Overclocked Or Not?

[Flamethrower1972]

In X1 I have my memory OC to 5301Mhz and my Graphics OC to 2021Mhz. In X1 it shows that and when I run HWMonitor it shows the two over clocks but when I run HWINFO64 v7.72-5355 it never shows the over clocks so what is going on? The 2 1080 TIS are only used folding.

 

[EarthDog]

 

[Flamethrower1972]

Sorry about that.

 

[EarthDog]

... how about a picture of the problem application.

 

[Flamethrower1972]

... how about a picture of the problem application.

Here you go.

 

[EarthDog]

I see both applications showing the same clocks... ~1950... what X1 shows you set it to. It won't be exact.

Post a screenshot of gpuz that includes the first/main page and the sensors tab while its under load.

Also note, it's hitting the power limit (normal) and running hot (normal) so it will sometimes drop boost clocks t9 maintain the temp/power.

 

[Woomack]

Throttling ... throttling everywhere ... just don't expect the card to boost higher when you try to overclock it while there is thermal, power, and utilization throttling/limit at once.

 

[Flamethrower1972]

Throttling ... throttling everywhere ... just don't expect the card to boost higher when you try to overclock it while there is thermal, power, and utilization throttling at once.

Is it too much voltage?

 

[Woomack]

I don't remember if you can even adjust the voltage on this card. 1.05V was a limit made by Nvidia. The hot spot is close to 100°C. It will limit OC regardless of the core temp.
1. You won't OC this card on air/stock much as it runs hot
2. OC won't give you much
3. I assume you use it for folding; then, a more efficient way is to set a lower power limit and check how much PPD will drop. There is a chance that when you lower the power limit, the card will run cooler, and if you set a higher frequency, then it will boost higher. It works much better on newer cards, and I wasn't folding on 1080/Ti for a long time, so you have to check it on your own. I can only tell you that on RTX30/40, you can drop the power limit/power draw by 20% and keep almost the same PPD as at stock settings.
4. I see your card's 12V rail goes down to 11.7V. Or the PSU is heavily loaded, or it will end in some time. I just replaced the 1200W Corsair as it was going down to 11.4-11.5V under a 200-300W load. Once it was going under 11.5V, the PC was restarting. 11.7V is not tragic, but I would keep an eye on that too.

 

[Flamethrower1972]

I don't remember if you can even adjust the voltage on this card. 1.05V was a limit made by Nvidia. The hot spot is close to 100°C. It will limit OC regardless of the core temp.
1. You won't OC this card on air/stock much as it runs hot
2. OC won't give you much
3. I assume you use it for folding; then, a more efficient way is to set a lower power limit and check how much PPD will drop. There is a chance that when you lower the power limit, the card will run cooler, and if you set a higher frequency, then it will boost higher. It works much better on newer cards, and I wasn't folding on 1080/Ti for a long time, so you have to check it on your own. I can only tell you that on RTX30/40, you can drop the power limit/power draw by 20% and keep almost the same PPD as at stock settings.
4. I see your card's 12V rail goes down to 11.7V. Or the PSU is heavily loaded, or it will end in some time. I just replaced the 1200W Corsair as it was going down to 11.4-11.5V under a 200-300W load. Once it was going under 11.5V, the PC was restarting. 11.7V is not tragic, but I would keep an eye on that too.

I put the 2 cards back to stock.

 

[Woomack]

My point 4 can be a wrong reading from the software/motherboard/graphics card. It is better to check it directly on the PSU with a multimeter when the PC is under load. I just wanted to add it, as everything is probably fine with the PSU.

 

[Flamethrower1972]

My point 4 can be a wrong reading from the software/motherboard/graphics card. It is better to check it directly on the PSU with a multimeter when the PC is under load. I just wanted to add it, as everything is probably fine with the PSU.

I only have a VRM no clamp type voltage reader. Would it be the GPU connectors or the PCIE ??

 

[Woomack]

I only have a VRM no clamp type voltage reader. Would it be the GPU connectors or the PCIE ??

The easiest one is probably 4-pin molex, but it can be CPU, PCIe/GPU or a 24 pin mobo connector too. Since every modern PSU is a single rail one then it shouldn't make a difference where you check it.

 

[Flamethrower1972]

The easiest one is probably 4-pin molex, but it can be CPU, PCIe/GPU or a 24 pin mobo connector too. Since every modern PSU is a single rail one then it shouldn't make a difference where you check it.

Ok thank you.

 

[Flamethrower1972]

I used a power supply tester, and all the voltages were right where they should be.

 

[EarthDog]

I used a power supply tester, and all the voltages were right where they should be.

What power supply tester? How did you test, exactly (under load, too...right??)? Just want to make sure you did it right.

 

[Flamethrower1972]

The one that only has leds on it. I have used them before, and they work very well to me.

 

[EarthDog]

The one that only has leds on it.

This isn't helpful. Please list the brand/model or link it...

I have used them before, and they work very well to me.

That's great. But I'm asking how exactly you tested. Did you load up the PSU or leave it idle?

 

[Flamethrower1972]

No load on the psu and the tester has a plug for each pin set example 24 pin 8 pins etc....

 

[EarthDog]

Again, please list the brand/model or link the power supply tester...describing the ports it and others like all have, also isn't helpful.

No load on the psu

Glad I asked... if your problem is while the unit is under load, how does idle testing help you?????? Woomack mentioned in post 11 already you need to test under load. Start folding and see what it says...though if it's just dummy lights, you won't get a voltage reading (why we suggested a DMM).