At techjazzup, we’ve been reviewing computers for a long time, and we evaluate a lot of them. Some of the technique has evolved throughout time, but our essential dedication to thorough product assessments has not. But how precisely do we evaluate things that we have tested?
Two steps make up the review process for laptops, desktops, tablets, and other computer-like devices: performance testing under controlled settings in the techjazzup Labs, and comprehensive hands-on use by our expert reviewers. This entails assessing the functionality, ergonomics, and aesthetics of a gadget. These two types of evaluations—objective and subjective—are combined to provide a final review judgment.
We set up computers that come to the techjazzup Labs—typically a laptop, desktop, two-in-one hybrid, or Chromebook—in the same way, that the average user would. As a best practice, we turn off as many intrusive privacy and data-collecting settings as we can during setup. Then, we update the OS, GPU drivers, BIOS, and manufacturer utilities if necessary. Finally, we utilize a program like SiSoftware’s Sandra to gather data on the system’s parts, including the CPU, GPU, and RAM.
We execute a basic set of benchmark tests on each compatible system. There is also a more extensive set of tests for certain use cases, including gaming or content production, where systems would need to be assessed for more potent GPUs or screens with higher resolutions.
As the devices we evaluate advance over time, so does the array of benchmarking programs we employ. We are now doing the following core checks on each suitable computer:
Monkey Labs 5 Geekbench
We execute the Vulcan (Windows) or Metal (MacOS) Compute test, as well as single-core and multicore CPU tests. We do the CPU and compute tests on Android, Apple, and Chromebooks. The CPU tests on Geekbench gauge a mixed workload’s performance.
Movie Bench R23
On computers running Windows and macOS, we conduct both single-core and multi-core tests. Cinebench gauges the efficiency of the CPU in 3D rendering.
10 PCMark
This Windows benchmark is being phased out, but we continue to use the previous edition, which mimics a variety of activities such as online surfing, video conferencing, picture editing, video editing, and more.
Wild Life Extreme 3DMark
One of the few cross-platform benchmarks for measuring graphics performance, we conduct this test on macOS (Apple silicon), Windows, Android, and iPadOS computers. Additionally, we run it in Unlimited mode, which removes screen resolution as a factor when comparing results across different devices.
Time Spy, 3DMark Fire Strike Ultra, and Port Royal
To assess a system’s DirectX 11 or DirectX 12 graphics performance, which is crucial for gaming PCs, we conduct these tests on any system with a discrete GPU. Port Royal, a raytracing performance test initially created for Nvidia GPUs, is being phased out in favor of 3DMark’s DXR test. Additionally, we are implementing tests for PCI Features, Storage, and CPU Profiler.
Far Cry 5 Standard
This game benchmark is being phased out, however for the time being it is being performed at the High-quality setting at 1,920×1,080 resolution. We are gradually adding the complicated simulation game RiftBreaker’s CPU and GPU benchmarks as well as Marvel’s Guardians of the Galaxy in-game benchmark, both in 1080p resolution and with the corresponding High-quality settings. Systems with discrete GPUs at the low end can support both games.
Benchmark: Shadow of the Tomb Raider
This older game may function effectively on less expensive gaming hardware since it balances the CPU and GPU loads rather than depending solely on the GPU. On computers with a discrete GPU, we run the game’s built-in benchmark at the highest quality setting and 1,920×1,080 resolution.
UL Procyon tests for video and photo editing
We utilize Adobe Premiere Pro and Photoshop with Photoshop Lightroom Classic benchmarks at a 1,920×1,080-pixel resolution to assess a system’s potential for content production if it satisfies the minimal requirements to run these programs.
Test of battery life
We adjust the settings on all battery-operated computers so that they won’t go to sleep or hibernate, turn off pop-ups and alerts that can obstruct the test, and set the screen brightness and volume (output to headphones) to 50%. After that, we use Chrome to broadcast a looping, customized YouTube movie over Wi-Fi while keeping track of how long the system is operational with a timer app.
MotionMark, WebXPRT 3, and JetStream
To assess Chromebook performance, we conduct these browser-based tests. Occasionally, we also run them on Windows machines for comparison.
Even more testing
To offer further context for a system’s performance, we may conduct a variety of extra tests or variants on the common tests, such as Cinebench and Geekbench 5 on battery power.
If supported by extra in-game benchmarks like Watch Dogs: Legion or DLSS (on Nvidia) or FidelityFX Super Resolution (on AMD), discretionary testing may additionally incorporate those technologies. we occasionally run SpecViewPerf 2020 on computers with midrange and above GPUs (professional content creation and analysis beyond photo and video editing).
As part of a review, we usually include a comparison table of findings from relevant tests performed on connected products. Whenever we make substantial changes to testing, such as moving from one version of a test to another, we double-test both iterations of the test or the whole old and new set in order to build a database of comparative data.
These test results typically match our predictions since there are so many machines utilizing the same few CPUs and GPUs, the same operating systems, and comparable amounts of RAM and storage. That implies you may get a good notion of a system’s performance by glancing at its specifications on paper. The intriguing part begins when it deviates from those predictions, and we look into the reason why. One of the reasons we test so thoroughly is to find these edge circumstances so you don’t buy a lemon. Sometimes it’s bad hardware, other times its bad optimization or a compatibility problem.
Here are some of our current top picks, which we’ve chosen based on our own usage, this testing, the system’s design, and usability, and, most importantly, its total value. Because I’ve been a part of our team for a while, I want everyone who reads a techjazzup review to know that I value your money just as much as you do.
READ MORE HERE:
Best New Budget Laptop Under $500 for 2022
5 Best Gaming Laptop for 2022.
