Cinebench R15 Multi Loop

Therefore, the score curve of the P16G1 under BIOS 1.13 and 1.15, and the Cinebench R15 multi-core performance 50-cycle test is also significantly different, compared with other models, as shown in the figure below:

P16G1 BIOS 1.13 is a red curve, with a peak of 3144.65, a valley of 2753.49, and an average of 2845.35;

P16G1 BIOS 1.15 is a white curve, with a peak value of 2984.63, a valley of 2711.13, and an average value of 2758.99;

HP ZBook Fury 16 G9 is a blue curve with a peak of 2970.22, a trough of 2788.43, and an average of 2884.93;

In this way, under long-term high pressure, BIOS 1.15 P16G1, its CPU performance play, has been in the downside of F16G9, only higher than P1G5... This is exactly the same as the comparison result of the 3DMark CPU Profile link above.

The above are separate stress tests of CPU performance or graphics performance using performance benchmarking software.

The author continues to use the Stress software, and when the two are stress-tested separately or simultaneously, I will see how the P16G1 values are in many aspects.

In the following test, data is recorded using HWiNFO64 and data is analyzed using Generic Log Viewer.

At the time of the test, the room temperature was about 15 degrees Celsius, and the test object was P16G1-DG-C (discrete graphics card - 4x32GB memory) with BIOS of 1.13, 1.14 and 1.15, respectively.

　　
AIDA64 Stress FPU

Using the Stress FPU in the system stability test module of AIDA64, the CPU was subjected to a stress test for more than 15 minutes, and one screenshot was not sent during the test, so as not to affect the page display speed and layout.

Using Generic Log Viewer, the first 15 minutes of HWiNFO64 recorded data were analyzed to obtain the following figures, where red is BIOS 1.13 data curve, green is BIOS 1.14 data curve, and blue is BIOS 1.15 data curve.

The first is a comparison chart of CPU Core Temperatures-avg.

You can see:

The average CPU core temperature of BIOS 1.13 has a peak of 92 degrees Celsius and a valley of 74 degrees Celsius (the first 10 seconds are standby, not counted, the same below), and the average value is 87.47 degrees Celsius;

The average CPU core temperature of BIOS 1.14, with average peak valley values of 90, 88 and 87.12, respectively;

The average CPU core temperature of BIOS 1.15, with average peak valley values of 85, 83 and 82.88, respectively;

The average average temperature of CPU cores at 1.13 and 1.14 is about the same, but 1.15 is 5 degrees lower.

Among them, the curve of 1.13 has seen several sharp declines and rises, while 1.14 and 1.15 are relatively stable and do not have violent ups and downs, in other words, the latter two are relatively better stable than 1.13.

The second graph is a CPU Package Power comparison chart.

CPU power consumption of BIOS 1.13, peak 156.8 watts, valley value of 60 watts (the first 10 seconds is standby, not counted, the same below), the average value is 99.98 watts;

The average CPU power consumption of BIOS 1.14 is 127.9, 87 and 89.18, respectively;

The average CPU power consumption of BIOS 1.15 is 108, 75 and 72.61, respectively;

Based on the above measured data, the author believes that the long-term targets of CPU power consumption of the three are

BIOS 1.13 is 100 watts,

1.14 is 90 watts;

And 1.15 is only 75 watts.

The third graph is a Core Clocks-avg comparison chart:

The average CPU core frequency of BIOS 1.13, the peak value is 3965MHz, the valley value is 2520MHz (the first 10 seconds are in the standby state, not counted, the same below), and the average value is 3127MHz;

The average CPU core frequency of BIOS 1.14, with average peak and trough values of 3547, 3000 and 3059, respectively;

The average CPU core frequency of BIOS 1.15, the average peak and trough values are 4184, 2800 and 2790, respectively.

The operating frequency of the CPU determines the performance of the CPU, which is why the three CPU performance has a huge difference based on the above average, the performance of the three CPUs under long-term high voltage, theoretically will be 3127:3059:2790---- that is, about 100:98:89---- in other words, the long-term performance of the CPU of BIOS 1.15, only 89% of 1.13, down 11%!

In the process of the above stress test, the total power consumption of each of the three (Total System Power) is compared as shown in the figure below:

The total power consumption of BIOS 1.13 is 221.3 watts at a peak value, with an average value of 155.7 watts;

The peak power consumption of BIOS 1.14 is 183.3 watts, and the average value is 141.5 watts, which is 14 watts less;

The total power consumption of BIOS 1.15 is 159.3 watts at a peak value, with an average value of 120.3 watts, which is 22 watts less than 1.14, which is 36 watts less than 1.13.

During the above stress test, the comparison of the respective CPU PL1 values is shown in the figure below:

1.14 and 1.15 remained unchanged at 157 watts, only 1.13 showed a drastic change, the peak was the same at 157 watts, but the trough fell to 60 watts, and the average was 119.1 watts.

This seems to indicate that the CUP performance of the P16G1 is not controlled by PL1 setting a power consumption wall.

　　
FurMark GPU Stress Test

Using the FurMark GPU Stress Test, the three GPUs were subjected to a stress test lasting nearly 20 minutes, and the HWiNFO64 recorded data was analyzed using Generic Log Viewer to obtain the following figure, where red is BIOS 1.13 data curve, green is BIOS 1.14 data curve, and blue is BIOS 1.15 data curve.

You can see:

The peak GPU temperature of the three is a little more than 81 degrees Celsius, and the average value is about 76 degrees Celsius---- which is much higher than the HP ZBook Fury 16 G9 (53.24 degrees Celsius), and the ThinkPad P1 Gen 5 (81.25 degrees Celsius);

The peak GPU power consumption (GPU Power) of the three is 130 watts, and the average value is about 128 watts--- which is much higher than the HP ZBook Fury 16 G9 (88.73 watts) and ThinkPad P1 Gen 5 (79.69 watts);

The peak operating frequency (GPU clock) of the GPU is 1700+MHz, and the average value is 1500+MHz.

At this time, the peak total system power consumption of the whole machine (Total System Power) is about 194 watts, and the average value is about 186 watts.

Therefore, changes in BIOS versions have no effect on CPU performance in single-stress tests.

　　
AIDA64 Stress FPU+ FurMark GPU Stress Test

Using AIDA64 Stress FPU+ FurMark GPU Stress Test, the CPU and GPU were subjected to a continuous stress test of no less than 30 minutes at the same time, and the HWiNFO64 recorded data was analyzed using Generic Log Viewer to obtain the following figures, where red is BIOS 1.13 data curve, green is BIOS 1.14 data curve, and blue is BIOS 1.15 data curve.

The first figure is a comparison chart of the three elements of the CPU---- temperature, power consumption, and frequency.

The peak temperature of the three was 92, 84 and 89 degrees Celsius, and the average value was 84.8, 76.93 and 77.04 degrees Celsius.

The peak power consumption is 119.8, 118.1 and 108.2 watts, and the average value is 71.64, 40.8 and 41.64 watts, respectively.

THE FREQUENCY PEAKS ARE 3666, 4184, AND 4221MHZ, AND THE MEAN VALUES ARE 2707, 1971, AND 1986 MHz.

BIOS 1.13 is much higher in temperature, power consumption, and frequency.

According to the average of the CPU core frequency (Core Clocks-avg), it can be speculated that the CPU performance under BIOS 1.13 will theoretically be 127.34% of 1.14 and 136.30% of 1.15.

The second figure is a three-element comparison chart of the GPU.

The GPU temperature below BIOS 1.13 is lower than the latter two, and the average power consumption is 113.9 watts, which is also lower than the 126+ watts of the latter two;

Therefore, the average GPU frequency is 1440MHz, which is also lower than the latter two 1502.

Theoretically, the GPU performance of BIOS 1.13 is only 95.87% of the latter two.

At this time, it can be known that under double pressure, compared with BIOS 1.13, the CPU performance of the latter two is nearly thirty percent lower, and the GPU performance is 5% higher.

What you get is the stability of the system.

Continue to the third figure: CPU PL1 and the whole machine power consumption comparison chart.

At the time of the dual-stress test, CPU PL1 below BIOS 1.13 experienced severe fluctuations, with a valley of 55 watts and an average of 114.4 watts. While 1.14 and 1.15 do not move--- although it does not help the actual CPU power consumption of 40+ watts, it maintains stability without ups and downs, which cannot be denied.

At the time of the double stress test, the power consumption curve of BIOS 1.13 also had frequent ups and downs, and the latter two, after the outbreak in the initial stage, were relatively stable.

The peak power consumption of the whole machine, 1.13 and 1.14 are 280 watts, 1.15 is 267.6 watts;

The average power consumption of the whole machine, 1.13 is 228.5 watts---- which is almost equal to the 230 watts of the standard power supply of P16G1, 1.14 and 1.15 are more than 28 watts lower, both are about 199 watts---- from the power supply point of view, of course, the lower the load, the more stable.

BIOS 1.13 double-stress test when the average power consumption of the whole machine is close to 230 watts, but also brings a hidden danger---- P16G1 at this time is extremely thirsty for energy, in the case of external power supply, but also use its built-in main battery---- built-in main battery began to continue to discharge.

As shown in the figure below, in the 30-minute double stress test, under BIOS 1.13, the power of the built-in main battery dropped from 99.9% to 96.98%, 2.92% less, and used about 0.1% of the battery power per minute.

And for the same 30 minutes, under BIOS 1.14, the power dropped by only 0.4%, which is negligible.

The most funny thing is that under BIOS 1.15, after 38 minutes, its power still has spare power, and within 30 minutes the battery is charged from 94.5% to 98.8%, and the power has increased by 4.3%... Steady! 1.15 Brother, you are really stable!