Jump to content
Search In
  • More options...
Find results that contain...
Find results in...

How do you check if the OC of a CPU is stable?

If I raise clock speed and little bit of voltage how can I tell if it's enough voltage?

 

Is the system not crashing while playing games enough of an indicator or is there something more to check?

 

Can not enough voltage hurt the CPU or performance in games?

 

Thanks.

When I give advice I use my experience and google, I know alot, but not everything.

Have a question? - quote me = I'll respond instantly or in a day (if it's longer than 2 days, don't expect me to come back soon)

Specs I have experience with: (highlited build is being used)

Pentium G4500 | 4GB/8GB DDR4 2133Mhz | H110 | GTX 1050

Ryzen 3 1200 3,5Ghz / OC:4Ghz | 8GB DDR4 2133Mhz / 16GB 3200Mhz | B450 | GTX 1050

Link to post
Share on other sites
6 minutes ago, podkall said:

If I raise clock speed and little bit of voltage how can I tell if it's enough voltage?

 

A lot the time you won't even need to raise the voltage. Only raise the voltage if you run into instability. There are a number of programs that stress the CPU pretty well, the one I tend to use is Prime95, but Aida64 is another big one out there.

8 minutes ago, podkall said:

Is the system not crashing while playing games enough of an indicator or is there something more to check?

 

That's the first hurdle, but again, you want to run through a stress test. Long blender renders are another one that stresses the CPU a lot.

8 minutes ago, podkall said:

Can not enough voltage hurt the CPU or performance in games?

 

The voltage is really just to stabilize the CPU. Less voltage is actually better for the CPU in the long term, but not enough will usually cause crashing, so performance in games will therefore be pretty bad.

Link to post
Share on other sites
28 minutes ago, podkall said:

If I raise clock speed and little bit of voltage how can I tell if it's enough voltage?

 

Is the system not crashing while playing games enough of an indicator or is there something more to check?

 

Can not enough voltage hurt the CPU or performance in games?

 

Thanks.

There are many, many OC guides on YouTube and the internet, and I couldn’t possibly derail everything here for you.

 

To overclock correctly, stably, and safely, it takes a good amount of reading and understanding. Thankfully, the internet is full of information. Don’t just overclock because you heard the word and think you can wing it, that’s a recipe for a fried CPU. You need to understand what settings your changing, have a plan to test for stability and how to adjust settings to gain stability, check temps, and most importantly check voltages.

 

Google around for OC guides for your socket and chipset; there will be lots of articles to read. Then once you sorta get what’s going on, you can start fiddling with BIOS settings 🙂  

Rig: i7 10700k @ 5.1Ghz, 4.8 Ring - - Z490 Vision G - - EVGA RTX 2080 XC Ultra @ 2025Mhz - - 4x8GB Vengeance Pro 3000Mhz 15-17-17-34 @ 3500MHz 16-19-19-38 - - Samsung 950 Pro 512 NVMe Boot + Main Programs - - Samsung 830 Pro 256 RAID 0 Lightroom + Photo work - - WD Blue 1 TB SSD for Games - - Corsair RM850x - - Sound BlasterX EA-5 - - EK Supremacy Evo - - XT45 X-Flow 420 + UT60 280 rads - - EK Full Cover GPU Block - - EK XRES RGB PWM - - Fractal Define S2 - - Acer Predator X34 -- Logitech G502 - - Logitech G710+ - - Logitech Z5500 - - Steel Series QcK XXL

 

Headphones/amp/dac: Schiit Lyr 3 - - Fostex TR-X00 - - Sennheiser HD 6xx

 

Homelab/ Media Server: ESXi 6.5 - - 250 GB SSD for VM's/ESXi boot - - FreeNAS - - HPE Proliant ML10 Gen 9 backbone - - i3 6100 - - 28 GB ECC - - 10x4 TB WD Red RAID Z2 - - 10TB WD Red for expendable data - - Corsair 750D - - Corsair RM650i - - Dell H310 6Gbps SAS HBA - - Intel RES2SC240 SAS Expander

 

White Lightning (original full watercooled build) - Homelab / Media Server - The Blue Beast (my car and other expensive hobby...) - iPhone Xs - 2018 MacBook Air

Link to post
Share on other sites
9 hours ago, RONOTHAN## said:

Long blender renders are another one that stresses the CPU a lot.

the BMW one?

 

9 hours ago, RONOTHAN## said:

The voltage is really just to stabilize the CPU. Less voltage is actually better for the CPU in the long term, but not enough will usually cause crashing, so performance in games will therefore be pretty bad.

Can side effect of not enough of voltage be occasional stuttering when playing game?

9 hours ago, LIGISTX said:

There are many, many OC guides on YouTube and the internet, and I couldn’t possibly derail everything here for you.

 

To overclock correctly, stably, and safely, it takes a good amount of reading and understanding. Thankfully, the internet is full of information. Don’t just overclock because you heard the word and think you can wing it, that’s a recipe for a fried CPU. You need to understand what settings your changing, have a plan to test for stability and how to adjust settings to gain stability, check temps, and most importantly check voltages.

 

Google around for OC guides for your socket and chipset; there will be lots of articles to read. Then once you sorta get what’s going on, you can start fiddling with BIOS settings 🙂  

I already rolled my silicon dice, it's pretty decent as you can see in my signature.

I was asking because I only OC like half the amount for CPU demanding games and don't wanna hurt it in a longrun if if low voltage was one of possible ways to kill CPU, but thankfuly it isn't.

When I give advice I use my experience and google, I know alot, but not everything.

Have a question? - quote me = I'll respond instantly or in a day (if it's longer than 2 days, don't expect me to come back soon)

Specs I have experience with: (highlited build is being used)

Pentium G4500 | 4GB/8GB DDR4 2133Mhz | H110 | GTX 1050

Ryzen 3 1200 3,5Ghz / OC:4Ghz | 8GB DDR4 2133Mhz / 16GB 3200Mhz | B450 | GTX 1050

Link to post
Share on other sites
1 minute ago, podkall said:

the BMW one?

That's one of them

1 minute ago, podkall said:

Can side effect of not enough of voltage be occasional stuttering when playing game?

Not that I know of.

Link to post
Share on other sites
9 hours ago, podkall said:

If I raise clock speed and little bit of voltage how can I tell if it's enough voltage?

 

Is the system not crashing while playing games enough of an indicator or is there something more to check?

 

Can not enough voltage hurt the CPU or performance in games?

 

Thanks.

Is the question in regards to your sig rig? Ryzen 1200?

 

If so I've overclocked one. 4ghz is about the limit of this chip on ambient temps consuming nearly 1.40v for a steady 4ghz.

 

You probably won't find more performance from it. 

 

Low voltage typically results is a hang up. The pc freezes and sometimes accompanied by a black screen. 

 

Yes you may see degraded performance from a low v-core, but with early Ryzen chips I noticed the reference clock which should be 99 to 100mhz will often times dip even below 95mhz. In this case I was able to measure the performance loss with benchmarking. 

- If it ain't broken, don't fix it! - - Your post codes and beep codes in the drop down below -

Spoiler

 

 

Link to post
Share on other sites
2 hours ago, podkall said:

the BMW one?

 

Can side effect of not enough of voltage be occasional stuttering when playing game?

I already rolled my silicon dice, it's pretty decent as you can see in my signature.

I was asking because I only OC like half the amount for CPU demanding games and don't wanna hurt it in a longrun if if low voltage was one of possible ways to kill CPU, but thankfuly it isn't.

Blender is not at all a stressful enough application to prove stability. A very unstable OC can pass blender. Unless you have properly stress tested your CPU, you don’t actually know where your chip lands… if it doesn’t pass blender, it’s guaranteed not stable, but passing it does not guarantee it IS stable. 
 

Not enough voltage could be stuttering, or crashing, or BSOD, or memory errors. Plenty of ways for it to “fail” when not stable, which is why properly stress testing is an integral part of the overclocking equation.

 

Low voltage I don’t believe can hurt the hardware, but it can cause corruption of software… no reason to “only OC half of what the CPU is capable of”, just have to actually thoroughly stress test and being the bolts down as low as possible to actually be stable. Assuming temps are ok at that level while in your every day workload, and voltage is within what is considered safe limits, the chip will be fine. My old 9900k ran at 98c while in aida64 FPU stress test for 4 hours, low 90’s with normal default Aida, and low 80’s for blender (blender isn’t very stressful….), but in the 60’s for games. If I was always doing renders or number crunching, I would have backed it off from 5ghz to something lower, but since my main use is gaming, and my voltage was not high but I had a hot running chip while in stress test but temps were fine in gaming, I stuck with 5ghz without issue. Just because the chip is hot in unrealistic stress tests doesn’t mean it’ll be dangerous since you are never in that type of workload...

Rig: i7 10700k @ 5.1Ghz, 4.8 Ring - - Z490 Vision G - - EVGA RTX 2080 XC Ultra @ 2025Mhz - - 4x8GB Vengeance Pro 3000Mhz 15-17-17-34 @ 3500MHz 16-19-19-38 - - Samsung 950 Pro 512 NVMe Boot + Main Programs - - Samsung 830 Pro 256 RAID 0 Lightroom + Photo work - - WD Blue 1 TB SSD for Games - - Corsair RM850x - - Sound BlasterX EA-5 - - EK Supremacy Evo - - XT45 X-Flow 420 + UT60 280 rads - - EK Full Cover GPU Block - - EK XRES RGB PWM - - Fractal Define S2 - - Acer Predator X34 -- Logitech G502 - - Logitech G710+ - - Logitech Z5500 - - Steel Series QcK XXL

 

Headphones/amp/dac: Schiit Lyr 3 - - Fostex TR-X00 - - Sennheiser HD 6xx

 

Homelab/ Media Server: ESXi 6.5 - - 250 GB SSD for VM's/ESXi boot - - FreeNAS - - HPE Proliant ML10 Gen 9 backbone - - i3 6100 - - 28 GB ECC - - 10x4 TB WD Red RAID Z2 - - 10TB WD Red for expendable data - - Corsair 750D - - Corsair RM650i - - Dell H310 6Gbps SAS HBA - - Intel RES2SC240 SAS Expander

 

White Lightning (original full watercooled build) - Homelab / Media Server - The Blue Beast (my car and other expensive hobby...) - iPhone Xs - 2018 MacBook Air

Link to post
Share on other sites

Some things in my tool box for testing:

 

Linpack Xtreme

OCCT

TM5

SuperPi 32M

Some other crap.

 

Not Cinebench lol..

AMD R9 5900X @ + 150MHz | Thermalright Frost Commander 140/NF-A14 iPPC/SYY-157
Asus Strix B550-F Gaming | 4x8GB G.Skill Trident Z Mix @ 1866MHz 14-14-14-34 1T 1.5v
EVGA RTX 3070Ti FTW3 Ultra @ 2085/1212MHz | WD SN750 500GB, 1000GB, 545S 256GB
Seasonic Prime GX-750 | Fractal Design Meshify C | TY-143, TL-D14X, TL-C12 Pro, NF-F12 iPPC

Link to post
Share on other sites
17 hours ago, LIGISTX said:

Blender is not at all a stressful enough application to prove stability. A very unstable OC can pass blender. Unless you have properly stress tested your CPU, you don’t actually know where your chip lands… if it doesn’t pass blender, it’s guaranteed not stable, but passing it does not guarantee it IS stable. 
 

Not enough voltage could be stuttering, or crashing, or BSOD, or memory errors. Plenty of ways for it to “fail” when not stable, which is why properly stress testing is an integral part of the overclocking equation.

 

Low voltage I don’t believe can hurt the hardware, but it can cause corruption of software… no reason to “only OC half of what the CPU is capable of”, just have to actually thoroughly stress test and being the bolts down as low as possible to actually be stable. Assuming temps are ok at that level while in your every day workload, and voltage is within what is considered safe limits, the chip will be fine. My old 9900k ran at 98c while in aida64 FPU stress test for 4 hours, low 90’s with normal default Aida, and low 80’s for blender (blender isn’t very stressful….), but in the 60’s for games. If I was always doing renders or number crunching, I would have backed it off from 5ghz to something lower, but since my main use is gaming, and my voltage was not high but I had a hot running chip while in stress test but temps were fine in gaming, I stuck with 5ghz without issue. Just because the chip is hot in unrealistic stress tests doesn’t mean it’ll be dangerous since you are never in that type of workload...

Well playing BR games gives my CPU quite the effort, basicaly constant 100% utralisation

 

15 hours ago, freeagent said:

Some things in my tool box for testing:

 

Linpack Xtreme

OCCT

TM5

SuperPi 32M

Some other crap.

 

Not Cinebench lol..

Not Cinebench? damn, I thought it was decent since it runs the cpu at highest possible speed and utralisation, and takes little time

like for a quick test could be fine right? not for definitive answer of stability ofc

 

Btw @freeagent @LIGISTX values when Cinebench was running:

obrazek.png.6b415dbf26579f327877c1322adfad07.png

When I give advice I use my experience and google, I know alot, but not everything.

Have a question? - quote me = I'll respond instantly or in a day (if it's longer than 2 days, don't expect me to come back soon)

Specs I have experience with: (highlited build is being used)

Pentium G4500 | 4GB/8GB DDR4 2133Mhz | H110 | GTX 1050

Ryzen 3 1200 3,5Ghz / OC:4Ghz | 8GB DDR4 2133Mhz / 16GB 3200Mhz | B450 | GTX 1050

Link to post
Share on other sites

You should use HWinfo64... its uber gooder.

 

Capture.PNG.b5bf203f92d54e193a07f5906be432a7.PNG

AMD R9 5900X @ + 150MHz | Thermalright Frost Commander 140/NF-A14 iPPC/SYY-157
Asus Strix B550-F Gaming | 4x8GB G.Skill Trident Z Mix @ 1866MHz 14-14-14-34 1T 1.5v
EVGA RTX 3070Ti FTW3 Ultra @ 2085/1212MHz | WD SN750 500GB, 1000GB, 545S 256GB
Seasonic Prime GX-750 | Fractal Design Meshify C | TY-143, TL-D14X, TL-C12 Pro, NF-F12 iPPC

Link to post
Share on other sites
4 hours ago, podkall said:

Well playing BR games gives my CPU quite the effort, basicaly constant 100% utralisation

 

Not Cinebench? damn, I thought it was decent since it runs the cpu at highest possible speed and utralisation, and takes little time

like for a quick test could be fine right? not for definitive answer of stability ofc

 

Btw @freeagent @LIGISTX values when Cinebench was running:

obrazek.png.6b415dbf26579f327877c1322adfad07.png

Just because the CPU is at 100% utilization doesn’t mean it is being heavily worked. It’s all about the actual wattage load being placed on the chip. Games are no where near stressful enough to prove actual stability, and CB isn’t either. It’s just not enough of a wattage draw. If that package power is correct in your screenshot, it’s only pulling 65 watts of power. Try asus real bench, or AIDA64 with only the FPU option checked. 
 

Download link is there. 

https://rog.asus.com/rog-pro/realbench-v2-leaderboard/

Rig: i7 10700k @ 5.1Ghz, 4.8 Ring - - Z490 Vision G - - EVGA RTX 2080 XC Ultra @ 2025Mhz - - 4x8GB Vengeance Pro 3000Mhz 15-17-17-34 @ 3500MHz 16-19-19-38 - - Samsung 950 Pro 512 NVMe Boot + Main Programs - - Samsung 830 Pro 256 RAID 0 Lightroom + Photo work - - WD Blue 1 TB SSD for Games - - Corsair RM850x - - Sound BlasterX EA-5 - - EK Supremacy Evo - - XT45 X-Flow 420 + UT60 280 rads - - EK Full Cover GPU Block - - EK XRES RGB PWM - - Fractal Define S2 - - Acer Predator X34 -- Logitech G502 - - Logitech G710+ - - Logitech Z5500 - - Steel Series QcK XXL

 

Headphones/amp/dac: Schiit Lyr 3 - - Fostex TR-X00 - - Sennheiser HD 6xx

 

Homelab/ Media Server: ESXi 6.5 - - 250 GB SSD for VM's/ESXi boot - - FreeNAS - - HPE Proliant ML10 Gen 9 backbone - - i3 6100 - - 28 GB ECC - - 10x4 TB WD Red RAID Z2 - - 10TB WD Red for expendable data - - Corsair 750D - - Corsair RM650i - - Dell H310 6Gbps SAS HBA - - Intel RES2SC240 SAS Expander

 

White Lightning (original full watercooled build) - Homelab / Media Server - The Blue Beast (my car and other expensive hobby...) - iPhone Xs - 2018 MacBook Air

Link to post
Share on other sites
4 hours ago, podkall said:

Well playing BR games gives my CPU quite the effort, basicaly constant 100% utralisation

 

Not Cinebench? damn, I thought it was decent since it runs the cpu at highest possible speed and utralisation, and takes little time

like for a quick test could be fine right? not for definitive answer of stability ofc

 

Btw @freeagent @LIGISTX values when Cinebench was running:

 

There is no "One" thing that can completely test it, as long as the system is running fine with no errors or crashing it's fine.
The usual suspects are Prime 95, OCCT and such but you have to be careful, some tests you can use are capable of doing damage to a system like Furmark.
Do not use Furmark and if you do, you've been warned.

This is the current and complete list of benchmarks ran at the bot: Overclocking, overclocking, and much more! Like overclocking.
Lots of stuff to try with some of it being free, some not but whatever tickles your fancy you can go for it if you want.

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×