I only get 34k cinabench with a 13900k

[Jackaroe]

I don't think it's throttling all my fans are set to performance in bios my aio is always at 100% and it's a 360mm, my motherboard is nzxt z790 my temps are usually 85-90

[RONOTHAN##]

What motherboard do you have? Odds are there's some sort of power limit in place, since IIRC with a ~250W power limit the 13900K should score ~35K, and 1000 points lower can be explained by a dirty OS. 

[Jackaroe]

Just now, RONOTHAN## said:

What motherboard do you have? Odds are there's some sort of power limit in place, since IIRC with a ~250W power limit the 13900K should score ~35K, and 1000 points lower can be explained by a dirty OS. 

I have an nzxt z790

 

[RONOTHAN##]

6 minutes ago, Jackaroe said:

I have an nzxt z790

 

Check in the BIOS then, there is likely some sort of power limit setup by default since that board shouldn't be having the VRM overheating in Cinebench. 

 

If that doesn't help, maybe try doing a BIOS update as it's not unheard of for a BIOS revision to have significantly worse performance for whatever reason. 

[tdkid]

well the 13900 is a hot running CPU but 85-90 seems like your cooling is not working as well as it should be. maybe reapply the thermal paste. as a suggestion.

[thekingofmonks]

Could you please give us the system specs?

[Mark Kaine]

So but basically?  irs working as intended? 1000 points on a benchmark like cb is nothing...

 

 

can you show us where it says what your cpu should score on average?  cause i don't believe its performing abnormally as for now.

 

 

53 minutes ago, Jackaroe said:

my temps are usually 85-90

hot chip is gonna be hot, yes.

 

 

 

maybe try undervolting/  set a power limit, turn turbo boost off, etc. 

[RONOTHAN##]

Just now, Mark Kaine said:

So but basically?  irs working as intended? 1000 points on a benchmark like cb is nothing...

 

 

can you show us where it says what your cpu should score on average?  cause i don't believe its performing abnormally as for now.

Most people get ~40K in R23 with a 13900K when power limits are disabled. 34K is in 13700K territory (that chip does ~31K stock, ~33K when overclocked fully stable, ~34K when maxed out for Cinebench). 

[Mark Kaine]

10 minutes ago, RONOTHAN## said:

Most people get ~40K in R23 with a 13900K when power limits are disabled. 34K is in 13700K territory (that chip does ~31K stock, ~33K when overclocked fully stable, ~34K when maxed out for Cinebench). 

oh ok,  so power limits need to be disabled?  what a horrible design lol.

 

And then they'll need adequate cooling or the chip will melt. got it.

 

 

 

[Jackaroe]

11 hours ago, tdkid said:

well the 13900 is a hot running CPU but 85-90 seems like your cooling is not working as well as it should be. maybe reapply the thermal paste. as a suggestion.

I don't think it's cooling, my temps are pretty normal for underload if anything a little cooler then what most people's are while.using the 13900k in cinabench 

 

[Jackaroe]

11 hours ago, AndreStray said:

Obviously there having an issue with power supply or Cooling system.

I don't think it's either, my power supply is 1200 watts my temps are pretty normal and if not better then most in every game and cinabench 

[Jackaroe]

10 hours ago, thekingofmonks said:

Could you please give us the system specs?

13900k 

Msi suprim liqud x 4090

Corsair 4 16gb sticks of 6000mhz ddr5

Nzxt z790 motherboard 

Nzxt 1200 watt powersupply 

Nzxt kracken z73 360mm

[Jackaroe]

11 hours ago, RONOTHAN## said:

Check in the BIOS then, there is likely some sort of power limit setup by default since that board shouldn't be having the VRM overheating in Cinebench. 

 

If that doesn't help, maybe try doing a BIOS update as it's not unheard of for a BIOS revision to have significantly worse performance for whatever reason. 

I tried looking for something like that I can't find anything, this is what I was thinking but I spent like a half hour looking and I don't know what it would be called

 

[tdkid]

11 minutes ago, Jackaroe said:

I don't think it's cooling, my temps are pretty normal for underload if anything a little cooler then what most people's are while.using the 13900k in cinabench 

 

with 80-90 temps if you are running like cinabench or any one of the other stress testing programs then i can see that but just normal every day use, no that is not normal.

 

edit: 4 sticks of 16GB RAM? so you are running 64GB? well there is one issue right there on the score. running all 4 DIMM slots limits the speed of the RAM basically to the base speed as far as i know

[RONOTHAN##]

2 minutes ago, Jackaroe said:

I tried looking for something like that I can't find anything, this is what I was thinking but I spent like a half hour looking and I don't know what it would be called

 

It would be the PL1 and PL2 settings. As for where they would be, I haven't used an NXZT board so I don't really know. From what I know they do use a very similar BIOS to ASRock, and those settings IIRC are in the overclocking tab under CPU and towards the bottom of that menu, but even then it's been quite a while since I've used an ASRock Intel board so I'm very fuzzy on specific setting locations. 

[DANK_AS_gay]

8 minutes ago, tdkid said:

edit: 4 sticks of 16GB RAM? so you are running 64GB? well there is one issue right there on the score. running all 4 DIMM slots limits the speed of the RAM basically to the base speed as far as i know

That's false. It is also highly dependant on setup. So long as all the sticks of RAM are the same brand and kit they should be fine.

[tdkid]

11 minutes ago, DANK_AS_gay said:

That's false. It is also highly dependant on setup. So long as all the sticks of RAM are the same brand and kit they should be fine.

you might want to watch some tech reviewers like jaystwcents, linus gamers nexus and many others as they all say the same thing that the DIMM slots all run the same and that you are dividing the power to each stick, the frequency and the overall speed. 

[Jackaroe]

That's my cinabench scores for games it's like normal 50-60, my ram is set to 5200mhz so it is more stable, everything with my RAM is good I had same scores when I was running 2 sticks, and I heard RAM doesn't really effect cinabench to much

[tdkid]

8 minutes ago, Jackaroe said:

That's my cinabench scores for games it's like normal 50-60, my ram is set to 5200mhz so it is more stable, everything with my RAM is good I had same scores when I was running 2 sticks, and I heard RAM doesn't really effect cinabench to much

well everything effects those stress test programs. if the sticks dont effect the score than getting 2 or more sticks of 32Gb or 64Gb each would increase the score by a lot as far as i can tell but i dont stress my PC. 

[DANK_AS_gay]

18 minutes ago, tdkid said:

you might want to watch some tech reviewers like jaystwcents, linus gamers nexus and many others as they all say the same thing that the DIMM slots all run the same and that you are dividing the power to each stick, the frequency and the overall speed. 

He's running a z790 motherboard, it's not going to have RAM power issues

[tdkid]

16 minutes ago, DANK_AS_gay said:

He's running a z790 motherboard, it's not going to have RAM power issues

he is running

13900k 

Msi suprim liqud x 4090

Corsair 4 16gb sticks of 6000mhz ddr5

Nzxt z790 motherboard 

Nzxt 1200 watt powersupply 

Nzxt kracken z73 360mm

 

but you can easily see the RAM issue when you have a cheap laptop only running 4GB and upgrade it to 8+. not only because of speed and space but you will able to do more

[RONOTHAN##]

1 hour ago, tdkid said:

you might want to watch some tech reviewers like jaystwcents, linus gamers nexus and many others as they all say the same thing that the DIMM slots all run the same and that you are dividing the power to each stick, the frequency and the overall speed. 

First off, those issues reported were either a) from the early days of DDR5 where the training algorithms for 4x16GB configs were borderline broken or b) talking about 4x32GB configs where it's significantly harder on the memory controller than the 4x16GB configs OP is running. It's still more finicky than 2x32GB so it isn't optimal, but it's not unusable like it was initially and for the most part just works. 

 

As for everything else you said, I have no idea what you're going on about. Each stick gets the same amount of power whether it's in 2x16GB or 4x16GB. 

[unclewebb]

@Jackaroe

 

Run HWiNFO and see what the PL1 and PL2 turbo power limits are set to. The default values will result in significantly reduced performance. Most decent motherboards have an unlimited setting which sets both power limits to the max, 4095W. 

 

[DANK_AS_gay]

18 minutes ago, RONOTHAN## said:

First off, those issues reported were either a) from the early days of DDR5 where the training algorithms for 4x16GB configs were borderline broken or b) talking about 4x32GB configs where it's significantly harder on the memory controller than the 4x16GB configs OP is running. It's still more finicky than 2x32GB so it isn't optimal, but it's not unusable like it was initially and for the most part just works. 

 

As for everything else you said, I have no idea what you're going on about. Each stick gets the same amount of power whether it's in 2x16GB or 4x16GB. 

Thank you, I haven't looked into RAM stuff but I didn't think what he said was the case. I'm glad someone more knowledgeable than myself could step in.