Everything has a well-defined process to reward. We rate products, mainly laptops, for this article, based on our own judgments that come to by testing in several ways, collecting the information we require and analyze it using the reference set by experts at Thanalysis. Generally, sites that review the products, don't public the process they follow to rate the products. But, in this article, I'm going to make it public. So that you can know the various factors we consider to show you the final evaluation. So, let's get started.

The laptop is consists of many big things like a keyboard, display, touchpad, speakers, and many more. But, several smaller components directly affect the user experience. We analyze the laptop devices by considering 15 main factors and 52 combined sub-factors.

It's very hard to reach 100% in the test that we follow for calculation!

Stars on the blank wall.
Table of Contents

So, I will start to explain with a build.

Build Quality

How we rate the build quality of laptops?

The build quality is the primary factor that directly influences the durability of the machine. To rate the build, we analyze the machine at the following different angles.

  • Flex level
  • Flex extension
  • Material
  • Weight distribution
  • Hinge
  • Weight

Flex is something like when you push the surface and it bends down to withstand the pressure until the surface breaks down. So, in the build quality assessment, we have accounted for flex at two components: Keyboard and Display.

For both keyboard and display flex, if I press it, the flex generated falls into one of the following categories.

  • No flex
  • Little flex but not problematic
  • Average - problematic
  • A lot problematic

Hence, we have assigned the numeric values to the terms listed above. The values are 10 for no flex, 7 for little flex but not problematic, 5 for average flex, and 2 for problematic flex out of 10. Remember that these values apply to both keyboard and display's flexes. Now, how we assess the flex?

For keyboard flex, we press the keyboard down hard with fingers with a fair amount of pressure at three sides simultaneously. First, we press the WASD part followed by the middle part that consists the keys like YUGHJ and the last part is num-pad in case of a full-size keyboard, and the right-hand side part in case of a short keyboard. At the same time, we also the extension level. When you press the keyboard, there are chances that it extends towards the palm rest area.

So, we have taken it into account too. We have given a separate numeric reference value to the flex expansion to the palm rest area for judgment. The values are 1 for no extend, 0.3 for little extend, and 0 for extend out of 1.

For display flex, we just move the display up and down by holding it with hands from the top left and right corners with fair pressure. And, we have applied the same numeric value that we've applied to judge the keyboard flex mathematically. Display flex indirectly gives us information about the strength of the hinge and most probably its size and placement. In the next paragraph, I'll explain the process to judge the material.

For material, we have applied 10 scores for all-metal (Aluminum only) design, 8 scores for all-metal (Magnesium only) design, 8.5 scores for the metal base (Any metal) + plastic lid, 7 scores for metal (any material) lid, and plastic base, and up to 6 scores for plastic out of 10. The score for all-plastic body ranges because different types of plastic material have different strength values. They have different elasticity and fracture points.

Next, we look at whether the display can be opened with just one finger. For that, we have given 10 scores for easily open, 8 for just open, 5 for those having a problem initially, and 0 for those who don't open up out of 10. We judge this because it is directly related to the weight distribution across the hinge. Fairly evenly distributed weight allows to open, but unevenly distributed one doesn't or sometimes allows with problems. So, that's the factor we believe important to consider.

In addition to that, we also take a loot at the hinge for a degree of rotation. Generally, it has no performance value except for user experience. Hence, we have applied 1 score for 180-degree and 2 for 360-degree out of 10. Here is a 0 score for the hinge with less than 180-degree. Next, we have considered the weight of the laptop. The way we followed for weight's judgment is a bit complicated.

For weight, we have taken 1.8 kg or 3.96 lbs ~ 4 lbs as a base. So, if the weight of the laptop falls below 1.8 kg, it will be rewarded in proportional points out of 10. If falls out of the boundary, the points exceed the limit will be deducted from the final calculation. So, indirectly we have set a mechanism to punish the laptop that exceeds the weight limit. In contrast, for reward, we set the point lower that goes close to the base value. Next, I will explain the mechanism we follow to judge the display quality.

Display

How we judge the display mathematically?

The display is one of the components that affect the user experience heavily. We have decided to judge display by considering the following factors.

  • Nits
  • Resolution
  • SRGB %
  • AdobeRGB %
  • Contrast ratio
  • Anti-glare panel
  • Display panel
  • Touch feature
  • Refresh rate
  • G-Sync, V-Sync of FreeSync

For Nits, we have taken 300 as a base since 300 nits are what we consider the lowest value for outdoor usage without experiencing pain in the eyes. So, the laptop that falls behind the limit will be punished and those who cross the limit will be rewarded proportionally. The mechanism we follow here is similar in nature but in contrast to one we follow for the mathematical judgment of weight.

For resolution, the scores were assigned as following out of 4: 0 = HD, 1 = FHD, 2 = QHD, 3 = 4K, 5 = 5K. Here, the display resolution that doesn't fall exactly into any of what we've specified will be calculated in the proportion manner. For example, Apple Macbook's Retina display doesn't fall to the exact option that we have laid down. Hence, we use a proportional approach to get the mathematical value. In this way, we ensure that no device is left behind. Over time, we update the mechanism to include (mostly) other factors.

For color gamut, we consider SRGB and AdobeRGB values. You can check more information about color gamut on our dedicated article here. For judgment, we directly insert the value we got in the form of a percentage out of 100.

For contrast ratio, we have taken 1000:1 as an ideal value. Hence, like Nits' calculation, laptops will be punished if the value falls behind the ideal. For a reward, it must found to be crossed the limit. In this calculation, we have followed the proportion manner.

For reflectivity, we considered the presence of an anti-glare panel here. The limit is only 1, although. So, if not present then will be punished by inserting 0 else rewarded with one point. And, the same mechanism we are following for Panels like IPS. For the panel, we require the display to be an IPS panel ideally.

For the touch feature, we have assigned the scores 10 for smooth, responsive, and accurate, 0 for no availability of the feature, 4 for smooth only, 2.5 for responsive only, and 3.5 for accuracy. Here, there are chances that the display may be a little responsive and accurate but we have considered the lack even if it's a little bit, as a serious issue here. Hence, the display with little less accuracy won't be granted the score proportion to 3.5, it will directly 0. In this way, our approach is to make those laptops highlighted that are at least not designed with half of each feature.

For refresh rate, we have set 0 scores for 60Hz, 1 for 120Hz, 1.3 for 144Hz, and 2 for 240Hz. A 60Hz refresh rate is the most common the laptops. Hence, it doesn't make sense if I reward at least one point there. Take note that the score limit is 2. Over time, it's a matter of upgrade if higher refresh rates come in the market than 240Hz.

For G-Sync, V-Sync, or FreeSync, we just seek the presence. If found, the laptop will be rewarded with 10 points else 0.

Keyboard

The keyboard contributes directly to the user experience (UX). The keyboard must be tactile for better UX. Hence, in this section of this article, I am going to explain the mechanism we follow to decide its ratings. So, we have considered the distance of key's travel, consistency, size and placement of arrow keys, other misplaced keys, backlit and its intensity, keyboard size, and design as an extra.

For the key's travel distance, we have set 1.5mm as an ideal since most laptops with 1.5mm travel reflect tactile feedback. But, that's not always true. So, we reward if travel distance crosses the limit else punish and stay neutral if found exact.

For consistency, we have set 10 points, 5 for inconsistent key travel or tactile feedback, and 0 for no tactile feedback out of 10.

For arrow keys, we seek them to be enough big in size and not closely placed to prevent accidental wrong press. For that, out of 10, we assign 6 scores for its size if it falls under the criteria else we assign 0, and 4 scores for not well placed.

In addition to that, there is a chance that the laptop's keyboard may have other keys than arrows incorrectly equipped. For that, we have set the 2 as a base value which means we allow the maximum of 2 keys for incorrect placement. Although, the ratings will be 0 out of 10. If found more than 2 keys may be pressed incorrectly, we punish the laptop by proportionally deducting the score. Take note that we don't consider the sizes here.

For backlit, we give 10 scores for Per-Key RGB of Single backlit, 5 for zonal, and 0 for no backlit keyboard out of 10. Here, no score deduction will take place.

For the intensity of backlit, we set 10 scores if the backlit illuminates the primary and secondary functions fully, 7 if little less, 5 for medium illumination, 2 for bad, and 0 for no out of 10. However, most of the laptops found to be reached a minimum of 5 scores in this test.

For an extra reward to design, we consider the key's highlight, the shape of the keys, curve or flat on the key (experience will be first considered here), multimedia keys, and many more. For mathematical ratings, we apply a score out of 100 in this test.

For keyboard size, out of 1, we assign 1 for full-size keyboard and 0 for else. Next, I will explain the mechanism we have set to judge the quality of the touchpad.

Touchpad

The touchpad, like the keyboard, contributes heavily to the UX. Hence, it must be evaluated. For that, we only reward the touchpad if it's smooth, responsive, accurate, and uses Windows precision drivers. In addition to that, it must be able to click down anywhere. Furthermore, we have considered its size here too, but very minimum weight is given to it.

For smoothness, we reward it with a 10 score for excellent, 7 for good, 5 for average, 2 for bad, and 1 for extremely poor out of 10. And, the same mechanism to evaluate accuracy and responsiveness.

For Windows Precision drivers, we consider the presence very seriously here. Hence, if present, will be rewarded with 10 points else 0 out of 10.

In addition to that, if it is found to be clicking down anywhere, will get 1 score else 0.

For size, we have considered its width. The width of the touchpad should be at least 4.1-inch. So, if any value falls below the ideal, will be deducted proportionally as a punishment else compensated in the same manner.

Fingerprints on the surface

Fingerprints, if found hard to wipe-off, often devalue the product. Hence, we believe it's very important to take into account before the final evaluation. If you use any product, there will any of the following cases happen in terms of fingerprints.

  • Don't show up
  • Show up but easy to clean
  • Not easy to clean
  • Color change happens but easy to get rid of it
  • Color change but hard to revert

Hence, we have set different mathematical values for the different cases out of 10. We will reward 10 points if it does not show up, 8 if visible but easy to wipe-away, 4 if not easy to clean, 2 if the color change takes place but easy to revert, and 0.5 if hard to revert the color changes.

Noise Emission

Noise emission is only contributed by either fans or hard disks. But, most probably hard disks here. To test the noise level, we use Digitech's sound level meter that will measure the noise in dB. The fans' noise level is always different during a max performance than idle. Hence, each of the modes must be evaluated differently. Since all modes directly contribute to the noise, we have only one way remain - a proportional evaluation - to reward or punish the laptop.

We have set 34dB of fan noise as a reference or ideal and, 46dB for maximum performance provided no CPU undervolting. The evaluation method is going to be the same here. If the test value falls below the reference one, the laptop will be rewarded else punished in the form of deduction in the traditional manner we are using.

Ports

The type of port always matters a lot. In this test, we have chosen three-port as following to evaluate the laptop's final score.

  • A thunderbolt port
  • A display port
  • A graphics amplifier port

Each of the ports mentioned above has set to different points based on the consideration of the importance of presence. In this test, we can only consider the availability here. There is no proportional deduction or addition. Hence, if found not present, will be eligible for 0 points else the preset value.

For the thunderbolt port, we have assigned 50 points since it's the most important port of the machine, irrespective of its generations like Gen-1 and Gen-2. For DisplayPort, 20 points are given. And last, only 10 points are given to the Graphics Amplifier port.

However, take note that the absence of any of the unimportant ports doesn't mean we will eliminate the base value from the final base total. The laptops will get benefit if three of the ports are available else the 0 points will surely make damage their final score.

Gaming Benchmarks

In gaming benchmarks, we consider The Witcher 3 PC game. Because our tests and other experiments found it using the CPU and GPU in a good combination. Hence, if not playable on the machine then the laptop will definitely get 0 points of 100. But, if playable, the score will be decided concerning the maximum refresh theoretical refresh rate.

So, in this test, We have skipped to include the 3DMark score since it doesn't put any real-world sense. Next, we will look at the mechanism we follow to evaluate the CPU temperature for final ratings.

Temperature

Here, in this test, we have considered the CPU temperature at max and idle and exterior temperature. Take a note that we haven't considered an external GPU temperature if present. The method for ratings is not different here. We have set the boundary and if the laptop's component tries to cross it, it will be punished in the form of proportional deduction.

For CPU temperature, we have set 80 degrees C as a reference value for maximum mode during CPU+GPU stress test using AIDA64 and 50 degrees C for idle mode. And, for exterior temperature, we have set 35 degrees C for the keyboard and 25 for the palm rest during the stress test.

Battery life

The battery is one of the major contributors to the laptop's performance. We always seek a higher battery life in the laptop. To judge the battery life, we follow the ratio of battery capacity in Wh to battery life in minutes. Here, the battery life that we consider in the ratio is measured while watching YouTube videos at 100% of speakers level, 50% of screen brightness, background apps disabled, and keeping keyboard backlit and other back-lighting off. Take note that there is no punishment as we don't deduct the score.

Camera

We assess the camera quality mainly in the room with natural light for video conference. After that, the score will be assigned out of 10. We give a fixed score of 10 for an excellent result, 7 for just good, 5 for average, 2 for bad, and 1 for extremely poor.

In addition to that, Windows Hello is also gaining a massive attraction nowadays. So, we check its presence on the laptop. If found, the 10 scores directly credited for the final evaluation else 0. Also, note that the absence won't make the laptop immune from counting 10 in the total score.

Microphone

We assess the microphone quality mainly outdoor for video conferences. After that, the score will be assigned out of 10. We give a fixed score of 10 for an excellent result, 7 for just good, 5 for average, 2 for bad, and 1 for extremely poor.

Hence, consider yourself lucky if you got the laptop with a 10 score assigned to the microphone.

Fingerprint Reader

The fingerprint reader doesn't contribute much to the performance of the laptop. However, it does contribute to the user experience. And, it was also found to be absent on many laptops.

So, we assign a 10 score for its presence and another 10 for its support for Windows Hello.

Wireless Connectivity

For the evaluation of wireless connectivity, we only consider the latest Wi-Fi version. In addition to that, we also consider here Bluetooth and Infrared, but both have given the lower score weight.

For the presence of Wi-Fi 6, we reward the laptop with a 10 score. And, there will be only 1 score assigned for the availability of Bluetooth 5 and 0.5 for IR.

Standard Benchmarks

We do take into account the proportion of several standard benchmarks like Cinebench R20, Geekbench 5, and many more.

Read other related articles:

Laptops