QUOTE(gengstapo @ Jan 1 2014, 09:47 PM)
Glad to hear that you managed to "solve" it.Don't forget to share with us your OC results here.
Haswell, Devil's Canyon, Pentium G3258 OC V1, 4790K/4760K/4770K/4670K/G3258 | LGA1150
|
|
Jan 1 2014, 10:49 PM
Return to original view | Post
#201
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
|
|
|
|
|
|
Jan 4 2014, 06:43 PM
Return to original view | Post
#202
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
QUOTE(meons @ Jan 4 2014, 11:25 AM) rebuild back my system.. early rig was destroy by kebabom psu The BCLK @ 99.98MHz can be fixed to 100 if you disable the CPU Spread Spectrum in BIOS.some quick stress testing .. 50minute cooler via air cooler Prolimatech Supermega dual fan setup asrock z87 fatal1ty professional cheapo ram ... early ram was destroy too it okie or fail which one better adjust multiple @bclk AIDA64 stress test must be set manually, to stress the RAM at maximum amount of capacity. Go to Preference/General/.....Memory Block Size (512 MB, 1 GB, 2 GB etc). By default it is 16 MB only which is not sufficient to stress the Integrated Memory Controller aka IMC built on the chip. |
|
|
Jan 12 2014, 12:13 AM
Return to original view | Post
#203
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
QUOTE(stasio @ Jan 10 2014, 12:57 AM) - Intel i7-4790, i5-4690, i5-4590, i5-4460 (Haswell refresh). - Intel serie 9 chipset (Z97). Z97 is coming in April 2014. This post has been edited by owikh84: Jan 12 2014, 12:13 AM |
|
|
Jan 13 2014, 05:45 PM
Return to original view | Post
#204
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
QUOTE(stasio @ Jan 13 2014, 04:53 PM) Prime95v.29.3 Build 1 Beta (32/64-bit) 1.5% more voltage or temperature? ftp://mersenne.org/gimps/p95v283.win32.zip ftp://mersenne.org/gimps/p95v283.win64.zip ftp://mersenne.org/gimps/p95v283.linux64.tar.gz ftp://mersenne.org/gimps/p95v283.MacOSX.zip |
|
|
Jan 19 2014, 05:50 PM
Return to original view | Post
#205
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
QUOTE(hairy9802 @ Jan 19 2014, 04:10 PM) hi all.. i got i5 4570@3.2ghz using asrock z87 pro3 and gskill ripjawsx 2133 8gb ram.. i know that i cant oc non k haswell but i tried changing the cpu ratio to 33 instead of 32 and i changed blck to 104.2.. i got 3.4ghz and it is stable.. is it ok to do this? is it will damage my proc? sorry noob here.. 1st time overclocking.. thanks 3.4ghz is a mild OC and is not gonna hurt your chip.http://imageshack.com/a/img546/6324/o4aj.png |
|
|
Jan 19 2014, 08:01 PM
Return to original view | Post
#206
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
|
|
|
|
|
|
Jan 26 2014, 07:10 AM
Return to original view | Post
#207
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
|
|
|
Feb 7 2014, 12:53 AM
Return to original view | Post
#208
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
QUOTE(gengstapo @ Jan 26 2014, 09:42 AM) Owh, one more thing Related: http://www.hardocp.com/article/2014/01/29/.../7#.UvO8hLSgh9YI cant fix my Bus Speed to 100 even I put manual value & Spread Spectrum Disable Is that common for Asrock board? QUOTE The Z87 KILLER FATAL1TY also runs its BCLK a bit shy of 100MHz, so you will see some slightly reduced clocks when you check those. What is pointed out above, which is not much, is basically all the "issues" I had with this motherboard. The OS and driver install were near perfect as noted. It worked exactly as it should and never gave me an issue with BIOS changes or hot-swapping data drives. A temporary "fix" is by setting the BCLK one notch higher than 100MHz such as 100.1 or 100.2 etc. My Z87 OCF doesn't have this issue though. |
|
|
Feb 7 2014, 09:03 AM
Return to original view | Post
#209
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
QUOTE(gengstapo @ Feb 7 2014, 08:33 AM) Thanks for the tips but is that possible to set the value by 0.1 increment? I don't have the board so no idea. But it seems like you have fixed it already with that BCLK @ 100.15MHz I cant test right now coz working, might check up later Wow w0ot?! Such nice review almost cannot believe what exactly this board could offers ![]() |
|
|
Feb 10 2014, 06:51 AM
Return to original view | Post
#210
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
QUOTE(gengstapo @ Feb 8 2014, 02:13 PM) owikh84 Yes, I have heard of that too but won't take it as a good direction.Ive read over the net, few forum regarding OC not recommended Prime95 for Haswell as they mention Prime95 gone really beyond the limits They suggesting AIDA64 as it were optimized for Haswell Wht dya think? As for me, what's the point of using a second most stressful app if there's already a better one on top of it. |
|
|
Feb 10 2014, 11:31 PM
Return to original view | Post
#211
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
QUOTE(gengstapo @ Feb 10 2014, 08:32 AM) So how bout IBT? It seems that IBT further extreme compared to Prime95 IBT/LinX generates more heat but consumes less voltage. So can say that IBT/LinX is good for testing cooling limit. But it finds error faster than Prime95.Example, at same setting, my rig fails stressed with Prime at 40 minutes But for IBT, it seems much faster, at 7th round it BSOD Prime95 on the other hand is good for testing vCore but you need to run it longer to find the error. But afaik according to cstkl1 all these can't beat the mighty Linpack 11. |
|
|
Feb 11 2014, 07:25 PM
Return to original view | Post
#212
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
QUOTE(gengstapo @ Feb 11 2014, 02:01 AM) Prime is more into power consumption? vCore is directly proportional power draw so yes.That means it stress the PSU as well? Is your rig passed the Linpack 11? Prime95 will affect the PSU lets say a weak PSU will produce worse coil whine when stress under Prime95. I tried Linpack11 @ 4.5 GHz but my cooling couldn't take the heat Haswell is not my 24/7 rig so I didn't put much effort to find its stability. |
|
|
Feb 11 2014, 09:22 PM
Return to original view | Post
#213
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
QUOTE(gengstapo @ Feb 11 2014, 08:29 PM) Ahh ic, in my situation, when I ran IBT, setup 43x100 it could passed Pls refer to the first post of this thread for BSOD codes & OC guides.But when I increase the multi to 44x100, it crashed/ BSOD, both are same vCore This mean I need to increase the voltage right? Btw, does the cache multi affecting our OC stability?? BSOD codes for overclocking QUOTE 0x101 = increase vcore 0x124 = increase/decrease vcore or QPI/VTT... have to test to see which one it is 0x0A = unstable RAM/IMC, increase QPI first, if that doesn't work increase vcore 0x1E = increase vcore 0x3B = increase vcore 0x3D = increase vcore 0xD1 = QPI/VTT, increase/decrease as necessary, can also be unstable Ram, raise Ram voltage 0x9C = QPI/VTT most likely, but increasing vcore has helped in some instances 0x50 = RAM timings/Frequency or uncore multi unstable, increase RAM voltage or adjust QPI/VTT, or lower uncore if you're higher than 2x 0x109 = Not enough or too Much memory voltage 0x116 = Low IOH (NB) voltage, GPU issue (most common when running multi-GPU/overclocking GPU) 0x7E = Corrupted OS file, possibly from overclocking. Run sfc /scannow and chkdsk /r ASUS recommends cache frequency to stay at 300MHz lower than core clock for stability. QUOTE Running CPU Cache Ratio faster than processor core frequency does not show any performance gains so we recommend staying in sync with the CPU ratio or within 300MHz (lower than CPU frequency). Maximus VI Series OC Guide |
|
|
|
|
|
Feb 21 2014, 07:15 PM
Return to original view | Post
#214
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
QUOTE(raydenex @ Feb 21 2014, 01:25 PM) Might not be the right place, but since it has some relation to the BIOS update to H87/B85 chipset that enables overclocking of K series Haswell; I suggest you to contact ASUS support service since this is mobo specific question.i5-4440 in an Asus H87M-E motherboard with BIOS version 0604 out of the box (dated 30/8/2013). Detected as i5-4440 in BIOS, Windows 7 (My Computer -> Properties), CPU-Z, HWmonitor. Then I flashed the BIOS to the latest one at https://www.asus.com/Motherboards/H87ME/#support to 0904 (dated 17/1/2014). The changes I observed; Still detected as i5-4440 in BIOS, Windows 7. But in CPU-Z and HWmonitor it now shows i5-4670S. Still clocked as 3.1ghz with 3.3ghz turbo like a 4440 and 84w TDP, as 4670s suppose to have 3.8ghz turbo and 65w TDP. Anybody knows what happened here? It could be either BIOS or CPU-Z/HWMonitor issue. QUOTE(DesktopLimousine @ Feb 21 2014, 03:32 PM) what is the common stable voltage for 4770k at 4.5ghz 24/7? Voltage required is depends on chip and it's unique from chip to chip. does it need to run 4c/8t? may i know best batch for this CPU? I need to know as im planning to get 4770k. From what I observed a 4770K @ 4.5GHz around 1.15v is a good chip, bad chip like mine requires 1.25v+. If you don't need 4C/8T then might consider to take the 4570K 4C/4T. There is no best batch it's all depends on luck. QUOTE(g88 @ Feb 21 2014, 04:53 PM) Intel locks overclocking to K-series Haswell chips This post has been edited by owikh84: Feb 21 2014, 07:15 PM |
|
|
Feb 26 2014, 07:34 PM
Return to original view | Post
#215
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
|
|
|
Mar 13 2014, 10:14 PM
Return to original view | Post
#216
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
QUOTE(coolkwc @ Mar 13 2014, 01:07 PM) My 3770K is running without the IHS until now.I am using EK Supreme HF waterblock which requires me to remove the washers in order to make a better contact between the waterblock base and the chip (die). I'm not going to delid my 4770K until I found a good OC chip. |
|
|
Mar 14 2014, 06:50 AM
Return to original view | Post
#217
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
QUOTE(coolkwc @ Mar 14 2014, 02:32 AM) Sad OC result. Don't know whether due to cooler no sitted properly or what. 4.4Ghz with 1.25Vcore CPU already start throttling. Yeah can't really play much with undelidded chip. It's time to give it a move by removing the cap In stock, temp delta between idle and load is too big, which is around 45'C. Load temp during LinX is from 70-85'C, However when i touch my H110 radiator with hand it actually just slightly warm only. Just feel the heat just trap inside the IHS...if really no improvement after i resit the cooler, might as well consider delid the damn thing. Sure you won't regret with the improvement. By the way, 1.25V seems too much for 4.4. A good chip will need only 1.15-1.2V for that OC. |
|
|
Mar 17 2014, 12:52 AM
Return to original view | Post
#218
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
QUOTE(coolkwc @ Mar 16 2014, 09:37 PM) Resit my H110 cooler today, still the same, the temp is very high at load even on stock frequency and voltage. I think you have missed the Ivy Bridge overheat issue. Same thing for Haswell.Stock temp idle 33, load 75-80. Delta over 40'C. Is this normal? The thermal paste used is Cooler Master X1 Extreme Fusion. Back in core2 era, OC is just a simple task for me. For this Haswell I have absolutely no idea for it, it just like a stranger to me. day 3 i OC my CPU, i understand something from the sensor. 1. memory voltage 1.65V will absolutely increase the CPU temp, hence those performance RAM is actully can run at tighter timing and faster clock with lower voltage (ex 1.5V), am i right? Anyway, i really feel disappointing with the temp, although sensor reading is 90'C, but the heat cannot transfer to the cooler, just waste my cooler capability. The overheat issue is caused by the excessive gap between the die and IHS that makes heat dissipation poorer than expected. Blame Intel for that. All you have to do is to void warranty and delid the IHS out of craps then only you will see an 30C drop on temperature. Same as CPU, overvolt the RAM will benefit overclocking & tighter timing. |
|
|
Mar 17 2014, 07:07 PM
Return to original view | Post
#219
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
QUOTE(coolkwc @ Mar 17 2014, 09:21 AM) Actually i have no plan to OC RAM, but well, running my RAM at 'default' 2400Mhz already mean i OC from Haswell point of view. Normal TIM will reduce up to 20C. Liquid metal TIM like Coolaboratory Liquid Pro will give another 10C improvement.Regarding temp drop after delid, -30'C is just an absolutely best case can achieve, i doubt everyone delid can get the same result. I search the forum, mostly drop around 10~ 15'C only, correct me if i'm wrong. I just worry lesser gain when i delid (<10'C drop), not worth if like that. Anyway, yesterday i got frustrated by the result, so i pump 1.4v and clock to 4.7Ghz, can boot and running GRID 2 without issue. Today will try to hit 5Ghz with this voltage. |
|
|
Apr 11 2014, 06:17 AM
Return to original view | Post
#220
|
|
Elite
8,711 posts Joined: Nov 2007 From: Butterworth, PG / Machang, Kelantan |
QUOTE(DesktopLimousine @ Apr 8 2014, 05:10 PM) just want to ask, what is the common voltage for i5 4670k at 4.0ghz? does it need to manually set the Voltage if overclock speed at 4.0ghz? Average is 1.0-1.15v depending on chip.You can have it set to auto but the BIOS might supply it excessively. So set it manually is better than letting mobo to decide for you. |
| Change to: | 0.0432sec
0.36
7 queries
GZIP Disabled
Time is now: 3rd December 2025 - 05:15 PM |