BenchmarkXPRT Blog banner

Category: AI

AI is the heartbeat of CES 2019

This year’s CES features a familiar cast of characters: gigantic, super-thin 8K screens; plenty of signage promising the arrival of 5G; robots of all shapes, sizes, and levels of competency; and acres of personal grooming products that you can pair with your phone. In all seriousness, however, one main question keeps coming to mind as I walk the floor: Are we approaching the tipping point where AI truly starts to affect most people in meaningful ways on a daily basis? I think we’re still a couple of years away from ubiquitous AI, but it’s the heartbeat of this year’s show, and it’s going play a part in almost everything we do in the very near future. AI applications at this year’s show include manufacturing, transportation, energy, medicine, education, photography, communications, farming, grocery shopping, fitness, sports, defense, and entertainment, just to name a few. The AI revolution is just starting, but once it gets going, AI will continually reshape society for decades to come. This year’s show reinforces our decision to explore the roles that the XPRTs, beginning with AIXPRT, can play in the AI revolution.

Now for the fun stuff. Here’s a peek at a couple of my favorite displays so far. As is often the case, the most awe-inducing displays at CES are those that overwhelm attendees with light and sound. LG’s enormous curved OLED wall, dubbed the Massive Curve of Nature, was truly something to behold.

IMG_0268 - Copy

Another big draw has been Bell’s Nexus prototype, a hybrid-electric VTOL (vertical takeoff and landing) air taxi. Some journalists can’t resist calling it a flying car, but I refuse to do so, because it has nothing in common with cars apart from the fact that people sit in it and use it to travel from place to place. As Elon Musk once said of an earlier, but similar, concept, “it’s just a helicopter in helicopter’s clothing.” Semantics aside, it’s intriguing to imagine urban environments full of nimble aircraft that are quieter, easier to fly, and more energy efficient than traditional helicopters, especially if they’re paired with autonomous driving technologies.

Version 2

Finally, quite a few companies are displaying props that put some of the “reality” back into “virtual reality.” Driving and flight simulators with full range of motion that are small enough to fit in someone’s basement or game room, full-body VR suits that control your temperature and deliver electrical stimulus based on game play (yikes!), and portable roller-coaster-like VR rides were just a few of the attractions.

IMG_0203 - Copy

It’s been a fascinating show so far!

Justin

New XPRTs for the new year

Happy 2019! January is already a busy time for the XPRTs, so we want to share a quick preview of what community members can expect in the coming months.

The MobileXPRT 3 community preview (CP) is still open, but draws to a close on January 18th. If you are not familiar with the updates and changes we implemented in the newest version of MobileXPRT, you can read more in the blog. Members can access this APK on the MobileXPRT tab in the Members’ Area. We also posted an installation guide that provides both a general overview of the app and detailed instructions for each step. The entire process takes about five minutes on most devices. If you haven’t already, give it a try!

We also recently published the first AIXPRT Request for Comments (RFC) preview build, an early version of one of the tools we’re developing to evaluate machine learning performance. You can find more details in Bill’s most recent blog post and on AIXPRT.com. Only BenchmarkXPRT Development Community members have access to our RFCs and the opportunity to provide feedback. However, because we’re seeking broad input from experts in this field, we’ll gladly make anyone interested in participating a member. To gain access to the AIXPRT repository, please send us a request.

Work on the HDXPRT 4 CP candidate build continues, and we hope to publish the preview for community members this month. We appreciate everyone’s patience as we work to get this right. We think it will be worth the wait.

On a general note, I’ll be travelling to CES 2019 in Las Vegas next week. CES is a great opportunity for us to survey emerging tech and industry trends, and I look forward to sharing my thoughts from the show. If you’ll be there and would like to discuss any aspect of the XPRTs in person, let me know.

Justin

The AIXPRT Request for Comments preview build

In the next few days, we’ll be publishing the first AIXPRT tool as a Request for Comments (RFC) preview build, an early version of one of the AIXPRT tools we’re developing to help evaluate machine learning performance.

We’re inviting folks to run the workload and send in their thoughts and suggestions. Only BenchmarkXPRT Development Community members have access to our RFCs and the opportunity to provide feedback. However, because we’re seeking broad input from experts in this field, we’ll gladly make anyone interested in participating a member.

This AIXPRT RFC preview build includes support for the Intel OpenVINO computer vision toolkit to run image classification workloads with ResNet-50 and SSD-MobileNet v1 networks. The test reports FP32 and FP16 levels of precision. The system requirements are:

  • Operating system = Ubuntu 16.04
  • CPU = 6th to 8th generation Intel Core or Xeon processors, or Intel Pentium processors N4200/5, N3350/5, N3450/5 with Intel HD Graphics


We welcome input on all aspects of the benchmark, including scope, workloads, metrics and scores, user experience, and reporting. We will add support for TensorFlow and TensorRT to the AIXPRT RFC preview build during the preview period. We are accepting feedback through January 25th, 2019, after which we’ll collect and evaluate responses before publishing the next build. Because this is an RFC release, we ask that testers do not publish scores or use the results for comparison purposes.

We’ll send out a community announcement when the RFC preview build is officially available, and we’ll also post an announcement and RFC preview build user guide on AIXPRT.com. We’re hosting the AIXPRT RFC preview build in a dedicated GitHub repository, so please contact us at BenchmarkXPRTsupport@principledtechnologies.com to gain access.

This is just the next step for AIXPRT. With your help, we hope to add more workloads and other frameworks in the coming months. We look forward to receiving your feedback!

Bill

XPRT collaborations: North Carolina State University

For those of us who work on the BenchmarkXPRT tools, a core goal is involving new contributors and interested parties in the benchmark development process. Adding voices to the discussion fosters the collaboration and innovation that lead to powerful benchmark tools with lasting relevance.

One vehicle for outreach that we especially enjoy is sponsoring a student project through North Carolina State University. Each semester, the Senior Design Center in the university’s Department of Computer Science partners with external companies and organizations to provide student teams with an opportunity to work on real-world programming projects. If you’ve followed the XPRTs for a while, you may remember previous student projects such as Nebula Wolf, a mini-game that shows how well different devices handle games, and VR Demo, a virtual reality prototype workload based on a room escape scenario.

This fall, a team of NC State students is developing a software console for automating machine learning tests. Ideally, the tool will let future testers specify custom workload combinations, compute a performance metric, and upload results to our database. The project will also assess the impact of the framework on performance scores. In fact, the console will perform many of the same functions we plan to implement with AIXPRT.

The students have worked very hard on the project, and have learned quite a bit about benchmarking practices and several new software tools. The project will wrap up in the next couple of weeks, and we’ll share additional details as soon as possible. Early next year, we’ll publish a video about the experience.

If you’d like to join the NC State students and hundreds of other XPRT community members in the future of benchmark development, please let us know!

Justin

Notes from the lab: Updates on HDXPRT 4, MobileXPRT 3, and AIXPRT

The next couple of months will be very busy with XPRT activity, so we want to update readers on what to expect. Depending on a number of factors, we expect to release HDXPRT 4 and MobileXPRT 3 community previews (CPs) within the next four to six weeks. We’re also hoping to publish an early AIXPRT request-for-comment (RFC) build on GitHub within the same time frame. Here’s a little more detail about each of these developments.

HDXPRT 4: We originally planned to release the HDXPRT 4 CP several weeks ago. As we recently discussed in the blog, a lot has changed in the Windows 10 development world within a short period of time, and Microsoft has released a number of new Redstone 5/October 2018 Update builds in quick succession. While our HDXPRT 4 CP candidate testing went well overall, we observed some inconsistent workload scores when testing on some of the new Windows builds. Since then, we believe we’ve narrowed down the list of possible causes to a few specific graphics driver versions, but we’re still testing to make sure there are no other immediate issues. As soon as we’re confident in that assessment, we’ll release the CP along with any relevant information about the affected graphics drivers.

MobileXPRT 3: MobileXPRT 3 development is progressing nicely, and we’re close to completing a CP candidate build. We’ll test that build extensively on our library of Android phones and tablets, and barring any unforeseen issues, we plan to release the CP in the next few weeks.

AIXPRT: AIXPRT is the umbrella name for a set of tools we’re developing to help evaluate machine learning performance. After a great deal of research, we’re getting closer to releasing a build – tentatively called the AIXPRT RFC – for community members and other interested parties to download and review. For a number of reasons, the AIXPRT RFC process will be a little different than our normal XPRT RFC and CP process. We’ll be offering more information on the AIXPRT RFC build over the next several weeks.

We’re grateful to everyone who’s contributed in any way to each of these projects, and we look forward to sharing the benchmarks with the world. If you have any questions about the XPRTs, please don’t hesitate to ask!

Justin

News from the MobileXPRT 3 team

A few months ago, we shared some of our thoughts during the early planning stages of MobileXPRT 3 development. Since then, we’ve started building the new benchmark with Android Studio SDK 27. We’re now at a place where we can share more details about what to expect in MobileXPRT 3. In a nutshell, one of the five workloads in the previous version, MobileXPRT 2015, is getting a major overhaul, the remaining four workloads are getting updated test content, and we’re adding one completely new workload.

One of the first challenges we tackled was to completely rebuild the Create Slideshow workload. In MobileXPRT 2015, the workload uses FFmpeg to convert photos into video. FFmpeg utilizes a C++ executable, and it needs to be compiled differently for different architectures such as x86, x64, arm32, arm64, etc. With each new Android version, the task of maintaining FFmpeg compatibility with numerous architectures and Android versions becomes more complex. MobileXPRT 2015 still works well on most Android devices, but we wanted a more future-proof solution. In MobileXPRT 3, the Create Slideshow workload will use the Android MediaCodec API instead of FFmpeg. This change enables the workload to run successfully on devices that could not complete the workload in MobileXPRT 2015.

We are updating the test content of the following workloads: Apply Photo Effects, Create Photo Collages, Encrypt Personal Content, and Detect Faces to Organize Photos. We will replace items such as photos and videos with more contemporary file resolutions and sizes where applicable.

In the mobile device market, artificial intelligence and machine learning capabilities are rapidly moving from the level of novelty to being integrated into many daily tasks, so we wanted to include an AI or ML element in MobileXPRT 3. Our new workload uses Google’s Mobile Vision API to perform optical character recognition (OCR) tasks involving scanning receipts for personal records or an expense report. The scenario is similar to the OCR receipt-scanning task in WebXPRT 3, though the two workloads are based on different text-recognition technologies.

Finally, we’re updating the MobileXPRT UI to improve the look of the benchmark and make it easier to use. We’ll share a sneak peek of the new UI here in the blog around the time of the community preview. If you have any questions about MobileXPRT 2015 or MobileXPRT 3, please let us know!

Justin

Check out the other XPRTs:

Forgot your password?