Continued

This is the third and final article in the ThinkPad P1 Gen 5 review. The full text of 8300 words and 40 pictures, the Chinese version was first published on the "Fair Evaluation Chinese Station", the English version will be released later on the "FairReviews English Station", and will also be issued in the "Fair Evaluation" of Zhihu, Weibo and WeChat public accounts, this article is divided into the following sections:

• synthesis

• pressure

• tactics

• temperature

• noises

• power consumption

• battery

• appraise

• expect

Synthesis

　　PCMark 8

Using PCMark 8 for the comprehensive performance test of the whole machine, the scores of P1G5 in the two states of hybrid graphics card and discrete graphics card are as follows:

Home accelerated 3.0 scores: 5946 and 5478;

Creative accelerated 3.0 score: 10026 and 8804;

Work accelerated 3.0 score: 6032 and 6031;

Note: The scores for discrete graphics cards are lower than those for hybrid graphics.

Comparing the above test results with the comparison models selected in the author's historical test database, the results are as follows:

Based on the overall score of the P1G5 discrete graphics card with the lowest overall score:

Compared with the comparison model, the P1G5 is basically completely defeated?!

Just higher than the previous generation P1G3 (Intel i9-10885H+Quadro T2000)?!

The author repeated it many times, and the result is still the same.

After contacting UL's technical department, I was told that it is not recommended to use PCMark 8 for benchmarking current PCs---- and UL has officially been completely free of charge for PCMark 8, which means that UL does not care about this, because PCMark 8 is old and retired.

After the author's thinking, it should be that Intel started from the 12th generation CPU, using a heterogeneous core architecture, and possibly some of the impact caused by the Windows 11 system, so that PCMark 8 could not play the performance of P1G5 at all.

In this way, the author's future evaluation content will probably abandon the use of PCMark 8, the old benchmark software...

Here, I would like to express my deep respect to FutureMark, which was acquired by UL!

　　再见了!多年的PCMark 8!

　　PCMark 10

When tested with PCMark 10, the P1G5 scored in both hybrid and discrete graphics states, as follows:

PCMark 10: 7350 and 7969;

PCMark 10 Express: 6099 and 6594;

PCMark 10 Extended: 9349 and 9677;

PCMark 10 Applications: 14452 and 14699;

The above test results are compared with the comparison models selected in the author's historical test database, and the results are as follows:

Based on the total score of 38939 for the P1G5 discrete graphics card, three models have slightly higher total scores among the comparison models:

The total score of the P17G2 discrete graphics card status is 102.57%;

ASUS ROG S17 GX703 for its 100.83%; ，

The total score of the 7760 discrete graphics card status is 100.38%;

Everything else is lower than the P1G5---- the total score of the previous generation P1G3 is only 77.23% of that of the P1G5.

　　SPECwpc 2.1

Using the industry's comprehensive performance benchmarking software SPECwpc 2.1, the P1G5 discrete graphics card state was tested, and the 5 scores obtained were compared with other models, and the situation was:

The author was surprised to find that the P1G5, whether it is the total score or each sub-score, is higher than all the comparison models!

Based on the overall score of the P1G5 discrete graphics card:

Even the P15G2 discrete graphics card status that is closest to P1G5 in total score is only 69.02%!

The furthest away is P1G3, with a total score of only 53.22%.

According to the test results of CPU and GPU in the middle part, the author can only speculate that this is because of the powerful blessing effect brought by Intel's 12th generation CPU, or perhaps there are reasons for the different Windows versions: the comparison models are all installed with Windows 10, while P1G5 is Windows 11...

　　SPECworkstation 3.10

Using the industry's comprehensive performance benchmarking software SPECwpc 2.1, the P1G5 discrete graphics card state was tested, and the result was that the test part failed---- CPU Blender always reported an error, the author repeated many times is still the same, and I don't know why I checked the test log:

In the end, I had to abandon Media and Entertainment's score and compare the remaining 6 scores with other models, as follows:

The P1G5's four sub-scores, Product Development, Life Sciences, Financial Services and Energy, were higher than all comparison models.

Only two sub-categories, General Operations and GPU Compute, did not score as well as some heavy-duty flagships.

The author believes that this comparison result, coupled with the comparison result of SPECwpc 2.1 above, is expected.

The power and performance of the new hardware of the P1G5 is indeed remarkable.

　　UL Procyon

UL Procyon is UL's newly released benchmark software that tests the P1G5 discrete graphics card with the following scores:

　　Office Productivity score：7426;

　　Photo Editing score：8341;

　　Video Editing score：6527;

Compare the results obtained with the author's test history models, as follows:

Due to the continuous update of UL Procyon, some models are missing.

Based on the overall score of the P1G5 discrete graphics card:

The total score of the P17G2 discrete graphics card status is 105.89%,

The total score of 7760 discrete graphics card status DDR4-3466 memory is 100.23%;

The total score of the P15G2 discrete graphics card status is 103.49%;

The overall score of the other models is inferior to the P1G5.

Based on the above UL Procyond score comparison, it can be seen that the gap between the P1G5 and the previous generation of mobile workstation flagships (P17G2, P15G3 and 7760) is small.

　　UserBenchmark

UserBenchmark is an online benchmarking software that can be used for free without registration, and any user's test results will enter its official website database and provide users with open query comparison.

Using UserBenchmark, the P1G5 was tested in the independent display state, and the test results were as follows:

Gaming score: 97%, rated as a nuclear submarine;

Desktop score: 109%, UFO rating;

Workstation score: 102%, rated UFO level.

At the same time, UserBenchmark has included 5 data of the same model (21DC0030US) as this P1G5 so far, and its three sub-scores are average: 100%, 105% and 106%, which is basically the same as the score of this station measured by the author.

　　PassMark PerformanceTest

PassMark PerformanceTest is a paid software, if you do not pay, there is a time limit, the test results can be uploaded to the official online database, and the existing data can be downloaded for comparison, the author tested the P1G5 in the discrete graphics card display state.

With a PassMark Performance Test CPU Mark score of 31350, the i9-12900H above the P1G5 outscored all comparisons, as in the mid-length CPU performance test.

The PassMark Performance Test 2D Graphics Mark scored 1199, which is 1309 below 7760, which is similar to the score of P17G2 and P15G2.

The PassMark Performance Test 3D Graphics Mark score was 18,274

The scores of mid-range P17G2, P15G2 and high-end P17G2 were 20344, 21093 and 24156, respectively; 7760 scored 23986, 7560 scored 19690, ASUS ROG Ice Blade 5 Plus 21024;

It can be seen that the 3D performance of the P1G graphics card RTX A5500 is indeed limited and not fully played.

The PassMark Performance Test Memory Mark scored 3389, which is at the upper middle level across the note-taking pair.

The PassMark Performance Test Disk Mark score of 33377 is the result of switching to a Western Digital SN850 2TB SSD.

Based on the above, the final score given by the PassMark Performance Test is 9087, which is higher than all the comparison models.

From the above PCMark 10, SPECwpc 2.1, SPECworkstation 3.10, UL Procyon and UserBenchmark and PassMark Performance Test, a total of 6 comprehensive performance test results and comparison, the author believes that P1G5 plus comparison models, P1G5 comprehensive performance ranks first, but graphics card performance is limited. In graphicality still can not fully suppress all comparison models.

The above comprehensive test can be said to be a test of performance explosion ability, under the long-term, high-load stress test (commonly known as "copying machine"), how will the performance of P1G5 be?

See below.

pressure

　　3DMark Stress Tests

At room temperature of around 25 degrees Celsius, the P1G5 with BIOS 1.05/EC 1.03 was tested in the discrete graphics card state.

The stability of Time Spy Stress Test and Fire Strike Stress Test was 95.9% and 94.8%, respectively---- P1G5 failed the test.

A few days later, the P1G5 got BIOS 1.11/EC 1.06 update:

After completing the above update, the author conducted the same test again, and the stability of the Time Spy Stress Test was 98.3%, and the P1G5 passed the test.

The Fire Strike Stress Test was 98.1% stable, and the P1G5 also passed.

3DMark added a Lenovo Savior-sponsored Speed Way test module, as well as a Speed Way Stress Test with a 98.3% test stability, and the P1G5 passed the test again.

It can be seen that the new BIOS and EC of P1G5 have optimized the stability of the system.

So, how is it optimized?

See below.

　　Cinebench R15 Multi Loop

The author once again conducted two Cinebench R15 multi-core performance 50 round-robin tests on the updated P1G5, and the score curve comparison chart is as follows, the curve of the P1G5 three tests is higher than for all comparison models, where:

The gray curve of State A---- is P1G5 with BIOS 1.05/EC 1.03, the room temperature is 20 degrees Celsius at the time of testing, and the average Cinebench R15 multi-core performance score is 2621.79;

State B---- P1G5 with a white curve of BIOS 1.11/EC 1.06, a room temperature of 25 degrees Celsius at the time of testing, and an average Cinebench R15 multi-core performance score of 2493.37;

The blue curve of state C ---- P1G5 with BIOS 1.11/EC 1.06, the room temperature was 15 degrees Celsius at the time of testing, and the average Cinebench R15 multicore performance score was 2362.61;

Looking closely at the above comparison graph, the author found:

1. After updating to the new version of BIOS/EC, the CPU performance of P1G5 has decreased;

1. P1G5 with the same BIOS/EC, the performance of the CPU with a slightly lower ambient temperature is even lower...

Point 1 can also be understood that the official new version of BIOS/EC has made a certain degree of restrictions on performance for stability; The second point is not understood, the ambient temperature is within the normal use range, the ambient temperature is low, but the performance is even lower? What the hell is this?!

Continue to see how the three sets of tests compare below.

The following test uses the AIDA64 stress function, uses HWiNFO64 to record data, and uses Generic Log Viewer to analyze data.

During the test, the P1G5 is a discrete graphics card, different from the BIOS/EC version and ambient temperature

The P1G5 of State A is: BIOS 1.05/EC 1.03 with an ambient temperature of 20 degrees Celsius

The P1G5 of State B is: BIOS 1.11/EC 1.06 with an ambient temperature of 25 degrees Celsius.

　　AIDA64 Stress FPU

Using the Stress FPU in the system stability test module of AIDA64, the CPU was stress tested for 15 minutes, and the screenshot of the two-state P1G5 is shown in the following two figures:

Using Generic Log Viewer, the HWiNFO64 record data was analyzed to obtain the following figure, where red is status A and green is status B.

You can see:

1. The CPU temperature of state A and state B are similar, both peak and average, both 96 degrees Celsius and 94.9 degrees Celsius, trough (that is, the temperature before the start of the stress test), state A is 46 degrees Celsius, state B is 51 degrees Celsius---- which is in line with the difference between the different ambient temperatures of the two states---- just 5 degrees Celsius;

2. However, there is a clear difference in CPU package power consumption between the two, with a peak of 109 watts in state A, 99.16 watts in B, an average of 85.62 watts in state A, and only 74.10 watts in state B---- with a gap of 11.52 watts!

3. Then, the gap in power consumption is reflected above the average operating frequency of the CPU (CPU Clocks-avg---- the average value of state A is 3141MHz, and state B is only 2980MHz... The gap is 5.4%, which determines the final performance of the two states;

In the Cinebench R15 Multi Loop test above, there was a significant score gap between ----the two states: the mean of state A was 2621.79, and the mean value of state B was 2493.97, and the difference between the two was 5.15%.

　　AIDA64 GPU Stress Test

Using AIDA64 GPU Stress, the GPU was stress tested for 15 minutes, and the screenshot of the two-state P1G5 is shown in the following two figures:

Using Generic Log Viewer, the HWiNFO64 record data was analyzed to obtain the following figure, where red is status A and green is status B.

You can see:

1. There is a significant difference between the GPU temperature, peak and mean value of state A and state B:

The peaks were 80.5 degrees Celsius and 84 degrees Celsius, respectively;

The mean values were 76.63 degrees Celsius and 81.25 degrees Celsius, respectively;

But the trough (that is, the temperature before the start of the stress test) is not, 47.7 degrees Celsius and 42.5 degrees Celsius, respectively---- which is just the opposite of the difference in ambient temperature between the two states---- but state A is 5 degrees Celsius higher than state B? The author speculates that this should be the P1G5 new and old BIOS/EC, which has different solutions for the GPU that was in a low-load state before the stress test, and the GPU cooling fan under the old and new cooling strategies.

2. There is a slight difference in GPU power consumption (GPU Power) between the two:

peaks were 89.7 watts and 84.37 watts, respectively;

The mean values were 79.69 watts and 78.99 watts, respectively;

The valleys were 14.16 watts and 13.42 watts, respectively;

Overall, state B consumes less power than state A, but the difference is small.

3. Then, the gap in power consumption is reflected above the average operating frequency of the GPU (GPU Clock---- the average state A is 1227MHz, and the average state B is slightly lower, 1182MHz, the gap is 45MHz---- the gap between the two states is 3.8%.

4. In the end, the GPU performance of the two states is basically the same, and the author's results in multiple actual tests are basically no gap, which is also empirical.

　　AIDA64 Stress FPU+ Stress GPU Test

Using the AIDA64 Stress FPU+ Stress GPU, the CPU and GPU were subjected to a continuous stress test for 30 minutes at the same time, and the screenshot of the two-state P1G5 is shown in the following two figures:

Using Generic Log Viewer, the HWiNFO64 record data was analyzed to obtain the following figure, where red is status A and green is status B.

You can see:

1. Under the long-term double torture of AIDA64 Stress FPU+ Stress GPU, the temperature, power consumption and operating frequency of the two states of the P1G5 CPU have fluctuated sharply, while the power consumption and operating frequency of the GPU are relatively stable, only the temperature fluctuates;

2. The CPU temperature curve of the two states, the beginning rises rapidly to the peak, maintains for a period of time and then plummets (state A remains for nearly 4 minutes, state B remains for nearly 8 minutes), and then periodic ups and downs, especially state B is more obvious, the final statistical value comparison is as follows:

The peak value is 96 degrees Celsius;

The mean values were significantly different, with state A being 78.83 degrees Celsius and state B being nearly 10 degrees Celsius higher at 88.76;

The difference in valley value is still the same as above, state A is higher than state B, 44 degrees Celsius and 49 degrees Celsius, respectively---- or a direct reflection of the different ambient temperatures of the two states.

3. The CPU power consumption curve of the two states shows a peak at the beginning, and then slowly declines, and then there is a periodic fluctuation state A for nearly 4 minutes, state B for nearly 8 minutes), and the final statistical values are compared as follows:

The peak value of state A is 109 watts, and the peak of state B is more than 5 watts lower is 104.9;

The mean value of state A is 39.94 watts, and state B is 7 watts higher than 47.02;

The valley value of State A is 9.178 watts, and the valley value of State B is 7.865 for more than 1 watt;

4. The average operating frequency curve of the two-state CPU is exactly the same as the above power consumption curve, and the result is that the average value of state B is 2426MHz, and the 2175MHz higher than state A is 251MHz, and the impact of this gap on performance will theoretically reach 10%.

5. The GPU temperature curve of the two states, on the whole, state B is higher than state A, and the average value is nearly 9 degrees Celsius higher;

6. The GPU power consumption curve of the two states, except for the abnormal value of a needle-like peak in the middle, is almost coincident, but the initial state B has a period of frequent fluctuations to form a depression curve, which seems to be less stable than state A, and the final average value of both is about 79 watts;

7. The operating frequency curve of the two-state GPU is that state A is higher than state B, and the average difference is 1234-1159=75MHz, and the impact of this gap on performance is theoretically about 6%.

8. The sum of the average power consumption of CPU + GPU in the two states is 40+80=120 watts and 47+80=127 watts, respectively.

Above, there is more data, and the description has to use a lot of words, which is easy to make readers who are not fully engaged in reading confused, so the author summarizes again, that is:

1. State A When the CPU or GPU is running independently and under high load, the individual results will be better than state B--- CPU is a little better than 5%, and the GPU is about 4% better;

2. However, when running the dual high load of CPU + GPU, the CPU performance of state B will be reversed, about 10% higher than state A, but the GPU continues to be weaker than state A, about 6% lower.

This is the difference brought by the official BIOS 1.05/EC 1.03 and BIOS 1.11/EC 1.06---- that is, the official vigorously improves the CPU performance at the time of double high load, and again suppresses the performance of the GPU---- GPU is about 6% lower, in exchange for a 10% increase in CPU---- at the same time, the independent results of CPU and GPU are about 5.4% and 3.8% lower respectively...

This is the so-called dynamic power adjustment function cooperated by Intel and Nvidia, which was played by the P1G5 official...

tactics

Such a power consumption strategy, the author does not like!

What I like is:

The average independent power consumption of CPU and GPU, both are 100 watts, CPU + GPU double high load operation, 80 + 80 = 160 watts, isn't it beautiful?

Doesn't your P1G5 CPU have a PL1 value of 109 watts?

Didn't your P1G5 TGP label 100 watts?

Isn't the power supply of your P1G5 230 watts? Where has it all gone?

Breath anger anger anger anger

The intel Core i9-12900H's TDP is 45 watts...

This is a laptop, not a desktop;

This is the P1G5 with a tendency to be thin and light, not the performance flagship Thick Brick P16G1 (ThinkPad P16 Gen 1/ThinkPad P16 2022)...

temperature

In the above, after analyzing the data of P1G5 in multiple stress tests (commonly known as copying machines), we have a relatively comprehensive understanding of the detailed temperature data of the two most critical components of P1G5 during high-load operation, CPU and GPU.

It can be seen that in all stress tests, the CPU temperature peak of P1G5 is up to 96 degrees Celsius, which is controlled below the intel CPU temperature threshold of 100 degrees Celsius, and when running under high load for a long time, the average CPU temperature is lower than 89 degrees Celsius---- this temperature control level, the author believes that it is more excellent---- if the CPU working temperature is too high, overheating and frequency reduction will occur frequently, and the peak value of P1G5 CPU is just a little away from the 100 setting.

If the temperature average of the CPU is very low when running at a high load, for the general laptop, it is often not that the heat dissipation system is very good, but that the laptop deliberately suppresses excessive power consumption in order to control the temperature, resulting in too much CPU performance limitation.

After all, laptops are usually not desktops, and not all can be casually equipped with liquid cooling systems.

Therefore, when the CPU of P1G5 is running under high load, the average temperature is around 89 degrees Celsius, and the peak value is 96 degrees Celsius, which is a relatively clever range.

Looking at the temperature of the GPU, basically, the temperature average and valley of the GPU (84.01/45.9) are lower than the CPU (88.76).

/49) about 3 degrees Celsius, only the peak gap is large, the GPU is 87.1, the CPU is 96, the gap reaches nearly 10 degrees Celsius.

In the eyes of some readers, the operating temperature of the GPU at 84 degrees Celsius seems to be on the high side.

In fact, this is the proof that the P1G5 cooling system is excellent---- it is precisely because of the excellent P1G5 cooling system, so that the CPU and GPU in the simultaneous high load operation, the heat dissipation system can try to exclude the heat of the two high temperature parts from the heat dissipation window evenly, so that the temperature gap of the CPU GPU is not very different.

The P1G5 uses a liquid metal cooling system (Vapor Chamber cooling is available only with NV RTX A3000 / A4500 / A5500 and GeForce RTX 3070 / 3080 GPUs) except for the low-end version (models with A1000 and A2000 graphics cards). ----- This so-called Vapor Chamber cooling, literally translated as steam chamber cooling, is actually a vapor chamber heat dissipation scheme---- P1G5 on this basis, using liquid metal as the thermal medium between the vapor chamber and the CPUGPU, instead of the conventional heat dissipation grease, further improve the heat dissipation efficiency.

Of course, it also brings the difficulty of the operation of conventional users trying to disassemble and assemble heat dissipation parts, liquid metal after removing the heat dissipation parts, want to install them back intact, is a certain degree of operation technology, but also must have the corresponding liquid metal materials to supplement---- which is also one of the important reasons why the author completely abandoned removing P1G5 heat dissipation components and seeing the true face of Lushan in the previous disassembly link.

The heat dissipation system shown in the P1G5 official hardware maintenance manual is as follows:

When the ambient temperature is 15 degrees Celsius, running the AIDA64 Stress FPU+ Stress GPU, the P1G5 double copy in the B state lasts for one hour, and the temperature of the body surface is measured immediately without stopping the double copy, and the results are as follows.

Power supply: 35.8 degrees Celsius at the highest point;

Screen: the average value is about 20 degrees Celsius;

The upper edge of the keyboard at the junction of the BC surface, near the center of the screen, is the highest temperature: 45.3 degrees Celsius;

The left and right sides of the keyboard each have a minimum temperature, about 18 degrees Celsius---- the reason is that these two places, the bottom interior corresponds to where the two cooling fans are;

The left and right sides of the palm rest are about 24 degrees Celsius;

There are no exhaust vents on the left and right sides of the fuselage, which does not affect the temperature of the nearby area.

The temperature of the C surface where the P1G5 keyboard is located is as follows:

Palm rest: the overall balance is relatively balanced, at about 24 degrees Celsius, comfortable temperature;

Keyboard: left and right at about 22 degrees Celsius, the center temperature is higher, reaching 43.4 degrees Celsius;

The middle between the keyboard and the screen face is the highest temperature, which reaches 46.6 degrees Celsius.

The temperature diagram at the bottom (D surface) of P1G5 is as follows:

Below the bottom palm rest: 24.6 degrees Celsius lower on the left side, 27.8 degrees Celsius on the right side, and 29.3 degrees Celsius higher in the middle;

Below the bottom keyboard: 25.7 and 31 degrees Celsius on the left and right, respectively, and 38.5 degrees Celsius in the center near the heat source;

The middle of the bottom rear is where the heat source is concentrated, showing that it reaches 48.62 degrees Celsius, but this highest temperature part is the heat dissipation grille, due to the physical principle of infrared temperature measurement, in fact, it has measured the temperature inside the body, that is, the temperature on the internal soaking plate.

When the P1G5 is subjected to a 1-hour double copy test, the surrounding temperature behind it is shown below:

You can see:

The billowing heat wave from the rear heat exhaust window is already placing the P1G5 tabletop, causing two obvious high temperature bands.

Because there is no heat dissipation vent on the left and right sides of the P1G5 fuselage, there is no temperature rise phenomenon on the left and right sides of the fuselage.

In summary, the author believes that the heat dissipation system of P1G5 belongs to the excellent grade, and the temperature difference between the CPU and GPU is small when the long-term double copy is small, which completely ensures that the CPU temperature does not reach the threshold;

At the same time, the temperature of most of the key parts of the body is within a good or acceptable range, only the C surface is close to the middle of the screen, the bottom of the corresponding part, there is a high temperature that cannot be ignored, but it is also suppressed below 50 degrees Celsius, in the long-term high-load operation, users only need to pay attention to try to avoid the P1G5 on the thigh to operate, so that there is no problem---- However, as a laptop (Laptop), really put on the thigh for a long time to run most of the scenes, There should be no external power supply to run with the battery, at this time the power consumption strategy of P1G5 will be optimized, the performance will be reduced, the external temperature of the whole machine will not be displayed, and the body temperature will be very cool.

As for other load states, the temperature of the CPU and GPU will of course be much cooler, and the external temperature of the P1G5 body is also the same, most readers will not pay too much attention, and the author will not test and analyze the temperature of other load states.

In the end, although the author is worried that the CPU + GPU cannot achieve 100 + 100 = 200 watts, I still have to admit that the official statement is basically true:

"Liquid metal dissipates heat, works for a long time, cooler and more stable!"

noises

According to the author's human ear records, the fan noise of P1G5 belongs to the good category, without excessive sound pressure and unbearable high-frequency noise.

Tip: Since the author is nearly half a hundred years old and has poor hearing, there is no guarantee that the above subjective views on fan noise can be applied to all users.

In the future, if there is more time, we will use professional instruments to record noise as much as possible, and make objective comments based on the recorded values.

During the stress test and standby, the ear was deliberately placed close to all parts of the P1G5 body to listen, and no mechanical or electronic noise was heard except for fan noise.

power consumption

The CPU + GPU power consumption before and after P1G5 update BIOS/EC has been explained in detail above, and it is summarized here:

BIOS 1.11/EC 1.06 CPU+GPU = 40+80=127W,

BIOS 1.05/EC 1.03 CPU+GPU = 40+80=120 watts.

The following figure is the data recorded after the author runs AIDA64 Stress FPU+ Stress GPU, and the P1G5 double copy of the B state lasts for one hour, according to the total system power (Total System Power) data recorded by HWiNFO64, the P1G5 of BIOS 1.11/EC 1.06 can be seen that its peak power consumption is about 220 watts, and the average value is about 155 watts---- so it seems that the standard power supply of this P1G5 is 230 watts , there is still the potential to increase the power consumption of the P1G5 by 10 watts.

At the same time, it can be seen that during the process of P1G5 double copy lasting one hour, the cooling fan of P1G5 is basically running at full speed---- starting from more than 7 minutes, the speed has increased again and has been stable at 5100RPM.

battery

The P1G5 has a built-in main battery capacity of 90Wh, and the official phrase for its battery life, the author feels that the language is vague and vague.

The author's current battery life test of the laptop uses PCMark 10 Professional Edition.

The battery life test of PCMark 10 has 5 scenarios, namely game (GAME), video playback (VIDEO), application (App), modern office (MO), idle (IDLE).

According to the actual situation, the author selected four scenarios for testing, and the relevant settings are as follows:

Modern Office (Modem Office): screen brightness 50% + battery performance balanced;

Applications: Same as above;

Gaming: Screen brightness 100% + best battery performance;

Idle: screen brightness 0% + battery performance energy saving;

The above 4 scenarios were tested in the P1G5 state of hybrid graphics card and discrete graphics card, and 8 results were obtained as follows:

1. The blue box is the score of the P1G5 hybrid graphics card status, and the red box is the score of the discrete graphics card status.

2. In the game scenario with the highest power consumption, the battery life of hybrid graphics cards and discrete graphics cards is 1:2 and 1:12 respectively, and the hybrid graphics card state is actually 10 minutes less;

3. In the scenario with the lowest power consumption, the hybrid graphics card and the discrete graphics card are 13:42 and 4:20 respectively, which is a huge gap;

4. Application scenarios, the two are 3:33 and 7 hours and 35 minutes respectively, and the gap is also huge;

5. In modern office scenarios, the gap is still huge at 3:15 and 6:18 respectively;

In summary, if it is not a necessary requirement for the application, the P1G5 runs in the hybrid graphics state, which is the best choice when working only on batteries.

P1G5's built-in main battery supports fast charging, the official propaganda can be fully charged 80% in 1 hour, the author has turned off the computer many times, the actual measurement of this official description is basically reliable.

appraise

Well, the author's review of P1G5 is basically completed here, if there is no accident, there will be no continuation after the next part, although when the article is ready to end, the author checked on the official website and found that the official release of P1G5's new BIOS 1.12...

The test time scheduled by the author is two weeks, and due to various affairs, the time has been forced to be extended and cannot be endless. After all, this article has exceeded 8,000 texts so far, there are many pictures, many readers can read them one by one, thinking that they are already haggard...

Whether the new BIOS 1.12 will bring further optimization, whether the author can ----play the potential of the 230-watt power supply and increase the power consumption of the P1G5 by 10 watts--- remains to be measured by the readers.

To review the author's translation of the official English version of the promotional copy at the end of the previous part:

Do you need performance?

I'm a dream!

Powerful, portable, and first-class screen, the ThinkPad P1 is beautiful and beautiful inside, which is admirable!

The P1 has a hidden spring color under the all-in-one formfactor: the latest Intel Core processor and the most powerful NVIDIA RTX A5500 graphics card.

Liquid metal heat dissipation, long-term work, cooler and more stable!

16-inch narrow border 16:10 screen, screen-to-body ratio greatly increased, low blue light, Dolby Vision and color calibration blessing, ultra-clear display effect, is the ideal dream machine for content creators!

Optional built-in 5G Internet access, the perfect combination of high performance and high portability!

Did the P1G5 do it all?

The author thinks: basically done, but the summary title is wrong, that is, "Do you need performance?" I'm a dream! No", it should be "you need performance!" My dream?".

Why do you say that?

The P1G5 is not a model whose primary purpose is performance, it is also about portability.

Therefore, we can see that the performance of P1G5 has to make a balanced abandonment, the GPU NVIDIA control panel TGP is labeled 100 watts, the official Platform Specifications can only write 80 watts, and the CPU long-term high voltage can only approximate TDP=45 watts---- BIOS 1.11/EC 1.06 is 47 watts, BIOS 1.05/EC 1.03 is 40 watts.

So, the author's final comment on P1G5 is the subtitle of this article:

I am not arrogant and light

I am not arrogant---- although my P1G5 cannot be fully developed in terms of performance,

You have to be light---- but in exchange for the user, you get the benefit of moving briskly.

"I lost my pride and Yang Jun lost the willow, and the Yang Willow lightly went straight to the heavy sky nine.

"I lose pride Yang", from "Butterfly Love Flower Answer Li Shuyi", the author changed "Yang" to "Yang", not Yang Kaihui's Yang, but carried forward Yang.

"Light", that is, gently fluttering, refers to the brisk rippling of boats in the water, from "Return to the Past".

expect

If you pursue the highest performance, you must carry the weight forward!

Therefore, the author's review articles for the flagship heavy-duty mobile workstation ThinkPad P16 Gen 1 with the primary goal of performance, as well as several flagship review articles from other brands, will be sent out one by one as soon as possible.