This question is sort of academic in nature, just to see if other folks have answers that I haven't been able to find.
I've been running some benchmarks on an Acer Predator Triton PT315-53 notebook with an i7-11800H and a 3060. I have an undervolt on the GPU (1875 @ .95v) to keep the laptop slightly cooler. It's been regularly scoring about 7,500 in TimeSpy (7500 graphics, 7400 cpu). I'm definitely getting below-average performance from the components. What began to puzzle me is the difference in core voltage, clock speed, and wattage between TimeSpy and real gaming.
In Time Spy, I get the following:
Clocks = 1545 mhz (avg) Core Voltage = .775v (avg) Wattage = 89-90 watts (avg) Temps = 71-75 c (avg)
In real gaming (Skyrim SE), I get readings like this:
Clocks = 1875 mhz (avg) Core voltage = .95v (avg) Wattage = 70 watts (avg) Temps = 71-75 c (avg)
Real gaming (Deathloop):
Clocks = 1875 mhz (while looking at the sky), 1550 mhz (while looking at any landscape) Core voltage = .95 (sky) .775 (landscape) Wattage and temps similar to Time Spy.
I'm basing my observations on both RivaTuner and HWInfo64. My question is, why is Time Spy's voltage and clockspeed so much lower than the same metrics in gaming? Am I getting CPU-bound on Deathloop and TimeSpy, leading to lower clocks when looking at more physics-demanding scenes? Am I missing some blatantly obvious thing about this cpu-gpu combo?
I've been running some benchmarks on an Acer Predator Triton PT315-53 notebook with an i7-11800H and a 3060. I have an undervolt on the GPU (1875 @ .95v) to keep the laptop slightly cooler. It's been regularly scoring about 7,500 in TimeSpy (7500 graphics, 7400 cpu). I'm definitely getting below-average performance from the components. What began to puzzle me is the difference in core voltage, clock speed, and wattage between TimeSpy and real gaming.
In Time Spy, I get the following:
Clocks = 1545 mhz (avg) Core Voltage = .775v (avg) Wattage = 89-90 watts (avg) Temps = 71-75 c (avg)
In real gaming (Skyrim SE), I get readings like this:
Clocks = 1875 mhz (avg) Core voltage = .95v (avg) Wattage = 70 watts (avg) Temps = 71-75 c (avg)
Real gaming (Deathloop):
Clocks = 1875 mhz (while looking at the sky), 1550 mhz (while looking at any landscape) Core voltage = .95 (sky) .775 (landscape) Wattage and temps similar to Time Spy.
I'm basing my observations on both RivaTuner and HWInfo64. My question is, why is Time Spy's voltage and clockspeed so much lower than the same metrics in gaming? Am I getting CPU-bound on Deathloop and TimeSpy, leading to lower clocks when looking at more physics-demanding scenes? Am I missing some blatantly obvious thing about this cpu-gpu combo?
Facebook
Twitter
Instagram