Overclock.net › Forums › Software, Programming and Coding › Operating Systems › Linux, Unix › Should i reset OC before installing ubuntu?
New Posts  All Forums:Forum Nav:

Should i reset OC before installing ubuntu? - Page 3

post #21 of 30
Quote:
Originally Posted by enorbet2 View Post
Greetz
The below linked article from Anandtech is so good it should probably have a thread all it's own. I heartily recommend it to anyone regardless of OpSys who has any interest in how CPUs function, stability and overclocking, and frankly, if you don't then why are you here? The article is highly technical but with the aid of graphs and excellent writing the n00best of n00bs will get something from it. I promise.

Se it here.... http://www.anandtech.com/show/2468/6
That's interesting, it's showing degradation much quicker than I expected it to be. Is that for all CPU's now? Just 45mn?

My one question, how does running stress tests and high over clocks work into the equation?

I ask that because that's what most people do here, they push insane voltages and clocks while "stress testing" with some pretty heafty programs. I'm sure that has to cut into the life of the chip some how, especially with all the gamers who do it.
Current Rig
(14 items)
 
  
CPUMotherboardGraphicsRAM
FX-8350 4.6GHz@1.44v GA-990FXA-UD3 R4.0 HD 7950 (1100/1450) 8G Muskin DDR3 1866@8CLS 
Hard DriveOptical DriveOSMonitor
1TB WD LiteOn DVD-RW DL Linux/Windows 19" Phillips TV 1080p 
PowerCaseMouseMouse Pad
OCZ 600W Generic Junk Logitech MX400 Generic Junk 
Audio
SBL 5.1 
  hide details  
Reply
Current Rig
(14 items)
 
  
CPUMotherboardGraphicsRAM
FX-8350 4.6GHz@1.44v GA-990FXA-UD3 R4.0 HD 7950 (1100/1450) 8G Muskin DDR3 1866@8CLS 
Hard DriveOptical DriveOSMonitor
1TB WD LiteOn DVD-RW DL Linux/Windows 19" Phillips TV 1080p 
PowerCaseMouseMouse Pad
OCZ 600W Generic Junk Logitech MX400 Generic Junk 
Audio
SBL 5.1 
  hide details  
Reply
post #22 of 30
Interesting article thanks for posting.


I don't like the idea of 24 hr stress tests that some people recommend. I usually test for a couple hours max and then rely on real life use to find any real problems. I do run aida64 and prime 95 together for 10-20 minutes and 99% of the time that will prove stability.


I actually get lower temps with the h70 at 4.1 ghz than I did at stock voltages and speeds with the stock cooler that has to count for something and I'm still well under intel's max voltages.
XEON POWER
(17 items)
 
  
CPUMotherboardGraphicsRAM
Xeon E3-1280 V2 Asus p8h77-i EVGA GTX 570  Corsair Dominator Platinum 1866 cl9 16gb 2x8gb 
Hard DriveHard DriveHard DriveOptical Drive
Samsung Spinpoint F3 Samsung 830 Crucial C300 Samsung USB Blu-Ray 
OSMonitorMonitorKeyboard
Windows 8 Dell U2711 HP 2709m Razer Blackwidow 
PowerCaseMouseMouse Pad
Silverstone Strider Essentials ST60ES-F Bitfenix Prodigy Razer Naga Molten razer sphex 
Audio
onboard 
  hide details  
Reply
XEON POWER
(17 items)
 
  
CPUMotherboardGraphicsRAM
Xeon E3-1280 V2 Asus p8h77-i EVGA GTX 570  Corsair Dominator Platinum 1866 cl9 16gb 2x8gb 
Hard DriveHard DriveHard DriveOptical Drive
Samsung Spinpoint F3 Samsung 830 Crucial C300 Samsung USB Blu-Ray 
OSMonitorMonitorKeyboard
Windows 8 Dell U2711 HP 2709m Razer Blackwidow 
PowerCaseMouseMouse Pad
Silverstone Strider Essentials ST60ES-F Bitfenix Prodigy Razer Naga Molten razer sphex 
Audio
onboard 
  hide details  
Reply
post #23 of 30
Quote:
Originally Posted by cavallino View Post
Interesting article thanks for posting.


I don't like the idea of 24 hr stress tests that some people recommend. I usually test for a couple hours max and then rely on real life use to find any real problems. I do run aida64 and prime 95 together for 10-20 minutes and 99% of the time that will prove stability.


I actually get lower temps with the h70 at 4.1 ghz than I did at stock voltages and speeds with the stock cooler that has to count for something and I'm still well under intel's max voltages.
run prime for longer than that

at least 6 hours for a stress test... ive had a cpu show stability for 4 hours then just die in prime
Kinda meh now...
(13 items)
 
  
CPUMotherboardGraphicsRAM
PhII 940 BE asus M4N82 Deluxe 2x 8800gts(g92) 2x 1g ocz ddr2 1066 
Hard DriveOptical DriveOSMonitor
/dev/sd[abc] /dev/sr0 WinXP/Linux 22in acer 
PowerCaseMouseMouse Pad
950w rosewill timebomb Antec 900 (2small) Voodoo Deathadder (RED!) X-Trac Pads Ripper XL 
  hide details  
Reply
Kinda meh now...
(13 items)
 
  
CPUMotherboardGraphicsRAM
PhII 940 BE asus M4N82 Deluxe 2x 8800gts(g92) 2x 1g ocz ddr2 1066 
Hard DriveOptical DriveOSMonitor
/dev/sd[abc] /dev/sr0 WinXP/Linux 22in acer 
PowerCaseMouseMouse Pad
950w rosewill timebomb Antec 900 (2small) Voodoo Deathadder (RED!) X-Trac Pads Ripper XL 
  hide details  
Reply
post #24 of 30
I've always had problems with OCs and Linux oses (but Debian didn't seem to give a damn about my clock speeds, it just worked no matter what). So if I wanted to use a Linux distro I had to reset my clock settings. What seemed right for Windows was bad for Linux. So heres hoping you won't have to.
post #25 of 30
I got a copy of OSX to play around with and it's even more picky haven't figured out the max OC yet but it's somewhere below the stable 4.35 in linux.

So so far it's 4.4 in windows 4.35 in Linux and somewhere less in OSX.
XEON POWER
(17 items)
 
  
CPUMotherboardGraphicsRAM
Xeon E3-1280 V2 Asus p8h77-i EVGA GTX 570  Corsair Dominator Platinum 1866 cl9 16gb 2x8gb 
Hard DriveHard DriveHard DriveOptical Drive
Samsung Spinpoint F3 Samsung 830 Crucial C300 Samsung USB Blu-Ray 
OSMonitorMonitorKeyboard
Windows 8 Dell U2711 HP 2709m Razer Blackwidow 
PowerCaseMouseMouse Pad
Silverstone Strider Essentials ST60ES-F Bitfenix Prodigy Razer Naga Molten razer sphex 
Audio
onboard 
  hide details  
Reply
XEON POWER
(17 items)
 
  
CPUMotherboardGraphicsRAM
Xeon E3-1280 V2 Asus p8h77-i EVGA GTX 570  Corsair Dominator Platinum 1866 cl9 16gb 2x8gb 
Hard DriveHard DriveHard DriveOptical Drive
Samsung Spinpoint F3 Samsung 830 Crucial C300 Samsung USB Blu-Ray 
OSMonitorMonitorKeyboard
Windows 8 Dell U2711 HP 2709m Razer Blackwidow 
PowerCaseMouseMouse Pad
Silverstone Strider Essentials ST60ES-F Bitfenix Prodigy Razer Naga Molten razer sphex 
Audio
onboard 
  hide details  
Reply
post #26 of 30
Quote:
Originally Posted by cavallino View Post
I got a copy of OSX to play around with and it's even more picky haven't figured out the max OC yet but it's somewhere below the stable 4.35 in linux.

So so far it's 4.4 in windows 4.35 in Linux and somewhere less in OSX.
that is probably due to the fact it uses parts from the bsd kernel, which is much more anal in terms of errors than its distant cousin linux

i'm actually quite surprised by the disparity between the operating systems tbh with you...i've had "stable" windows OC's fail in a heartbeat in linux...i can only imagine the data corruption you might be getting, as mushroomboy pointed out, i'm the same way, if it isn't stable in linux, i don't consider it stable in windows...but i'm not saying all windows users need to "double" check their oc's in linux, if they don't use it, then they can continue away imho
Bazinga Punk
(12 items)
 
ooh shiny!
(6 items)
 
 
CPUMotherboardGraphicsRAM
Intel Xeon 3440 AsRock P55 extreme Evga 8800 GT 512 MB Gskill Ripjaws 
Hard DriveCoolingOSMonitor
Western Digital Blue Antec Khuler 620 Ubuntu 11.10 Asus vw264H 
KeyboardPowerCaseMouse
GIGABYTE KM7600 CORSAIR TX 650 Cooler Master 590 GIGABYTE GM-M6800 
CPUMotherboardGraphicsRAM
Intel Core I5 6500 Gigabyte z170xp-SLI Nvidia 970gtx Corsair 16gb ddr4 2666mhz  
Hard DriveOS
250gb Samsung Evo 850 Windows 10 & Ubuntu 15.10 
  hide details  
Reply
Bazinga Punk
(12 items)
 
ooh shiny!
(6 items)
 
 
CPUMotherboardGraphicsRAM
Intel Xeon 3440 AsRock P55 extreme Evga 8800 GT 512 MB Gskill Ripjaws 
Hard DriveCoolingOSMonitor
Western Digital Blue Antec Khuler 620 Ubuntu 11.10 Asus vw264H 
KeyboardPowerCaseMouse
GIGABYTE KM7600 CORSAIR TX 650 Cooler Master 590 GIGABYTE GM-M6800 
CPUMotherboardGraphicsRAM
Intel Core I5 6500 Gigabyte z170xp-SLI Nvidia 970gtx Corsair 16gb ddr4 2666mhz  
Hard DriveOS
250gb Samsung Evo 850 Windows 10 & Ubuntu 15.10 
  hide details  
Reply
post #27 of 30
Quote:
Originally Posted by transhour View Post
that is probably due to the fact it uses parts from the bsd kernel, which is much more anal in terms of errors than its distant cousin linux

i'm actually quite surprised by the disparity between the operating systems tbh with you...i've had "stable" windows OC's fail in a heartbeat in linux...i can only imagine the data corruption you might be getting, as mushroomboy pointed out, i'm the same way, if it isn't stable in linux, i don't consider it stable in windows...but i'm not saying all windows users need to "double" check their oc's in linux, if they don't use it, then they can continue away imho
Yeah I run Windows and Linux on two separate PC's both now have clean installs that have been crash free once I got it more stable.

My usual test method is 20-30 minutes of aida64 and pirme95 together. Followed by 3-5 hours of prime95 alone. I admit though on my i3 rig I was a bit lax and that's now shared between Linux and osx on seperate hdds. I did do more extensive testing in Linux from a live usb on my sigrig and it was 5-6 hours stable in mprime at the same clocks as windows. I was more extensive with my initial testing on that PC so I guess it payed off. I still think it's a bit much to run a stress test for 24 hours.

I'm still not sure the OSX thing isn't some other issue as I dropped way down on the multi/bclk and left the voltage generous and still had an issue. It may be some other hardware/driver problem. It required a bit of rigmarole to get it running. I'm not knowledgeable enough about OS X to know how to identify specific issues. I will report back after I have more time to test it.
Edited by cavallino - 4/4/11 at 10:37pm
XEON POWER
(17 items)
 
  
CPUMotherboardGraphicsRAM
Xeon E3-1280 V2 Asus p8h77-i EVGA GTX 570  Corsair Dominator Platinum 1866 cl9 16gb 2x8gb 
Hard DriveHard DriveHard DriveOptical Drive
Samsung Spinpoint F3 Samsung 830 Crucial C300 Samsung USB Blu-Ray 
OSMonitorMonitorKeyboard
Windows 8 Dell U2711 HP 2709m Razer Blackwidow 
PowerCaseMouseMouse Pad
Silverstone Strider Essentials ST60ES-F Bitfenix Prodigy Razer Naga Molten razer sphex 
Audio
onboard 
  hide details  
Reply
XEON POWER
(17 items)
 
  
CPUMotherboardGraphicsRAM
Xeon E3-1280 V2 Asus p8h77-i EVGA GTX 570  Corsair Dominator Platinum 1866 cl9 16gb 2x8gb 
Hard DriveHard DriveHard DriveOptical Drive
Samsung Spinpoint F3 Samsung 830 Crucial C300 Samsung USB Blu-Ray 
OSMonitorMonitorKeyboard
Windows 8 Dell U2711 HP 2709m Razer Blackwidow 
PowerCaseMouseMouse Pad
Silverstone Strider Essentials ST60ES-F Bitfenix Prodigy Razer Naga Molten razer sphex 
Audio
onboard 
  hide details  
Reply
post #28 of 30
Quote:
Originally Posted by cavallino View Post
Yeah I run Windows and Linux on two separate PC's both now have clean installs that have been crash free once I got it more stable.

My usual test method is 20-30 minutes of aida64 and pirme95 together. Followed by 3-5 hours of prime95 alone. I admit though on my i3 rig I was a bit lax and that's now shared between Linux and osx on seperate hdds. I did do more extensive testing in Linux from a live usb on my sigrig and it was 5-6 hours stable in mprime at the same clocks as windows. I was more extensive with my initial testing on that PC so I guess it payed off. I still think it's a bit much to run a stress test for 24 hours.

I'm still not sure the OSX thing isn't some other issue as I dropped way down on the multi/bclk and left the voltage generous and still had an issue. It may be some other hardware/driver problem. It required a bit of rigmarole to get it running. I'm not knowledgeable enough about OS X to know how to identify specific issues. I will report back after I have more time to test it.
i think the running of the "stress" test for 24 hours has to do with how well it handles long periods of load and heat dissipation. i typically let it complete its set of cpu only test on prime95 (for windows, i also use the windows version in wine as it will run properly without problems in my expeirence.), i then reboot into memtest and let it run a few passes, as long as that all runs without error, i call it good.

i too don't understand the obsessive need to "burn in" a OC, cause i can't imagine what it proves or accomplishes tbh...i've seen OC's run prime stable for 12 hours, but then a few days later run prime again and it fails with in 5 minutes, restart prime and it will run for another 12 hours without problem...i think the running 24 hours once, gives a false sense of "stability" as i don't know many who will repeatedly let it run for 24 hours straight after it passes the first time...i've also seen 24 hour prime stable oc's, not being able to pass the second test on lin-x or intel burn test...so i guess i'm in the same boat as mushroomboy (gotta stop doing that, i tend to agree with that guy way too often! ) that there is no such thing as a rock solid "stable" OC only "acceptable" stable.

for good measures at time, i grab source code to large things i know will take awhile to "compile", typically i use the kernel source, throw a default config on it, and then run the:

make -j12

it will put a lot of stress on the cpu, probably more so than prime or lin-x can, or i leave my computer on over night and run programs i know will peak at 100% core usage or use 10 to 20% on average while running (chromium/FF on a flash site will do this most of the time.). there are lot of ways to "stress" your cpu without the use of "stress" programs, boinc comes to mind, so does F@H...to as how long you should do it to get an "acceptable" stable, well i look at this way, take the most cpu/resource intensive thing you do, figure on average how long you do it, then multiply that by 2, and then you should get an idea of how long you should "stress" for
Bazinga Punk
(12 items)
 
ooh shiny!
(6 items)
 
 
CPUMotherboardGraphicsRAM
Intel Xeon 3440 AsRock P55 extreme Evga 8800 GT 512 MB Gskill Ripjaws 
Hard DriveCoolingOSMonitor
Western Digital Blue Antec Khuler 620 Ubuntu 11.10 Asus vw264H 
KeyboardPowerCaseMouse
GIGABYTE KM7600 CORSAIR TX 650 Cooler Master 590 GIGABYTE GM-M6800 
CPUMotherboardGraphicsRAM
Intel Core I5 6500 Gigabyte z170xp-SLI Nvidia 970gtx Corsair 16gb ddr4 2666mhz  
Hard DriveOS
250gb Samsung Evo 850 Windows 10 & Ubuntu 15.10 
  hide details  
Reply
Bazinga Punk
(12 items)
 
ooh shiny!
(6 items)
 
 
CPUMotherboardGraphicsRAM
Intel Xeon 3440 AsRock P55 extreme Evga 8800 GT 512 MB Gskill Ripjaws 
Hard DriveCoolingOSMonitor
Western Digital Blue Antec Khuler 620 Ubuntu 11.10 Asus vw264H 
KeyboardPowerCaseMouse
GIGABYTE KM7600 CORSAIR TX 650 Cooler Master 590 GIGABYTE GM-M6800 
CPUMotherboardGraphicsRAM
Intel Core I5 6500 Gigabyte z170xp-SLI Nvidia 970gtx Corsair 16gb ddr4 2666mhz  
Hard DriveOS
250gb Samsung Evo 850 Windows 10 & Ubuntu 15.10 
  hide details  
Reply
post #29 of 30
Quote:
Originally Posted by mushroomboy View Post
That's interesting, it's showing degradation much quicker than I expected it to be. Is that for all CPU's now? Just 45mn?

My one question, how does running stress tests and high over clocks work into the equation?

I ask that because that's what most people do here, they push insane voltages and clocks while "stress testing" with some pretty heafty programs. I'm sure that has to cut into the life of the chip some how, especially with all the gamers who do it.
That is a very good question because one thing I don't detect in the graphs, unless it is hidden in "wiggle room", is the unexpected but experimentally verified concept of "work hardening". It was originally thought that digital electronics would not experience such an improvement as was long documented especially by the military for occurring in analog electronics as well as simple machines. An example in machines is the "break in period" for cars, where how the car is driven for the first thousand miles powerfully affects both performance and lifespan from then on. It has been proven that CPU's and mobos "burn in" and get used to operation conditions.

So it isn’t as simple to carry on their analogy that as soon as we are born we begin to die.
NewMain
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel i5 - 3550 Asrock Z77 Extreme4 Gigabyte GTX 760  4x2GB Corsair Vengeance 
Hard DriveOptical DriveCoolingOS
Seagate SATA 2TB x 2  Plextor PX-891SAW CM-Hyper N520 Slackware 14, Studio KUbuntu, OpenSuSe 12.3, Wi... 
MonitorKeyboardPowerCase
32" Vizio HDTV + DLP Logitech Wireless Corsair HX-850 Antec Sonata I 
MouseMouse PadAudioOther
Razer DeathAdder 2013 dual ESI Juli@ CoolGear ExtSata Enclosure w/ Optical and 3TB S... 
  hide details  
Reply
NewMain
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel i5 - 3550 Asrock Z77 Extreme4 Gigabyte GTX 760  4x2GB Corsair Vengeance 
Hard DriveOptical DriveCoolingOS
Seagate SATA 2TB x 2  Plextor PX-891SAW CM-Hyper N520 Slackware 14, Studio KUbuntu, OpenSuSe 12.3, Wi... 
MonitorKeyboardPowerCase
32" Vizio HDTV + DLP Logitech Wireless Corsair HX-850 Antec Sonata I 
MouseMouse PadAudioOther
Razer DeathAdder 2013 dual ESI Juli@ CoolGear ExtSata Enclosure w/ Optical and 3TB S... 
  hide details  
Reply
post #30 of 30
Quote:
Originally Posted by enorbet2 View Post
That is a very good question because one thing I don't detect in the graphs, unless it is hidden in "wiggle room", is the unexpected but experimentally verified concept of "work hardening". It was originally thought that digital electronics would not experience such an improvement as was long documented especially by the military for occurring in analog electronics as well as simple machines. An example in machines is the "break in period" for cars, where how the car is driven for the first thousand miles powerfully affects both performance and lifespan from then on. It has been proven that CPU's and mobos "burn in" and get used to operation conditions.

So it isn’t as simple to carry on their analogy that as soon as we are born we begin to die.
i've never seen any thing but circumstantial and anecdotal evidence to support "burn in" of silicon components, often time this "evidence" will negate, ignore, write off to other factors, why some chips do supremely well, when others fail or even explode after a few minutes/days. electron migration is a very real and provable "phenomenon", where "burn in" is an elitist way of saying "oh my cpu is better than yours!".

sorry to disagree with you, but even i have my limits bro
Bazinga Punk
(12 items)
 
ooh shiny!
(6 items)
 
 
CPUMotherboardGraphicsRAM
Intel Xeon 3440 AsRock P55 extreme Evga 8800 GT 512 MB Gskill Ripjaws 
Hard DriveCoolingOSMonitor
Western Digital Blue Antec Khuler 620 Ubuntu 11.10 Asus vw264H 
KeyboardPowerCaseMouse
GIGABYTE KM7600 CORSAIR TX 650 Cooler Master 590 GIGABYTE GM-M6800 
CPUMotherboardGraphicsRAM
Intel Core I5 6500 Gigabyte z170xp-SLI Nvidia 970gtx Corsair 16gb ddr4 2666mhz  
Hard DriveOS
250gb Samsung Evo 850 Windows 10 & Ubuntu 15.10 
  hide details  
Reply
Bazinga Punk
(12 items)
 
ooh shiny!
(6 items)
 
 
CPUMotherboardGraphicsRAM
Intel Xeon 3440 AsRock P55 extreme Evga 8800 GT 512 MB Gskill Ripjaws 
Hard DriveCoolingOSMonitor
Western Digital Blue Antec Khuler 620 Ubuntu 11.10 Asus vw264H 
KeyboardPowerCaseMouse
GIGABYTE KM7600 CORSAIR TX 650 Cooler Master 590 GIGABYTE GM-M6800 
CPUMotherboardGraphicsRAM
Intel Core I5 6500 Gigabyte z170xp-SLI Nvidia 970gtx Corsair 16gb ddr4 2666mhz  
Hard DriveOS
250gb Samsung Evo 850 Windows 10 & Ubuntu 15.10 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Linux, Unix
Overclock.net › Forums › Software, Programming and Coding › Operating Systems › Linux, Unix › Should i reset OC before installing ubuntu?