I have missed?

Not the greatest chip out there, obviously the voltage wall isn't golden. But I still wish people would post BIOS screenies so we could see all settings, including the cache, voltage type (adaptive v constant), etc. Auto settings?
 
Last edited:
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |