BenchmarkXPRT Blog banner

Category: browser performance

Browser-based AI tests in WebXPRT 4: face detection and image classification

I recently revisited an XPRT blog entry that we posted from CES Las Vegas back in 2020. In that post, I reflected on the show’s expanded AI emphasis, and I wondered if we were reaching a tipping point where AI-enhanced and AI-driven tools and applications would become a significant presence in people’s daily lives. It felt like we were approaching that point back then with the prevalence of AI-powered features such as image enhancement and text recommendation, among many others. Now, seamless AI integration with common online tasks has become so widespread that many people unknowingly benefit from AI interactions several times a day.

As AI’s role in areas like everyday browser activity continues to grow—along with our expectations for what our consumer devices should be able to handle—reliable AI-oriented benchmarking is more vital than ever. We need objective performance data that can help us understand how well a new desktop, laptop, tablet, or phone will handle AI tasks.

WebXPRT 4 already includes timed AI tasks in two of its workloads: the “Organize Album using AI” workload and the “Encrypt Notes and OCR Scan” workload. These two workloads reflect the types of light browser-side inference tasks that are now fairly common in consumer-oriented web apps and extensions. In today’s post, we’ll provide some technical information about the Organize Album workload. In a future post, we’ll do the same for the Encrypt Notes workload.

The Organize Album workload includes two different timed tasks that reflect a common scenario of organizing online photo albums. The workload utilizes the AI inference and JavaScript capabilities of the WebAssembly (Wasm) version of OpenCV.js—an open-source computer vision and machine learning library. In WebXPRT 4, we used OpenCV.js version 4.5.2.

Here are the details for each task:

  • The first task measures the time it takes to complete a face detection job with a set of five 720 x 480 photos that we sourced from commercial photo sites. The workload loads a Caffe deep learning framework model (res10_300x300_ssd_iter_140000_fp16.caffemodel) using the commands found here
  • The second task measures the time it takes to complete an image classification job (labeling based on object detection) with a different set of five 718 x 480 photos that we sourced from the ImageNet computer vision dataset. The workload loads an ONNX-based SqueezeNet machine learning model (squeezenet.onnx v 1.0) using the commands found here.

To produce a score for each iteration of the workload, WebXPRT calculates the total time that it takes for a system to organize both albums. In a standard test, WebXPRT runs seven iterations of the entire six-workload performance suite before calculating an overall test score. You can find out more about the WebXPRT results calculation process here.

We hope this post will give you a better sense of how WebXPRT 4 measures one kind of AI performance. As a reminder, if you want to dig into the details at a more granular level, you can access the WebXPRT 4 source code for free. In previous blog posts, you can find information about how to access and use the code. You can also read more about WebXPRT’s overall structure and other workloads in the Exploring WebXPRT 4 white paper.

If you have any questions about this workload or any other aspect of WebXPRT 4, please let us know!

Justin

Best practices for WebXPRT testing

One of the strengths of WebXPRT is that it’s a remarkably easy benchmark to run. Its upfront simplicity attracts users with a wide range of technical skills—everyone from engineers in cutting-edge OEM labs to veteran tech journalists to everyday folks who simply want to test their gear’s browser performance. With so many different kinds of people running the test each day, it’s certain that at least some of them use very different approaches to testing. In today’s blog, we’re going to share some of the key benchmarking practices we follow in the XPRT lab—and encourage you to consider—in order to produce the most consistent and reliable WebXPRT scores.

We offer these best practices as tips you might find useful in your testing. Each step relates to evaluating browser performance with WebXPRT, but several of these practices will apply to other benchmarks as well.

  • Test with clean images: In the XPRT lab, we typically use an out-of-box (OOB) method for testing new devices. OOB testing means that other than running the initial OS and browser version updates that users are likely to run after first turning on the device, we change as little as possible before testing. We want to assess the performance that buyers are likely to see when they first purchase the device and before they install additional software. This approach is the best way to provide an accurate assessment of the performance retail buyers will experience from their new devices. That said, the OOB method is not appropriate for certain types of testing, such as when you want to compare largely identical systems or when you want to remove as much pre-loaded software as possible. The OOB method is less relevant to users who want to see how their device performs as it is.
  • Browser updates can have a significant impact: Most people know that different browsers often produce different performance scores on the same system. They may not know that there can be shifts in performance between different versions of the same browser. While most browser updates don’t have a large impact on performance, a few updates have increased (or even decreased) browser performance by a significant amount. For this reason, it’s always important to record and disclose the extended browser version number for each test run. The same principle applies to any other relevant software.
  • Turn off automatic updates: We do our best to eliminate or minimize app and system updates after initial setup. Some vendors are making it more difficult to turn off updates completely, but you should always double-check update settings before testing. On Windows systems, the same considerations apply to turning off User Account Control notifications.
  • Let the system settle: Depending on the system and the OS, a significant amount of system-level activity can be going on in the background after you turn it on. As much as possible, we like to wait for a stable baseline (idle time) of system activity before kicking off a test. If we start testing immediately after booting the system, we often see higher variance in the first run before the scores start to tighten up.
  • Run the test more than once: Because of natural variance, our standard practice in the XPRT lab is to publish a score that represents the median of three to five runs, if not more. If you run a benchmark only once and the score differs significantly from other published scores, your result could be an outlier that you would not see again under stable testing conditions or over the course of multiple runs.
  • Clear the cache: Browser caching can improve web page performance, including the loading of the types of JavaScript and HTML5 assets that WebXPRT uses in its workloads. Depending on the platform under test, browser caching may or may not significantly change WebXPRT scores, but clearing the cache before testing and between each run can help improve the accuracy and consistency of scores.

We hope these tips will serve as a good baseline methodology for your WebXPRT testing. If you have any questions about WebXPRT, the other XPRTs, or benchmarking in general, please let us know!

Justin

February 2025 WebXPRT 4 browser performance comparisons

Once or twice per year, we refresh our ongoing series of WebXPRT comparison tests to see if software version updates have reordered the performance rankings of popular web browsers. We published our most recent comparison last June, when we used WebXPRT 4 to compare the performance of five browsers—Brave, Google Chrome, Microsoft Edge, Mozilla Firefox, and Opera—on a Lenovo ThinkPad T14s Gen 3. When assessing performance differences, it’s worth noting that all the browsers—except for Firefox—are built on a Chromium foundation. In the last round of tests, the scores were very tight, with a difference of only four percent between the last-place browser (Brave) and the winner (Chrome). Firefox’s score landed squarely in the middle of the pack.

Recently, we conducted a new set of tests to see how performance scores may have changed. To maintain continuity with our last comparison, we stuck with the same ThinkPad T14s as our reference system. That laptop is still in line with current mid-range laptops, so our comparison scores are likely to fall within the range of scores we would see from a typical user today. The ThinkPad is equipped with an Intel Core i7-1270P processor and 16 GB of RAM, and it’s running Windows 11 Pro, version 23H2 (22631.4890).

Before testing, we installed all current Windows updates, and we updated each of the browsers to the latest available stable version. After the update process was complete, we turned off updates to prevent any interference with test runs. We ran WebXPRT 4 five times on each of the five browsers. In Figure 1 below, each browser’s score is the median of the five test runs.

In this round of tests, the gap widened a bit between first and last place scores, with a difference of just over six percent between the lowest median score of 303 (Brave) and the highest median score of 322 (Firefox).

Figure 1: The median scores from running WebXPRT 4 five times with each browser on the Lenovo ThinkPad T14s Gen 3.

In this round of tests, the distribution of scores indicates that most users would not see a significant performance difference if they switched between the latest versions of these browsers. The one exception may be a change from the latest version of Brave to the latest version of Firefox. Even then, the quality of your browsing experience will often depend on other factors. The types of things you do on the web (e.g., gaming, media consumption, or multi-tab browsing), the type and number of extensions you’ve installed, and how frequently the browsers issue updates and integrate new technologies—among other things—can all affect browser performance over time. It’s important to keep such variables in mind when thinking about how browser performance comparison results may translate to your everyday web experience.

Have you tried using WebXPRT 4 in your own browser performance comparison? If so, we’d love to hear about it! Also, please let us know if there are other types of WebXPRT comparisons you’d like to see!

Justin

Recent XPRT mentions in the global tech press

One way we assess the XPRTs’ ongoing effectiveness is to regularly track the reach of our benchmarks in the global tech press. If tech journalists decide to include an XPRT benchmark in their suite of “go-to” performance evaluation tools, we know that decision reflects a high degree of confidence in the relevance and reliability of our benchmarks. It’s especially exciting for us to see the XPRTs win the trust of more tech press outlets in an ever-increasing number of countries around the world.

Because some of our newer readers may be unaware of the wide variety of tech press outlets that use the XPRTs, we occasionally like to share an overview of recent XPRT-related global tech press activity. For today’s blog, we want to give readers a sampling of the press mentions we’ve seen over the past few months.

Recent mentions include:

If you’d like to receive monthly updates on XPRT news, we encourage you to sign up for the BenchmarkXPRT Development Community newsletter. Each month, the newsletter delivers a summary of the previous month’s XPRT-related activity, including XPRT blog posts and new mentions of the XPRTs in the tech press. If you don’t currently receive the monthly BenchmarkXPRT newsletter but would like to join the mailing list, please let us know! It’s free to join. We won’t publish, share, or sell any of the contact information you provide, and we’ll send you only the monthly newsletter and occasional benchmark-related announcements, such as news about patches or new releases.

If you have any questions about the XPRTs, suggestions for improvement, or requests for future blogs, please just contact us.

Justin

The XPRTs: What would you like to see in 2025?

If you’re a new follower of the XPRT family of benchmarks, you may not be aware of one of the characteristics of the XPRTs that sets them apart from many benchmarking efforts—our openness and commitment to valuing the feedback of tech journalists, lab engineers, and anyone else that uses the XPRTs on a regular basis. That feedback helps us to ensure that as the XPRTs grow and evolve, the resources we offer will continue to meet the needs of those that use them.

In the past, user feedback has influenced specific aspects of our benchmarks, such as the length of test runs, UI features, results presentation, and the addition or subtraction of specific workloads. More broadly, we have also received suggestions for entirely new XPRTs and ways we might target emerging technologies or industry use cases.

As we look forward to what’s in store for the XPRTs in 2025, we’d love to hear your ideas about new XPRTs—or new features for existing XPRTs. Are you aware of hardware form factors, software platforms, new technologies, or prominent applications that are difficult or impossible to evaluate using existing performance benchmarks? Should we incorporate additional or different technologies into existing XPRTs through new workloads? Do you have suggestions for ways to improve any of the XPRTs or XPRT-related tools, such as results viewers?

We’re especially interested in your thoughts about the next steps for WebXPRT. If our recent blog posts about the potential addition of an AI-focused auxiliary workload, what a WebXPRT battery life test would entail, or possible WebAssembly-based test scenarios have piqued your interest, we’d love to hear your thoughts!

We’re genuinely interested in your answers to these questions and any other ideas you have, so please feel free to contact us. We look forward to hearing your thoughts and working together to figure out how they could help shape the XPRTs in 2025!

Justin

More than two million XPRT benchmark runs and downloads!

As we near the end of 2024, we’re excited to share that the XPRTs have passed another notable milestone—over 2,000,000 combined runs and downloads! The rate of growth in the total number of XPRT runs and downloads is exciting. It took about seven and a half years for the XPRTs to pass one million total runs and downloads—but it’s taken less than half that, three and a half years, to add another million. Figure 1 shows the climb to the two-million-run mark.

Figure 1: The cumulative number of total yearly XPRT runs and downloads over time.

As you would expect, most of the runs contributing to that total come from WebXPRT tests. If you’ve run WebXPRT in any of the 983 cities and 84 countries from which we’ve received completed test data—including newcomers El Salvador, Malaysia, Morocco, and Saudi Arabia—we’re grateful for your help in reaching this milestone! As Figure 2 illustrates, WebXPRT use has grown steadily since the debut of WebXPRT 2013. On average, we now record more than twice as many WebXPRT runs each month than we recorded in WebXPRT’s entire first year. With over 340,000 runs so far in 2024—an increase of more than 16 percent over last year’s total—that growth is showing no signs of slowing down.

Figure 2: The cumulative number of total yearly WebXPRT runs over time.

This milestone isn’t just about numbers. Establishing and maintaining a presence in the industry and experiencing year-over-year growth requires more than technical know-how and marketing efforts. It requires the ongoing trust and support of the benchmarking community—including OEM labs, the tech press, and independent computer enthusiasts—and those who simply want to know how good their devices are at web browsing.

Once again, we’re thankful for the support of everyone who’s used the XPRTs over the years, and we look forward to another million!

If you have any questions or comments about any of the XPRTs, we’d love to hear from you!

Justin

Check out the other XPRTs:

Forgot your password?