What's new
Carbonite

South Africa's Top Online Tech Classifieds!
Register a free account today to become a member! (No Under 18's)
Home of C.U.D.

Intel 9900k Owners Club

That's a massive difference. As I said something on your new system is holding you back. Redo Windows and have another crack at it, something ain't right =.
Play the windows game... LOL I dont like that game...
 
Firestrike shouldn't even be used anymore, only focus on Time Spy. It is very very accurate, go and check Red Dead Redemption 2 for example. Go and see how the 2080 and 2080 Ti is running away from the 1080 Ti. Newer games, especially DX12 is a lot better on RTX 2xxx.

Even my 1080 Ti gets 30800 on Fire Strike with a measly overclock, check the link I posted above, here it is again : I scored 25 310 in Fire Strike

I would say Yes AND no. I haven't played a game that implemented both show dx12 strictly outperforming dx11 on everything. In cases where dx12 gave higher average fps, it would also add stuttering or lower min 1% fps. I stuck with dx11 in most games because dx11 provided a more stable framerate and less lag or stuttering effect. Also, Ray Tracing is OUT OF THE WINDOW for now....it downgrades the performance for me and gives a very unstable gameplay so I don't even bother. For new technology to truly benefit from dx12, I believe the game needs to be developed from the get go with dx12 and not have updates to implement dx12 gameplay at a later stage because the boost in performance is not THAT surprisingly huge....well this is just my findings, I don't mind to be wrong but I personally experience this with games I play. I do agree that future games will be better that is purely dx12 development, but they won't just cut out dx11 in all development due to sales lost they will experience world wide. Maybe from 2020 things will change.
 
Last edited:
Would you say my score of 10253 is on par? AS I think I got a better score on the graphics side with my 8700k, byt that might have been on a clean install as well.... Not sure if that would make such a big difference....

9900k + 2080 Score

8700k + 2080 Score

The difference is even bigger than I first realized, I notice you have a great OC running on the 2080 on your 9900k system. That thing should score like 11.5k. You're way out. Do a clean install!

Also be sure to update to your latest bios while you're at it.
 
Last edited:
The difference is even bigger than I first realized, I notice you have a great OC running on the 2080 on your 9900k system. That thing should score like 11.5k. You're way out. Do a clean install!

Also be sure to update to your latest bios while you're at it.
I agree with @baasgene something is holding you back @HNO3. You have very much the same OC on your 2080 than I have as I also clock 2115MHz on the GPU clock and I do 12k + on Timespy so you should also be doing 12k + score atleast with that OC settings on the 9900K system. I will definitely try the 5.0GHz soon as I saw yesterday getting the colder temps in the evening now with winter approaching that my max volts for all cores dropped down to 1.3Volts on Air cooler so I should be running similar clocks and score on the CPU soon.
 
I agree with @baasgene something is holding you back @HNO3. You have very much the same OC on your 2080 than I have as I also clock 2115MHz on the GPU clock and I do 12k + on Timespy so you should also be doing 12k + score atleast with that OC settings on the 9900K system. I will definitely try the 5.0GHz soon as I saw yesterday getting the colder temps in the evening now with winter approaching that my max volts for all cores dropped down to 1.3Volts on Air cooler so I should be running similar clocks and score on the CPU soon.
I will be playing the windows game this weekend and do a fresh clean install and wipe all my HDD etc. to ensure I start from scratch etc.
 
I agree with @baasgene something is holding you back @HNO3. You have very much the same OC on your 2080 than I have as I also clock 2115MHz on the GPU clock and I do 12k + on Timespy so you should also be doing 12k + score atleast with that OC settings on the 9900K system. I will definitely try the 5.0GHz soon as I saw yesterday getting the colder temps in the evening now with winter approaching that my max volts for all cores dropped down to 1.3Volts on Air cooler so I should be running similar clocks and score on the CPU soon.
So I did a clean install on windows and wiped all my HDD except my backup drives, results doesnt seem to have improved, running everything at stock for now.
 
So I did a clean install on windows and wiped all my HDD except my backup drives, results doesnt seem to have improved, running everything at stock for now.

Try switching windows to high performance power mode, for some reason that made a difference for me. But it's very unlikely to have that same issue, but just give it a try anyway.

Download GPU-Z and just check if you're running PCI Express Gen 3.

Are you running the latest bios?

What's troubling to me is that it seems to be an overall system slow down, the cpu score is around 7-8% less than the average 5ghz 9900k, and the gpu is 15-20% off. That's massive and cannot be attributed to normal variance.

Would you mind running Time Spy Extreme and FireStrike standard as well? Just to compare those too.
 
Last edited:
So I did a clean install on windows and wiped all my HDD except my backup drives, results doesnt seem to have improved, running everything at stock for now.
What mobo are you running?
 
Try switching windows to high performance power mode, for some reason that made a difference for me. But it's very unlikely to have that same issue, but just give it a try anyway.

Download GPU-Z and just check if you're running PCI Express Gen 3.

Are you running the latest bios?

What's troubling to me is that it seems to be an overall system slow down, the cpu score is around 7-8% less than the average 5ghz 9900k, and the gpu is 15-20% off. That's massive and cannot be attributed to normal variance.

Would you mind running Time Spy Extreme and FireStrike standard as well? Just to compare those too.
I will run those settings for you once I get a gap at about 16h30, loadshedding is starting at 17h00 as well.

Here is my score on a clean install of windows EVERYTHING stock, not a single Mhz of OC.

 
I will run those settings for you once I get a gap at about 16h30, loadshedding is starting at 17h00 as well.

Here is my score on a clean install of windows EVERYTHING stock, not a single Mhz of OC.


Cpu score is sharp now. so there must've been some current limitations or thermal throttling on your 5ghz OC. No matter, the cpu seems fine, so the only culprit now is the GPU. As I said I would like you to confirm PCI-E Gen 3 and Windows on High Performance.
 
I agree your cpu is 100% on stock, but bliksem your gou is underperforming by at least 15%....something is not so ayoba.
That mobo is a sweet one and you are running the same Z390 chipset as I do. You should be hitting anything between 11000 - 11500 score on the stock gpu settings. Do yourself a favour, check with HWINFO after you did your Timespy benchmark that your gpu hits the required clock speed, voltages and check your thermals as well.
 
Here with a suggestion. Use dynamic voltage!

I'll post my Gigabyte settings tonight, managed to get 5ghz stable with temps at 81c on realbench, and as low as 64c on OCCT Medium with AVX2 (peaks for a second at 74c and drops back down to 64c)

When you use dynamic voltage you can set the LLC to normal on Gigabyte for example, which I believe eliminates most of the heat. On average I would get 87-90c on Realbench after 30 min, I now peak at 79-81c. It's a huge difference. And to me those are completely safe for a 24/7 system, especially since the temps won't even touch 70c on games (in my case like 50s because of 4k)

Nice thing about dynamic voltage of course is the fact that I was checking the voltage on hwmonitor whilst playing Ni No Kuni 2 for example, and the voltage was averaging between 1.0 and 1.1v, sometimes spiking for a moment to 1.24v. On Realbench it would sit at 1.284v, and OCCT at 1.3 (or just below 1.3).

It can peak for a second to 1.32v but never when it's under proper load (only reaches 1.32v on MHW really), and with the LLC on normal there's no more worrying about voltage overshoot if I'm not mistaken.

I don't see any drawback of dynamic voltage, I can't recall my exact settings but I will post them tonight. Think it's vcore on normal (1.2v) with dynamic at +0.040v and cpu internal AC/DC on power savings. Also make sure all your safety features like EIST and Speed stepping is enabled. Most important of course is to set your LLC to normal, not sure if that's level 2 on Asus, but it's the second lowest setting on Gigabyte.

Scores are still in line for 5ghz.

CPU-Z - 592 Single and 6010 Multi
Cinebench - 2157
Timespy - 12088
 
Last edited:
Here with a suggestion. Use dynamic voltage!

I'll post my Gigabyte settings tonight, managed to get 5ghz stable with temps at 81c on realbench, and as low as 64c on OCCT Medium with AVX2 (peaks for a second at 74c and drops back down to 64c)

When you use dynamic voltage you can set the LLC to normal on Gigabyte for example, which I believe eliminates most of the heat. On average I would get 87-90c on Realbench after 30 min, I now peak at 79-81c. It's a huge difference. And to me those are completely safe for a 24/7 system, especially since the temps won't even touch 70c on games (in my case like 50s because of 4k)

Nice thing about dynamic voltage of course is the fact that I was checking the voltage on hwmonitor whilst playing Ni No Kuni 2 for example, and the voltage was averaging between 1.0 and 1.1v, sometimes spiking for a moment to 1.24v. On Realbench it would sit at 1.284v, and OCCT at 1.3 (or just below 1.3).

It can peak for a second to 1.32v but never when it's under proper load, and with the LLC on normal there's no more worrying about voltage overshoot if I'm not mistaken.

I don't see any drawback of dynamic voltage, I can't recall my exact settings but I will post them tonight. Think it's vcore on normal (1.2v) with dynamic at +0.035v and cpu internal AC/DC on power savings. Also make sure all your safety features like EIST and Speed stepping is enabled. Most important of course is to set your LLC to normal, not sure if that's level 2 on Asus, but it's the lowest setting on Gigabyte.

Scores are still in line for 5ghz.

CPU-Z - 592 Single and 5960 Multi
Cinebench - 2152
Timespy - 12020

So by using dynamic/adaptive voltage and a lower LLC you basically dropped the peak voltage for stress tests which resulted in cooler temperatures.

This also means that in games that don't use the CPU fully it will dynamically lower voltages.
Is it still running at 5Ghz in games or does it drop the clocks in line with the lower 1.1v ?

Have you set your PC to high performance mode in the Windows power options ?
 
Yip I set it to high performance, which is also why I like Dynamic because even though the speed is constant @ 5ghz, the vcore is still fluctuating according to load which I honestly didn't expect.

I'm impressed, it's pretty cool.

Monster Hunter World Iceborne was a very weird experience for me, I could play other games just fine on a fixed voltage of 1.31v, OCCT gave zero errors, and Realbench was stable. But MHW crashed to destkop like 5 minutes into playing it with a fixed vcore of 1.31v (no bluescreen, just plain crash to desktop with no error)

To get it stable I had to up vcore to 1.33v which to me made no sense, because that game wasn't loading much on the cpu at all @ 4k.

I'm happy to say at dynamic settings it doesn't crash at all, but there are times that the vcore is bumped to 1.32v whilst playing monster hunter, so it's as if dynamic is preventing the crash by doing that. Other than that it also fluctuates between 1.0 and 1.25v whilst playing MHW, with the occasional spike to 1.32v.

Temps can spike for a second to 64c (probably in line with that vcore spike, maybe a game coding error?), but stays in the low 50s because it really doesn't tax the cpu much like most games at 4k.

Just to note that the spike on MHW was definitely there on my 9700k too (had a Asus mobo then), so not related to the OC at all. It's on DX12 mode though, which could explain a few things. I know devs have always had a tough time to implement perfect dx12 support.
 
Last edited:
So what cache ratio are you guys running ? I'm at 47x at the moment, wondering if its worth pushing it a little higher.
 

Users who are viewing this thread

Similar threads

[Wanted]
  1. LGA 1151
i9 9900k cpu only
  • Location:
    1. Cape Town (Northern Suburbs)
Replies
0
Views
58
[For Sale]
  1. LGA 1151
Intel Celeron Processor G3900
  • Location:
    1. Cape Town (Atlantic Seaboard)
Replies
1
Views
49
[For Sale]
  1. LGA 1200
i5- 10500T
  • Location:
    1. Hartebeespoort
    2. Johannesburg
    3. Pretoria
    4. Rustenburg
Replies
1
Views
124
[For Sale]
  1. LGA 1151
Intel Core i5 8500 CPU
  • Location:
    1. Mpumalanga
Replies
2
Views
113
Back
Top Bottom