[SOLVED] Weird Firestrike score / Physics score

[m.bsatgini]

Hi,
I just switched my Unstable 9900KF to a stable 9900k The KF used to run at 4.8 for 1.24 Vcore, the 9900K is running at 5Ghz for 1.25 stable (did a 1h occt test no issues)

I know firestrike isn't the best for benching and I should use timespy but it's for having a basis to compare.


I don't understand why my firestrike score is so different though,


the same OC has been applied on the gpu
Gsync is off
Vsync is off
power maxed plan on the gpu
Cpu oc with +1000 memory, 100 clock, 100 voltage,
With the former KF Cpu at 4.8 : https://www.3dmark.com/fs/22647094
3DMark Score 29420
Graphics Score 39102
Physics Score 25038
Combined Score 11340
Graphics Test 1 193.97 fps
Graphics Test 2 151.32 fps
Physics Test 79.49 fps
Combined Test 52.75 fps

with the new K cpu at 5gh : https://www.3dmark.com/fs/22761024
3DMark Score 28308
3DMark Score 28740
Graphics Score 38592
Physics Score 23561
Combined Score11119
Graphics Test 1 189.1 fps
Graphics Test 2 150.8 fps
Physics Test 74.8 fps
Combined Test51.72 fps


Can someone explain me why this difference? All scores seems lower than with a less clocked CPU

I've updated the driver and did the last windows update too can that explain the gap?.

Also is that score "good" for my rig? i've also done a timespy : https://www.3dmark.com/spy/12247134

Thanks

 

[m.bsatgini]

https://www.3dmark.com/compare/fs/22765409/fs/22647094

 

[apesoccer]

They are the same processor minus the onchip gpu, so it should definitely be more capable. That said, what are your temps (should be lower then the old one), and are you seeing ghz drops... Are your clocks locked in at 5?

Also...I'd grab an old ssd, unplug your current set of drive(s), reinstall Win7 64 on it, and rerun it specifically for testing purposes. Only update what you need, and don't install anything other then what you need...

If you don't want to spend 2.5 hrs doing that, you could clean out your graphics drivers and reinstall, include shutting off any anti-virus util's or gaming crap running in the background.

 

[m.bsatgini]

I have the K now, KF was before but the chip was garbage.

The temps are all good, around 70c max for what i saw
Here are my bios settings below.

i don't feel like reinstalling windows though but i've done a clean reinstall of the GPu driver. Apprently the microcode i have on this chip for what i read is a bit less efficiant "C6 is a very recent microcode update which is why you get lower performance at higher clock speeds compared to your 9900kf. I assume that your 9900kf was running earlier microcode which means that your 9900k is performing normally. "

But i guess those results are not bad though?

firestrike ; : https://www.3dmark.com/fs/22765409
timespy : https://www.3dmark.com/spy/12247134


I will try by running max power for CPU energy mode to see if there is a difference maybe?



ADVANCED FREQUENCY SETTINGS



CPU Base clock 100

CPU upgrade : auto

CPU vcore : 1.25

Enhanced multi core performance disabled

CPUclock ratio 50

Uncore ratio : 47

AVX offset :0

FCLK frequency for early power on : 1Ghz

XMP : profile 1

Vt-d disabled



Advanced cpu core settings

Intel Speed Shift Technology: enabled

Intel turbo boost technology : enable

C state : disabled

SpeedShift : enabled

Turbo per core limit control : auto

Turbo Power limits ; maxed

Active turbo rations ; all 50

Voltage optimisation : auto



ADVANCED VOLTAGE SETTINGS

Advanced power settings

CPU VCORE Loadline calibration turbo



CPU core voltage control

VCCIO 1.15

CPU SA : 1.15

CPU current limit 255

Cpu vcore PWM switch rate : 300

CSM : disabled

 

[apesoccer]

Interesting (and sorry yea...i read that the wrong way k vs kf). Yea could just be the microcode, but i'd try comparing against other similar setups in UL. May also be your external temps have simply gone up a bit.

Not sure i have much else to offer, keep making 1-2 changes at a time, testing for each...And yea not a bad score at all =).

Funny how close we are 5%~ score wise, on completely different setups...though i expect i'm using a lot more energy ><.

Also if you hadn't seen it...If he really tested and binned 200 9900k's, he should have something to share if he's around.
https://www.tomshardware.com/reviews/intel-9900kf-vs-9900k-overclock-comparison,6038.html

Edit(s):
Added one today: https://www.3dmark.com/fs/22771147

Kept playing with memory timings...and my gpus a small amount, mem@ 3600 @ cl16 are fairly tight, but only got a very small bump over running 3200 @ 14. /shrug...guess i'm not going to beat you to 30k without a cpu overclock...and i probably won't mess with that. folding@home has my system pegged out 24/7 now so probably don't want it running any hotter then it already is.
https://www.3dmark.com/fs/22771280