Overclock.net › Forums › AMD › AMD CPUs › is this normal for a 8320
New Posts  All Forums:Forum Nav:

is this normal for a 8320 - Page 4

post #31 of 35
Quote:
Originally Posted by mdocod View Post

Quote:
Originally Posted by Undervolter View Post

The first day they claim they can run 1.09. Then a day later, they say that "after all, it needed 1.12v to be stable". A day later they say "well, i actually had a BSOD and ran Prime95 for 2 hours and now i am at 1.15v". At the end they settle for 1.18v.

Bottom line. What people say in forums, when there is no standard testing method and each uses whatever he likes for whatever time he likes, becomes highly questionable and unreliable.




I ran 5ghz@1.43V for almost a month before I ran into stability problems. I was compiling, encoding, bench-marking without problems until one day, I started having issues.

The linux version of P95 doesn't "torture" test the way the win version does. I could never get it working right, my attempts to use it as a stability test didn't really work at all (no sign of instability, not much heat generated, etc). After getting the latest windows version of P95 working (it hadn't occurred to me till then to try it in wine, where it works fine) I realize now how unrealistic that 5ghz overclock was going to be for long term use. No doubt the chip and the board could do it, but I would need a lot more cooling and more voltage to actually be stable.

~4.7ghz@1.42V is (likely) about as good as it is really going to get on the Seidon 120V for "prime95" stability. I backed off and haven't "dialed in" a fine tune overclock yet.

That's one thing that I have always liked, these FX CPUs will run some pretty high clocks with low voltage and even do some work while there. Here's my 9370 doing almost 5.1 running Cinebench at 1.368v, I could probably game there too.
f430xh.jpg

But it's not even remotely stable at that speed/voltage add another .13v. Just like a previous post saying they can bench at 5.1 with so and so but benching and working are two different things when it come to computers.

Quote:
Originally Posted by Undervolter View Post

And another thing. The temperatures of the chokes or mofsets alone, don't guarantee that the VRM circuitry of the board can delivery the requested power. It doesn't mean that the VRM will necessarily overheat. It means that strcuturally, at some point, it can't deliver the power. In my link posted above, i didn't measure the VRM temp, but when i was at 4Ghz, i had measured it with IR thermometer and it wouldn't break 50C. But i still can't go to 4.4Ghz stable. In the first page of the link, where i "passed" IBT at 4.5Ghz, there was no throttling, hence the VRM must have been still in acceptable levels (top down cooler helps in that). Still, the board would fail Prime in 10 seconds. Same thing happened at 4.4Ghz. Passed at 4.3Ghz. Simply, the boards doesn't have enough power delivery to remain stable at anything beyond 4.3Ghz. I could blame Prime95 for being defective at 4.4 and 4.5, while being non defective at 4, 4.1, 4.2 and 4.3, but it would go beyond my logic.

The same is true with all boards they're limited by their circuitry. Sure they can go faster but not stable. The circuits get hot. I got berated when I first started posting in this forum when I said on an M5A99 board with an air cooler you'd be lucky to get 4.5 stable. The come backs were , well I get this and he gets that. They finally conceded when I said P95 blend stable so you could run F@H for three days straight and not crash. I have tried many different stressors and always come back to P95. When I first got my i74770k I followed a guide that suggested using AIDA64 for testing. I did and thought my CPU was effin amazing. I was running 4.7 @ 1.3v, ran it for 10 hours. Then I started up F@H and it crashed in seconds. I didn't understand, so I tried P95 and as soon as I hit the start, bam blue screen. I ended up at 4.5 with 1.32v, this was stable but I had wasted a whole day messing around re-doing my OC.
Bench
(9 items)
 
Intel Bench
(8 items)
 
DVD Player
(6 items)
 
CPUMotherboardGraphicsRAM
FX 9370,FX 8350, FX 6350, PII X4 965 ASUS Crosshair V Formula z, ASUS Sabertooth 990... 2x ASUS GTX580 DCUII, ASUS HD 6970 G.Skill Flare 1800, Pi Black 2000 
Hard DriveCoolingOSPower
Kingston Hyperx 3K Phobya Supernova 1260 Koolance 380a Win7 Pro 64 Leadex SuperFlower 1K 
Case
Banchetto 101 
CPUMotherboardGraphicsRAM
i7 6700k Asus Maximus VIII Hero Evga GTX 980Ti K|NGP|N G.Skill TridentZ 3600 
Hard DriveCoolingOSPower
Samsung 850 Pro Dragon Extreme F1 Windows 7 / Vista /XP Superflower Leadex 
  hide details  
Reply
Bench
(9 items)
 
Intel Bench
(8 items)
 
DVD Player
(6 items)
 
CPUMotherboardGraphicsRAM
FX 9370,FX 8350, FX 6350, PII X4 965 ASUS Crosshair V Formula z, ASUS Sabertooth 990... 2x ASUS GTX580 DCUII, ASUS HD 6970 G.Skill Flare 1800, Pi Black 2000 
Hard DriveCoolingOSPower
Kingston Hyperx 3K Phobya Supernova 1260 Koolance 380a Win7 Pro 64 Leadex SuperFlower 1K 
Case
Banchetto 101 
CPUMotherboardGraphicsRAM
i7 6700k Asus Maximus VIII Hero Evga GTX 980Ti K|NGP|N G.Skill TridentZ 3600 
Hard DriveCoolingOSPower
Samsung 850 Pro Dragon Extreme F1 Windows 7 / Vista /XP Superflower Leadex 
  hide details  
Reply
post #32 of 35
Quote:
Originally Posted by Johan45 View Post


But it's not even remotely stable at that speed/voltage add another .13v. Just like a previous post saying they can bench at 5.1 with so and so but benching and working are two different things when it come to computers.
The same is true with all boards they're limited by their circuitry. Sure they can go faster but not stable. The circuits get hot. I got berated when I first started posting in this forum when I said on an M5A99 board with an air cooler you'd be lucky to get 4.5 stable. The come backs were , well I get this and he gets that. They finally conceded when I said P95 blend stable so you could run F@H for three days straight and not crash. I have tried many different stressors and always come back to P95. When I first got my i74770k I followed a guide that suggested using AIDA64 for testing. I did and thought my CPU was effin amazing. I was running 4.7 @ 1.3v, ran it for 10 hours. Then I started up F@H and it crashed in seconds. I didn't understand, so I tried P95 and as soon as I hit the start, bam blue screen. I ended up at 4.5 with 1.32v, this was stable but I had wasted a whole day messing around re-doing my OC.

Unfortunately, people have "expectations". If they expect to hit say 5Ghz, most of them will hang tight on that idea and blame everything else they can before accepting the idea, that they can't... Just look at the HardOCP review, made by a "pro". Instead of starting with the mentality "let's see how high we can go", he started with the opposite mentality. "We will go to 4.5Ghz". Once he couldn't run Prime stable, he concluded there is something wrong in Prime95. rolleyes.gif Because "this happened to many others boards, except the Gigabyte XYZ". Because Prime as soon as it saw the Gigabyte, assumes a friendly behaviour and decides to collaborate. Prime95 is a Gigabyte fan. Never mind also the 4h hour with fan, 1h without fan episode. It doesn't mean anything. Prime has a problem. Conclusion: Editor's choice, Silver award for the MSI board!". This has "tester bias" written all over it.

My conclusion would be: "Listen guys, for those of you who still like the safety level that Prime95 provides, this board, with this chip, can't hit 4.5Ghz. We managed to run stable at 4.4Ghz. For those of you happy enough with IBT/OCCT/AIDA, we hit 4.5Ghz. We also had an episode with different endurance depending on cooling, which indicates, that some part of the board, was suffering during Prime. Goodbye".

If this is widespread on all models except Gigabyte, i would suspect that the new board chipset specs, have cheaped out on the VRM circuitry. Or, i would run Prime at stock, fail, and show the world, the bug in the software. That's how usually software bugs are proven, aren't they...
Main
(16 items)
 
Dedi-Coder
(15 items)
 
 
CPUMotherboardGraphicsRAM
FX-8320@4Ghz Asrock 970 Extreme3 Gainward GTX 750Ti Corsair XMS3 1600Mhz 16GB (4x4GB) 
Hard DriveHard DriveOptical DriveOptical Drive
Crucial MX100 250GB Western Digital Green 2TB LiteOn Blu-Ray Burner IHBS 112-2 LG BH16NS55 Blu-Ray Burner 
CoolingOSMonitorKeyboard
Scythe Katana 3 Windows 7 Pro 64bit ASUS 22" VS228HR Microsoft Wired Keyboard 600 
PowerCaseMouseAudio
EVGA 430W Sharkoon VG4-V Logitech M90 Onboard 
CPUMotherboardGraphicsRAM
FX-6300@4Ghz Asrock 970 Extreme3 HIS 6570 Silence Corsair XMS3 1600Mhz 8GB (2x4GB) CAS9 
Hard DriveHard DriveOptical DriveCooling
Plextor M6S 128GB Toshiba 2TB SATAIII LiteOn Blu Ray burner IHBS 112-2 Xigmatek Balder 
OSMonitorKeyboardPower
Windows 7 Pro 64bit Samsung S22B350H Microsoft Wired 600 Corsair VS350 
CaseMouseAudio
Lepa LPC 306 Logitech M90 Onboard 
  hide details  
Reply
Main
(16 items)
 
Dedi-Coder
(15 items)
 
 
CPUMotherboardGraphicsRAM
FX-8320@4Ghz Asrock 970 Extreme3 Gainward GTX 750Ti Corsair XMS3 1600Mhz 16GB (4x4GB) 
Hard DriveHard DriveOptical DriveOptical Drive
Crucial MX100 250GB Western Digital Green 2TB LiteOn Blu-Ray Burner IHBS 112-2 LG BH16NS55 Blu-Ray Burner 
CoolingOSMonitorKeyboard
Scythe Katana 3 Windows 7 Pro 64bit ASUS 22" VS228HR Microsoft Wired Keyboard 600 
PowerCaseMouseAudio
EVGA 430W Sharkoon VG4-V Logitech M90 Onboard 
CPUMotherboardGraphicsRAM
FX-6300@4Ghz Asrock 970 Extreme3 HIS 6570 Silence Corsair XMS3 1600Mhz 8GB (2x4GB) CAS9 
Hard DriveHard DriveOptical DriveCooling
Plextor M6S 128GB Toshiba 2TB SATAIII LiteOn Blu Ray burner IHBS 112-2 Xigmatek Balder 
OSMonitorKeyboardPower
Windows 7 Pro 64bit Samsung S22B350H Microsoft Wired 600 Corsair VS350 
CaseMouseAudio
Lepa LPC 306 Logitech M90 Onboard 
  hide details  
Reply
post #33 of 35
It is odd that it would work on one board and not others. To me indicates bios/chipset possibly. I say that because the majority of the voltage control has been implemented on die with the Hazwells and the refresh added a few more caps on the die.Basically you select an Input voltage to the CPU of say 1.8v then from there it's divided to feed cache/core etc. Yes the expectations were too high, most first gen hazwell couldn't do 4.5 on air so why would the refresh be any better.
Bench
(9 items)
 
Intel Bench
(8 items)
 
DVD Player
(6 items)
 
CPUMotherboardGraphicsRAM
FX 9370,FX 8350, FX 6350, PII X4 965 ASUS Crosshair V Formula z, ASUS Sabertooth 990... 2x ASUS GTX580 DCUII, ASUS HD 6970 G.Skill Flare 1800, Pi Black 2000 
Hard DriveCoolingOSPower
Kingston Hyperx 3K Phobya Supernova 1260 Koolance 380a Win7 Pro 64 Leadex SuperFlower 1K 
Case
Banchetto 101 
CPUMotherboardGraphicsRAM
i7 6700k Asus Maximus VIII Hero Evga GTX 980Ti K|NGP|N G.Skill TridentZ 3600 
Hard DriveCoolingOSPower
Samsung 850 Pro Dragon Extreme F1 Windows 7 / Vista /XP Superflower Leadex 
  hide details  
Reply
Bench
(9 items)
 
Intel Bench
(8 items)
 
DVD Player
(6 items)
 
CPUMotherboardGraphicsRAM
FX 9370,FX 8350, FX 6350, PII X4 965 ASUS Crosshair V Formula z, ASUS Sabertooth 990... 2x ASUS GTX580 DCUII, ASUS HD 6970 G.Skill Flare 1800, Pi Black 2000 
Hard DriveCoolingOSPower
Kingston Hyperx 3K Phobya Supernova 1260 Koolance 380a Win7 Pro 64 Leadex SuperFlower 1K 
Case
Banchetto 101 
CPUMotherboardGraphicsRAM
i7 6700k Asus Maximus VIII Hero Evga GTX 980Ti K|NGP|N G.Skill TridentZ 3600 
Hard DriveCoolingOSPower
Samsung 850 Pro Dragon Extreme F1 Windows 7 / Vista /XP Superflower Leadex 
  hide details  
Reply
post #34 of 35
Quote:
Originally Posted by Johan45 View Post

It is odd that it would work on one board and not others. To me indicates bios/chipset possibly. I say that because the majority of the voltage control has been implemented on die with the Hazwells and the refresh added a few more caps on the die.Basically you select an Input voltage to the CPU of say 1.8v then from there it's divided to feed cache/core etc. Yes the expectations were too high, most first gen hazwell couldn't do 4.5 on air so why would the refresh be any better.

Any explanation is more reasonable, than accusing the software. A user at the end of that discussion, says he ran 8h Prime with no problems. I am sorry, a software like that, doesn't turn on and off it's bugs depending on the board and Ghz. Either it's buggy or it's not. At that point, one should search the fault elsewhere. If you 're having the same problem with many boards, then you 're doing a "systematic testing error" and you should rethink the way you test.
Main
(16 items)
 
Dedi-Coder
(15 items)
 
 
CPUMotherboardGraphicsRAM
FX-8320@4Ghz Asrock 970 Extreme3 Gainward GTX 750Ti Corsair XMS3 1600Mhz 16GB (4x4GB) 
Hard DriveHard DriveOptical DriveOptical Drive
Crucial MX100 250GB Western Digital Green 2TB LiteOn Blu-Ray Burner IHBS 112-2 LG BH16NS55 Blu-Ray Burner 
CoolingOSMonitorKeyboard
Scythe Katana 3 Windows 7 Pro 64bit ASUS 22" VS228HR Microsoft Wired Keyboard 600 
PowerCaseMouseAudio
EVGA 430W Sharkoon VG4-V Logitech M90 Onboard 
CPUMotherboardGraphicsRAM
FX-6300@4Ghz Asrock 970 Extreme3 HIS 6570 Silence Corsair XMS3 1600Mhz 8GB (2x4GB) CAS9 
Hard DriveHard DriveOptical DriveCooling
Plextor M6S 128GB Toshiba 2TB SATAIII LiteOn Blu Ray burner IHBS 112-2 Xigmatek Balder 
OSMonitorKeyboardPower
Windows 7 Pro 64bit Samsung S22B350H Microsoft Wired 600 Corsair VS350 
CaseMouseAudio
Lepa LPC 306 Logitech M90 Onboard 
  hide details  
Reply
Main
(16 items)
 
Dedi-Coder
(15 items)
 
 
CPUMotherboardGraphicsRAM
FX-8320@4Ghz Asrock 970 Extreme3 Gainward GTX 750Ti Corsair XMS3 1600Mhz 16GB (4x4GB) 
Hard DriveHard DriveOptical DriveOptical Drive
Crucial MX100 250GB Western Digital Green 2TB LiteOn Blu-Ray Burner IHBS 112-2 LG BH16NS55 Blu-Ray Burner 
CoolingOSMonitorKeyboard
Scythe Katana 3 Windows 7 Pro 64bit ASUS 22" VS228HR Microsoft Wired Keyboard 600 
PowerCaseMouseAudio
EVGA 430W Sharkoon VG4-V Logitech M90 Onboard 
CPUMotherboardGraphicsRAM
FX-6300@4Ghz Asrock 970 Extreme3 HIS 6570 Silence Corsair XMS3 1600Mhz 8GB (2x4GB) CAS9 
Hard DriveHard DriveOptical DriveCooling
Plextor M6S 128GB Toshiba 2TB SATAIII LiteOn Blu Ray burner IHBS 112-2 Xigmatek Balder 
OSMonitorKeyboardPower
Windows 7 Pro 64bit Samsung S22B350H Microsoft Wired 600 Corsair VS350 
CaseMouseAudio
Lepa LPC 306 Logitech M90 Onboard 
  hide details  
Reply
post #35 of 35
Thread Starter 
i have to agree software is software im glad u guys all chimed in caz on the first 2 post my thought was something is buged with p95 ... iv never oced without it befor

on the amd fx id be amazed to see anything stable after 4.5 on air considering the sheer heat it dumps even in my waterloop im topping at 4.8

yea the m5a99fx is not doubt a warm board but its good if u cool it ... maby i should of bought another gigabyte ud but thats a topic for another time

its been going on for years tho prople dont p95 there rigs more than maby a hr and call it stabl and maby for them its fine ... if u dont ever stress ur cpu your all fine i guess

also tho just to point it out you can prove somewhat wether its a app or hardwhere check ur logs it will show critical hardware or software error i checked my logs shortly after u guys said prime is fine and critical hardware errors

also with hardware u got to calculate in age and stress as a overclocked componet ages it looses stability i got one rig i just run as a htpc and mine litecoins on the side with and after a week of no ac its no longer stable at stock speeds im guessing it overheated bad that ones almost 5 years old tho was under water high oc for the 5 years and the chip just cant proform without watercooling anymore
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD CPUs
Overclock.net › Forums › AMD › AMD CPUs › is this normal for a 8320