[SOLVED] My pc is massively underperforming

[trogloditul]

UserBenchmarks: Game 13%, Desk 39%, Work 10%
CPU: Intel Core i5-4670K - 25%
GPU: AMD RX 480 - 43.7%
SSD: Kingston SM2280S3 SATA M.2 120GB - 52.4%
HDD: Hitachi HUP722020APA330 2TB - 53%
RAM: Unknown CL11-11-11 D3-1600 4x4GB - 26.6%
MBD: MSI Z97 GAMING 7 (MS-7916)

So these are my scores guys. I have absolutely no idea why this is happening but everything is slow and annoying. I tried installing a fresh win10 cause i read on different forums that this might be an issue but it has barely improved. Any tips? I'm desperate.

 

[trogloditul]

UserBenchmarks: Game 22%, Desk 49%, Work 17%
CPU: Intel Core i5-4670K - 38.5%
GPU: AMD RX 480 - 48.1%
SSD: Kingston SM2280S3 SATA M.2 120GB - 63.9%
HDD: Hitachi HUP722020APA330 2TB - 52.5%
USB: Integral Neon USB 3.0 16GB - 11.7%
RAM: Unknown CL11-11-11 D3-1600 4x4GB - 44.7%
MBD: MSI Z97 GAMING 7 (MS-7916)


update i was able to get it up a bit by installing all of the latest drivers for everything and updating bios but still looks pretty..rubbish.

 

[kurdtnz]

can you post a link/picture of your userbenchmark results?

 

[Schlachtwolf]

Userbenchmark is poo, Run Cinebench to give a much better idea of what your CPU is doing, 3D Mark if you have it for your GPU.

Userbenchmark give you a % of what it is capable of against the BEST out there, it is not saying your CPU is running at 48.1% of it's capability but is only capable of 48.1% of what say an Ryzen 5950x can do. That is why the whole UB is misleading and you think your PC is running poo but it is not in reality.

Your SSD and HDD are according to UB "Performing as expected" correct ? The GPU is an older tech and is going to get slaughtered by being measured to an RTX 3090...... you can't compare DDR3 ram 1600mhz to DDR4 C16 3600mhz !!!! BUT USERBENCHMARK does !!!!!

Check your FPS in games, google what others get with similar setups and you will in all likelyhood see you PC is running just fine.

Userbenckmark and Bottleneck calculators are just plain no use to anyone, misleading and dumb.

Just as an example of the rubbish Userbenchmark come up with and I quote as an alternative to the Ryzen 5900x the suggest.... "Gamers that do not wish to pay “marketing fees” should investigate Intel’s $190 USD i5-9600K, the saved $370 USD would be far better spent on a higher tier GPU" ......... OK enough of that.......... I am not an AMD fanboy I also have Intel 9th gen i5 and i9 setups in the office so don't start on me !!!!

 

[trogloditul]

can you post a link/picture of your userbenchmark results?

MSI Z97 GAMING 7 (MS-7916) Performance Results - UserBenchmark

 

[trogloditul]

Userbenchmark is poo, Run Cinebench to give a much better idea of what your CPU is doing, 3D Mark if you have it for your GPU.

Userbenchmark give you a % of what it is capable of against the BEST out there, it is not saying your CPU is running at 48.1% of it's capability but is only capable of 48.1% of what say an Ryzen 5950x can do. That is why the whole UB is misleading and you think your PC is running poo but it is not in reality.

Your SSD and HDD are according to UB "Performing as expected" correct ? The GPU is an older tech and is going to get slaughtered by being measured to an RTX 3090...... you can't compare DDR3 ram 1600mhz to DDR4 C16 3600mhz !!!! BUT USERBENCHMARK does !!!!!

Check your FPS in games, google what others get with similar setups and you will in all likelyhood see you PC is running just fine.

Userbenckmark and Bottleneck calculators are just plain no use to anyone, misleading and dumb.

Just as an example of the rubbish Userbenchmark come up with and I quote as an alternative to the Ryzen 5900x the suggest.... "Gamers that do not wish to pay “marketing fees” should investigate Intel’s $190 USD i5-9600K, the saved $370 USD would be far better spent on a higher tier GPU" ......... OK enough of that.......... I am not an AMD fanboy I also have Intel 9th gen i5 and i9 setups in the office so don't start on me !!!!

Damn boy I had absolutely no idea that that's what userbenchmark does. thanks! Will update with results from cinebench

 

[kurdtnz]

Your cpu background usage is very high, are you running other programmes/ have other windows open when running userbenchmark?

 

[trogloditul]

It states that the cpu background usage is high altough if i open up my task manager the cpu usage is very low at all times(max 15%) and the temperatures are fine

 

[Phaaze88]

Yes, userbenchmark is poo, if looking to evaluate performance.
It does have its uses in the troubleshooting process, as much as some of us hate it, be it the developer, the software, or both.

1)You got flagged for very high background usage - this could be as simple as leaving the browser running while the benchmark is too.
This reduces benchmark accuracy. Ideally, it should be under 10% for the best accuracy.

2)The cpu's turbo clock is... wow. Either the cpu is thermal throttling, or the chipset drivers haven't been installed.
You haven't mentioned thermals at all.

 

[trogloditul]

Yes, userbenchmark is poo, if looking to evaluate performance.
It does have its uses in the troubleshooting process, as much as some of us hate it, be it the developer, the software, or both.

1)You got flagged for very high background usage - this could be as simple as leaving the browser running while the benchmark is too.
This reduces benchmark accuracy. Ideally, it should be under 10% for the best accuracy.

2)The cpu's turbo clock is... wow. Either the cpu is thermal throttling, or the chipset drivers haven't been installed.
You haven't mentioned thermals at all.

So I am using a software called NZXT Cam to see my components temperatures. I've observed it while idling for around 15 mins when idling and the CPU is staying at around 25 degrees with the load fluctuating in between 5%-15%(i saw windows defender running at some point so i'm assuming that's why it fluctuated) with the clock running at 800 Mhz constantly and the fan speed at 1109 Mhz at all times, never moved a bit in these 15 mins.

as of thermal throttling, I have absolutely no idea what that means. I can confirm that I installed all the drivers from my motherboard's website which includes the intel chipset driver.
Worth mentioning is that it's the first time when I have installed a processor on a motherboard, so i tried as much as possible to stick to the reccomendations when putting the thermal paste and everything else. My cooler is a x31 nzxt kraken.

 

[lga1156_ftw]

As Phaaze88 said, its easiest way for fast troubleshooting and comparing your components like cpu/gpu to see if you get performance similar to others (with same cpu/gpu). You cant however compare amd vs intel at all on that site since it relies on single thread performance. There is something very wrong , i would guess thermal throttling like other people in this thread too.

People here like to call userbenchmark crap if they have amd and i get it, its biased or atleast used to be. I mostly use it to quickly check what laptop processor or gpu i can upgrade since you can see all the possible combinations people got them working. I only game so any other benchmark is basically uselesss to me, none of the games i play use more than 8c / 8threads so userbenchmark gives "close enough" results, i actually had way less fps with setups DOUBLE the core/thread count which gave me insane scores in other benchmarks with WORSE game performance.

 

[trogloditul]

Yes, userbenchmark is poo, if looking to evaluate performance.
It does have its uses in the troubleshooting process, as much as some of us hate it, be it the developer, the software, or both.

1)You got flagged for very high background usage - this could be as simple as leaving the browser running while the benchmark is too.
This reduces benchmark accuracy. Ideally, it should be under 10% for the best accuracy.

2)The cpu's turbo clock is... wow. Either the cpu is thermal throttling, or the chipset drivers haven't been installed.
You haven't mentioned thermals at all.

update: MSI Z97 GAMING 7 (MS-7916) Performance Results - UserBenchmark here is the test without the browser running

 

[lga1156_ftw]

download aida64 extreme trial and open "tools" tab then click "system stability test" , start the test it will give you "thermal throttling" warning already in few secs if cpu overheating is the problem, stop the test immiddiately if that happens

 

[trogloditul]

download aida64 extreme trial and open "tools" tab then click "system stability test" , start the test it will give you "thermal throttling" warning already in few secs if cpu overheating is the problem, stop the test immiddiately if that happens

https://imgur.com/QCMIFCj

View: https://imgur.com/QCMIFCj

 

[Phaaze88]

CAM... ~Ohh, now the high cpu usage makes some sense.

https://imgur.com/QCMIFCj

View: https://imgur.com/QCMIFCj

Then it either means CAM is borked and giving you the wrong readings, or everything is 'fine' until you put the cpu under load.
In either scenario, the cooler is a problem.

How long has it been in use?
Can you actually feel the liquid moving throughout the unit?

 

[trogloditul]

Been in use less than an year, installed it myself. As for the liquid excuse my stupidness but..what liquid?

 

[Phaaze88]

The liquid in the cooler. The pump is supposed to keep it moving.
-Dead pump
-Clog from loose debris or gunk buildup
-Air trapped in the pump
All 3 of these can cause cpu throttling, and one of the best ways to tell in the case of unreliable software, is to get your hands in there, and feel the cpu block, both tubes, and the radiator.
The heat should be spread out through most of the unit, and not focused in the cpu block itself.

 

[trogloditul]

Everything feels pretty sound tbh but I do have an ancient cooler that I used before I got this one and it worked until replacement. I'll try swapping just to see if the x31 is faulty and update with the results tomorrow, now I have to leave for work.

 

[trogloditul]

One more thing until tomorrow, is the cpu clock speed supposed to stay constant no matter what am I doing on the computer? Even when i'm running a game is always at 800 mhz.

 

[Phaaze88]

is the cpu clock speed supposed to stay constant no matter what am I doing on the computer?

That only really applies when overclocking, otherwise they(frequency, power, voltage, etc) are supposed to change dynamically.

 

[trogloditul]

So then there might be a software issue? Cause I have been looking at the frequency and fan speed and they are always exactly the same, 800mhz no matter what i'm doing or how much strain i put on the cpu and the fan at 1103 rpm

 

[Phaaze88]

Fan speeds should be adjustable via CAM or bios.

When you say that it sits at 800mhz, that makes me think that the chipset drivers weren't updated, or Windows wasn't clean installed since your last motherboard change.

 

[trogloditul]

So I've done some digging and I found out that my motherboard and some other MSI mobos have a SLOW_MODE switch on the motherboard which limits your cpu to 8x multiplier.. Mine was turned on and after I turned it off and reset the bios to default settings and it still does the same thing. I saw that if you turn off the BD PROCHOT it fixes the issue and everything runs perfectly but I'm not 100% how safe is that. Also read that this might happen because of a bad heat sensor on the mobo.

 

[Phaaze88]

Geezus, they still haven't done away with that Slow Mode feature?

There's BD Prochot and there's also Prochot.
"PROCHOT (processor hot) is triggered by the CPU temperature. BD PROCHOT is never triggered by the CPU temperature. PROCHOT and BD PROCHOT sound the same but they are two different signals. BD PROCHOT can be triggered by ANY external sensor outside of the CPU. It could be the GPU but it could be ANYTHING. It could even be a keyboard temperature sensor. Someone might have decided that it would be a good idea to protect your fingers from getting too hot. The engineers that designed your hardware will know what sensors are connected to the BD PROCHOT signal line but you are not going to find this information in any forum. Anything to do with BD PROCHOT is rarely publicly documented.

Your CPU will still thermal throttle and slow down if it gets too hot whether BD PROCHOT is checked or not. The PROCHOT signal is generated within the CPU. PROCHOT always works regardless of BD PROCHOT.

If an external sensor has failed, it can constantly send throttling signals to the CPU. That is not good. Disabling BD PROCHOT allows the CPU to run at its full rated speed without being slowed down by a bad sensor.

Without knowing what sensor is triggering BD PROCHOT throttling, no one can guarantee you that disabling BD PROCHOT is 100% safe. If you do not feel comfortable disabling BD PROCHOT then leave it alone."

 

[trogloditul]

Geezus, they still haven't done away with that Slow Mode feature?

There's BD Prochot and there's also Prochot.
"PROCHOT (processor hot) is triggered by the CPU temperature. BD PROCHOT is never triggered by the CPU temperature. PROCHOT and BD PROCHOT sound the same but they are two different signals. BD PROCHOT can be triggered by ANY external sensor outside of the CPU. It could be the GPU but it could be ANYTHING. It could even be a keyboard temperature sensor. Someone might have decided that it would be a good idea to protect your fingers from getting too hot. The engineers that designed your hardware will know what sensors are connected to the BD PROCHOT signal line but you are not going to find this information in any forum. Anything to do with BD PROCHOT is rarely publicly documented.

Your CPU will still thermal throttle and slow down if it gets too hot whether BD PROCHOT is checked or not. The PROCHOT signal is generated within the CPU. PROCHOT always works regardless of BD PROCHOT.

If an external sensor has failed, it can constantly send throttling signals to the CPU. That is not good. Disabling BD PROCHOT allows the CPU to run at its full rated speed without being slowed down by a bad sensor.

Without knowing what sensor is triggering BD PROCHOT throttling, no one can guarantee you that disabling BD PROCHOT is 100% safe. If you do not feel comfortable disabling BD PROCHOT then leave it alone."

Marked this as the best solution as that's the only way to get the cpu working properly. Last thing in this thread, now that we know that there is a sensor f'ed up somewhere, does it mean that it's a small motherboard issue? Should I get another mobo?