BenchmarkXPRT Blog banner

Category: Performance benchmarking

Web AI frameworks: Possible paths for the AI-focused WebXPRT 4 auxiliary workload

A few months ago, we announced that we’re moving forward with the development of a new auxiliary WebXPRT 4 workload focused on local, browser-side AI technology. Local AI has many potential benefits, and it now seems safe to say that it will be a common fixture of everyday life for many people in the future. As the growth of browser-based inference technology picks up steam, our goal is to equip WebXPRT 4 users with the ability to quickly and reliably evaluate how well devices can handle substantial local inference tasks in the browser.

To reach our goal, we’ll need to make many well-researched and carefully considered decisions along the development path. Throughout the decision-making process, we’ll be balancing our commitment to core XPRT values, such as ease of use and widespread compatibility, with the practical realities of working with rapidly changing emergent technologies. In today’s blog, we’re discussing one of the first decision points that we face—choosing a Web AI framework.

AI frameworks are suites of tools and libraries that serve as building blocks for developers to create new AI-based models and apps or integrate existing AI functions in custom ways. AI frameworks can be commercial, such as OpenAI, or open source, such as Hugging Face, PyTorch, and TensorFlow. Because the XPRTs are available at no cost for users and we publish our source code, open-source frameworks are the right choice for WebXPRT.

Because the new workload will focus on locally powered, browser-based inference tasks, we also need to choose an AI framework that has browser integration capabilities and does not rely on server-side computing. These types of frameworks—called Web AI—use JavaScript (JS) APIs and other web technologies, such as WebAssembly and WebGPU, to run machine learning (ML) tasks on a device’s CPU, GPU, or NPU.

Several emerging Web AI frameworks may provide the compatibility and functionality we need for the future WebXPRT workload. Here are a few that we’re currently researching:

  • ONNX Runtime Web: Microsoft and other partners developed the Open Neural Network Exchange (ONNX) as an open standard for ML models. With available tools, users can convert models from several AI frameworks to ONNX, which can then be used by ONNX Runtime Web. ONNX Runtime Web allows developers to leverage the broad compatibility of ONNX-formatted ML models—including pre-trained vision, language, and GenAI models—in their web applications.
  • Transformers.js: Transformers.js, which uses ONNX Runtime Web, is a JS library that allows users to run AI models from the browser and offline. Transformers.js supports language, computer vision, and audio ML models, among others.
  • MediaPipe: Google developed MediaPipe as a way for developers to adapt TensorFlow-based models for use across many platforms in real-time on-device inference applications such as face detection and gesture recognition. MediaPipe is particularly useful for inference work in images, videos, and live streaming.
  • TensorFlow.js: TensorFlow has been around for a long time, and the TensorFlow ecosystem provides users with a broad variety of models and datasets. TensorFlow is an end-to-end ML solution—training to inference—but with available pre-trained models, developers can focus on inference. TensorFlow.js is an open-source JS library that helps developers integrate TensorFlow with web apps.

We have not made final decisions about a Web AI framework or any aspect of the future workload. We’re still in the research, discussion, and experimentation stages of development, but we want to be transparent with our readers about where we are in the process. In future blog posts, we’ll discuss some of the other major decision points in play.

Most of all, we invite you to join us in these discussions, make recommendations, and give us any other feedback or suggestions you may have, so please feel free to share your thoughts!

Justin

How to automate your WebXPRT 4 testing

We’re excited about the ongoing upward trend in the number of completed WebXPRT 4 runs that we’re seeing each month. OEM and tech press labs are responsible for a significant amount of that growth, and many of them use WebXPRT’s automation features to complete large blocks of hands-off testing at one time. We realize that many new WebXPRT users may be unfamiliar with the benchmark’s automation tools, so we decided to provide a quick guide to WebXPRT automation in today’s blog. Whether you’re testing one or 1,000 devices, the instructions below can help save you some time.

WebXPRT 4 allows users to run scripts in an automated fashion and control test execution by appending parameters and values to the WebXPRT URL. Three parameters are available:

  • test type
  • test scenarios
  • results

Below, you’ll find a description of those parameters and instructions for how you can use them to automate your test runs.

Test type

The WebXPRT automation framework accounts for two test types: (1) the six core workloads, and (2) any experimental workloads we might add in future builds. There are currently no experimental tests in WebXPRT 4, so always set the test type variable to 1.

  • Core tests: 1

Test scenario

The test scenario parameter lets you specify which subtest workloads to run by using the following codes:

  • Photo enhancement: 1
  • Organize album using AI: 2
  • Stock option pricing: 4
  • Encrypt notes and OCR scan using WASM: 8
  • Sales graphs: 16
  • Online homework: 32

To run a single subtest workload, use its code. To run multiple workloads, use the sum of their codes. For example, to run Stock options pricing (4) and Encrypt notes and OCR scan (8), use the sum of 12. To run all core tests, use 63, the sum of all the individual test codes (1 + 2 + 4 + 8 + 16 + 32 = 63).

Results format

The results format parameter lets you select the format of the results:

  • Display the result as an HTML table: 1
  • Display the result as XML: 2
  • Display the result as CSV: 3
  • Download the result as CSV: 4

To use the automation feature, start with the URL https://www.principledtechnologies.com/benchmarkxprt/webxprt/2021/wx4_build_3_7_3, append a question mark (?), and add the parameters and values separated by ampersands (&). For example, to run all the core tests and download the results, you would use the following URL: https://principledtechnologies.com/benchmarkxprt/webxprt/2021/wx4_build_3_7_3/auto.php?testtype=1&tests=63&result=4

We hope WebXPRT 4’s automation features will make testing easier for you. If you have any questions about WebXPRT or the automation process, please feel free to ask!

Justin

Shopping for back-to-school tech? The XPRTs can help!

For many students, the first day of school is just around the corner, and it’s now time to shop for new tech devices that can help set them up for success in the coming year. The tech marketplace can be confusing, however, with so many brands, options, and competing claims to sort through.

Fortunately, the XPRTs are here to help!

Whether you’re shopping for a new phone, tablet, Chromebook, laptop, or desktop, the XPRTs can provide industry-trusted performance scores that can give you confidence that you’re making a smart purchasing decision.

The WebXPRT 4 results viewer is a good place to start looking for device scores. The viewer displays WebXPRT 4 scores from over 700 devices—including many of the latest releases—and we’re adding new scores all the time. To learn more about the viewer’s capabilities and how you can use it to compare devices, check out this blog post.

Another resource we offer is the XPRT results browser. The browser is the most efficient way to access the XPRT results database, which currently holds more than 3,700 test results from over 150 sources, including major tech review publications around the world, manufacturers, and independent testers. It offers a wealth of current and historical performance data across all the XPRT benchmarks and hundreds of devices. You can read more about how to use the results browser here.

Also, if you’re considering a popular device, there’s a good chance that a recent tech review includes an XPRT score for that device. There are two quick ways to find these reviews: You can either (1) search for “XPRT” on your preferred tech review site or (2) use a search engine and input the device name and XPRT name, such as “Dell XPS” and “WebXPRT.”

Here are a few recent tech reviews that use one of the XPRTs to evaluate a popular device:

Lastly, here at Principled Technologies, we frequently publish reports that evaluate the performance of hot new consumer devices, and many of those reports include WebXPRT scores. For example, check out our extensive testing of HP ZBook G10 mobile workstations or our detailed comparison of Lenovo ThinkPad, ThinkBook, and ThinkCentre devices to their Apple Mac counterparts.

The XPRTs can help anyone stuck in the back-to-school shopping blues make better-informed and more confident tech purchases. As this new school year begins, we hope you’ll find the data you need on our site or in an XPRT-related tech review. If you have any questions about the XPRTs, XPRT scores, or the results database please feel free to ask!

Justin

Putting together a good WebXPRT workload proposal

Recently, we announced that we’re moving forward with the development of a new AI-focused WebXPRT 4 workload. It will be an auxiliary workload, which means that it will run as a separate, optional test, and it won’t affect existing WebXPRT 4 tests or scores. Although the inspiration for this new workload came from internal WebXPRT discussions—and, let’s face it, from the huge increase in importance of AI—we wanted to remind you that we’re always open to hearing your WebXPRT workload ideas. If you’d like to submit proposals for new workloads, you don’t have to follow a formal process. Just contact us, and we’ll start the conversation.

If you do decide to send us a workload proposal, it will be helpful to know the types of parameters that we keep in mind. Below, we discuss some of the key questions we ask when we evaluate new WebXPRT workload ideas.

Will it be relevant and interesting to real users, lab testers, and tech reviewers?

When considering a WebXPRT workload proposal, the first two criteria are simple: is it relevant in real life, and are people interested in the workload? We created WebXPRT to evaluate device performance using web-based tasks that consumers are likely to experience daily, so real-life relevance has always been an essential requirement for us throughout development. There are many technologies, functions, and use cases that we could test in a web environment, but only some are relevant to common applications or usage patterns and are likely to draw the interest of real users, lab testers, and technical reviewers.

Will it have cross-platform support?

Currently, WebXPRT runs on almost any web browser and almost every device that supports a web browser. We would like to keep that level of cross-platform support when we introduce new workloads. However, technical differences in how various browsers execute tasks make it challenging to include certain scenarios without undermining our cross-platform ideal. When considering any workload proposal, one of the first questions we ask is, “Will it work on all the major browsers and operating systems?”

There are special exceptions to this guideline. For instance, we’re still in the early days of browser-based AI, and it’s unlikely that a new browser-based AI workload will run on every major browser. If it’s a particularly compelling idea, such as the AI scenario we’re currently working on, we may consider including it as an auxiliary test.

Will it differentiate performance between different types of devices?

XPRT benchmarks provide users with accurate measures for evaluating how well target systems or technologies perform specific tasks. With a broadly targeted benchmark like WebXPRT, if the workloads are so heavy that most devices can’t handle them or so light that most devices complete them without being taxed, the results will be of little use for helping buyers evaluating systems and making purchasing decisions, OEM labs, and the tech press.

That’s why, with any new WebXPRT workload, we look for a sweet spot with respect to how computationally demanding it will be. We want it to run on a wide range of devices—from low-end devices that are several years old to brand-new high-end devices, and everything in between. We also want users to see a wide range of workload scores and resulting overall scores that accurately reflect the experiences those systems deliver, so they can easily grasp the different performance capabilities of the devices under test.

Will results be consistent and easily replicated?

Finally, WebXPRT workloads should produce scores that consistently fall within an acceptable margin of error and are easily replicated with additional testing or comparable gear. Some web technologies are very sensitive to uncontrollable or unpredictable variables, such as internet speed. A workload that measures one of those technologies would be unlikely to produce results that are consistent and easily replicated.

We hope this post will be useful if you’re thinking about potential new workloads that you’d like to see in WebXPRT. If you have any general thoughts about browser performance testing or specific workload ideas that you’d like us to consider, please let us know.

Justin

Contribute to WebXPRT’s AI capabilities with your NPU-equipped gear

A few weeks ago, we announced that we’re developing a new auxiliary WebXPRT 4 workload focused on local, browser-based AI technology. This is an exciting project for us, and as we work to determine the best approach from the perspective of frameworks, APIs, inference models, and test scenarios, we’re also thinking ahead to the testing process. To best understand how the new workload will impact system performance, we’re going to need to test it on hardware equipped with the latest generation of neural processing units (NPUs).

NPUs are not new, but the technology is advancing rapidly, and a growing number of PC and laptop manufacturers are releasing NPU-equipped systems. Several vendors have announced plans to release systems equipped with all-new NPUs in the latter half of this year. As is often the case with bleeding-edge technology, however, official release dates do not always coincide with widespread availability.

We want to evaluate new AI-focused WebXPRT workloads on the widest possible range of new systems, but getting a wide selection of gear equipped with the latest NPUs may take quite a while through normal channels. For that reason, we’ve decided to ask our readers for help to expedite the process.

If you’re an OEM or vendor representative with access to the latest generation of NPU-equipped gear and want to contribute to WebXPRT’s evolution, consider sending us any PCs, white boxes, laptops, 2-in-1s, or tablets (on loan) that would be suitable for NPU-focused testing. We have decades of experience serving as trusted testers of confidential and pre-release gear, so we’re well-acquainted with concerns about confidentiality that may come into play, and we won’t publish any information about the systems or related test results without your permission.

We will, though, be happy to share with you our test results on your systems, and we’d love to hear any guidance or other feedback from you on this new workload.

We’re open to any suitable gear, but we’re especially interested in AMD Ryzen AI, Apple M4, Intel Lunar Lake and Arrow Lake, and Qualcomm Snapdragon X Elite systems.

If you’re interested in sending us gear for WebXPRT development testing, please contact us. We’ll work out all the necessary details. Thanks in advance for your help!

Justin

Updating our WebXPRT 4 browser performance comparisons with new gear

Once or twice per year, we refresh an ongoing series of WebXPRT comparison tests to see if recent updates have reordered the performance rankings of popular web browsers. We published our most recent comparison in January, when we used WebXPRT 4 to compare the performance of five browsers on the same system.

This time, we’re publishing an updated set of comparison scores sooner than we normally would because we chose to move our testing to a newer reference laptop. The previous system—a Dell XPS 13 7930 with an Intel Core i3-10110U processor and 4 GB of RAM—is now several years old. We wanted to transition to a system that is more in line with current mid-range laptops. By choosing to test on a capable mid-tier laptop, our comparison scores are more likely to fall within the range of scores we would see from an typical user today.

Our new reference system is a Lenovo ThinkPad T14s Gen 3 with an Intel Core i7-1270P processor and 16 GB of RAM. It’s running Windows 11 Home, updated to version 23H2 (22631.3527). Before testing, we installed all current Windows updates and tested on a clean system image. After the update process was complete, we turned off updates to prevent any further updates from interfering with test runs. We ran WebXPRT 4 three times each on five browsers: Brave, Google Chrome, Microsoft Edge, Mozilla Firefox, and Opera. In Figure 1 below, each browser’s score is the median of the three test runs.

In our last round of tests—on the Dell XPS 13—the four Chromium-based browsers (Brave, Chrome, Edge, and Opera) produced close scores, with Edge taking a small lead among the four. Each of the Chromium browsers significantly outperformed Firefox, with the slowest of the Chromium browsers (Brave) outperforming Firefox by 13.5 percent.

In this round of tests—on the Lenovo ThinkPad T14s—the scores were very tight, with a difference of only 4 percent between the last-place browser (Brave) and the winner (Chrome). Interestingly, Firefox no longer trailed the four Chromium browsers—it was squarely in the middle of the pack.

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

Unlike previous rounds that showed a higher degree of performance differentiation between the browsers, scores from this round of tests are close enough that most users wouldn’t notice a difference. Even if the difference between the highest and lowest scores was substantial, the quality of your browsing experience will often depend on factors such as the types of things you do on the web (e.g., gaming, media consumption, or multi-tab browsing), the impact of extensions on performance, and how frequently the browsers issue updates and integrate new technologies, among other things. 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 to test the speed of different browsers on the same system? If so, we’d love for you to tell us about it! Also, please tell us what other WebXPRT data you’d like to see!

Justin

Check out the other XPRTs:

Forgot your password?