[SOLVED] Only getting 50% usage out of GPU and CPU during benchmarks

[Bigj73nsb]

Ok here's the deal
Ok to start off, this is a fresh install of Win 7 home Premium on a new SSD drive with a new GPU in an old system
Asus Z97-A DDR3 motherboard//
i5 4670k 3.4ghz OC to 4.3ghz CPU//
EVGA RTX 2060 KO Ultra
Corsair H100 sealed loop
CPU water cooler//
Kingston HyperX FURY 16GB Kit (2x8GB) 1866MHz DDR3
WD Blue 3D NAND 1tb SSD
Ok I just upgraded from a HDD to SSD and from a GEFORCE GTX 760 2gig to RTX 2060 KO Ultra GPU and went through a fresh install of Win 7. Installed all drivers for Windows and GPU, updated BIOs and Chipsets plus other MOBO drivers and did an EZ tuning wizard slight overclock to 4.3. Had to turn down the ram clock it gave me because it was too much from 1886 to 24 something. Failed to boot with that clock,so I tuned it down.
Anyway the problem I'm having now is that I rand benchmarks through Furmark, Kombuster and Heavens Benchmarks and I'm only getting like 50% usage out of GPU and CPU both capping at 60 fps on the dot. I was thinking that my RAM was holding me back some, but at the same time I don't believe the RAM would hold it back that much. I figured I might get some bottlenecking, but if that was the issue CPU would be running higher if not at 100% while GPU would be lagging behind. I went into NVIDIA control panel to check power settings and there at max performance.
If there's any one that could help me put a finger on this, it would be much appreciated.....
Ty for your time.....

 

[TerryLaze]

Anyway the problem I'm having now is that I rand benchmarks through Furmark, Kombuster and Heavens Benchmarks and I'm only getting like 50% usage out of GPU and CPU both capping at 60 fps on the dot.

The only reason for this would be if something is locked to vsync.
(Most probable the nvidia drivers)

 

[Bigj73nsb]

The only reason for this would be if something is locked to vsync.
(Most probable the nvidia drivers)

Ok you were correct, Vsync was on and it did fix past problem, but now i seem to have a backwards bottle neck going on.
Correct me if I'm wrong, If there's any bottle necking going on, it would be my I5 4670k at 4.3 clock should bottle neck the RTX 2060 KO Ultra? Meaning the CPU should be running at full, not letting the GPU run at full capability right? Well I just ran through Furmark benchmark again and my GPU ran at 100% at like 1845MHz while CPU was at only around 50% at 4300MHz. Correct me if I'm wrong about this, but this has got me confused?

 

[mdd1963]

Technically, at least a partial bottleneck will exist anytime the CPU cannot max out the GPU... (As the 2060 has a fair amount of VRAM, it will likely never show very high usage at 1080P in many games anyway)

As most quad core/quad threads are all potential bottlenecks in many games, the issue is now to what degree. (I'd venture to guess the i5 will get decent average frame rates, but, likely suffer in the minimum FPS, as measured in .1/1% lows, which is where the 4c/4t CPUs get hammered in certain parts of the game, especially with newer games that thrive on 6 or 8 cores and even 6-16 threads.)

The 2060 is a very nice/solid GPU...; it is probably time to start pondering assorted CPU and mainboard reviews..

(The R5-3600 and R7-3700X on a good B450 mainboard are popular combos for a reason!)

 

[sizzling]

Ok you were correct, Vsync was on and it did fix past problem, but now i seem to have a backwards bottle neck going on.
Correct me if I'm wrong, If there's any bottle necking going on, it would be my I5 4670k at 4.3 clock should bottle neck the RTX 2060 KO Ultra? Meaning the CPU should be running at full, not letting the GPU run at full capability right? Well I just ran through Furmark benchmark again and my GPU ran at 100% at like 1845MHz while CPU was at only around 50% at 4300MHz. Correct me if I'm wrong about this, but this has got me confused?

Benchmark software is not good for identifying a ‘bottleneck’. Benchmarks are very good at targeting predominantly the gpu or another component so are not representative of real world gaming. Also bottlenecks also depend on the game, game settings and resolution. How do actual games play?

Ive got a 4670k in a second system. Some games it does ok in. However the latest AAA games it struggles or is even unplayable because the cpu is pegged at 100%.

 

[Bigj73nsb]

Technically, at least a partial bottleneck will exist anytime the CPU cannot max out the GPU...

As most quad core/quad threads are all potential bottlenecks in many games, the issue is now to what degree. (I'd venture to guess the i5 will get decent average frame rates, but, likely suffer in the minimum FPS, as measured in .1/1% lows, which is where the 4c/4t CPUs get hammered in certain parts of the game, especially with newer games that thrive on 6 or 8 cores and even 6-16 threads.)

The 2060 is a nice GPU...; it is probably time to start pondering assorted CPU and mainboard reviews..

(The R5-3600 and R7-3700X on a good B450 mainboard are popular combos for a reason!)

Yes I figured that much and I do plan on slowly upgrading. Kind doing it slowly through payments. Buy a piece of hardware, pay off payment's, then on to the next piece.
Anyway what what my issue here is, is if my CPU bottlenecks my GPU, then CPU should be the one running at a high percentage not allowing my GPU to run at full capacity. But whats going here is backwards isn't it???? Instead my GPU at max% not allowing my CPU to run at full capacity. This seems backwards to me, like theres another issue here. Correct me if I'm wrong about this. Idk maybe it's I'm just up too late and thinking to much but......

 

[TerryLaze]

Yes I figured that much and I do plan on slowly upgrading. Kind doing it slowly through payments. Buy a piece of hardware, pay off payment's, then on to the next piece.
Anyway what what my issue here is, is if my CPU bottlenecks my GPU, then CPU should be the one running at a high percentage not allowing my GPU to run at full capacity. But whats going here is backwards isn't it???? Instead my GPU at max% not allowing my CPU to run at full capacity. This seems backwards to me, like theres another issue here. Correct me if I'm wrong about this. Idk maybe it's I'm just up too late and thinking to much but......

A GPU benchmark is made to push your GPU to 100% ,it does not put a lot of usage to the CPU,a CPU benchmark will do the opposite push your CPU to 100% while the GPU won't be doing much.