A Principled Technologies report: Hands-on testing. Real-world results.

HP ZBook Power 16-inch G11 Mobile Workstation PC: Accelerate growth and performance

vs. an HP ZBook Power 15.6-inch G10 Mobile Workstation PC

The HP ZBook Power 16-inch G11 mobile workstation PC is powered by an Intel® Core Ultra processor, which integrates central processing unit (CPU), graphic processing unit (GPU), and neural processing unit (NPU) architecture into a single chip.1 What benefits does this cutting-edge technology bring to the table?

To find out, we used industry-standard benchmarks to compare AI, 3D rendering, and content creation workload performance on a next-gen HP ZBook Power 16-inch G11 Mobile Workstation PC powered by an Intel® vPro® with Intel Core Ultra 7 processor 165H to that of its 15.6-inch G10 predecessor powered by an Intel Core i7-13800H processor. We found the new Intel Core Ultra processor-powered HP ZBook Power 16-inch G11 Mobile Workstation PC delivers wins in every category, helping separate the champions from the also-rans.

Take a victory lap with image identification tasks. 90.5 percent higher Geekbench AI GPU Quantized score. Burn rubber while creating an image. 34.8 percent less time to generate a Stable Diffusion image from a text prompt. Render 3D projects at speed. 44.7 percent higher 3DMark Time Spy Extreme score.

How we tested

To measure the benefits of upgrading to the newest 16-inch HP ZBook Power G11 Mobile Workstation PC, we equipped both Windows 11 Pro systems with 32 GB of memory, 512 GB of storage, and the closest discrete NVIDIA® graphics available at the time of testing:

HP ZBook Power 16-inch G11

Mobile Workstation PC

  • Intel vPro with Intel Core Ultra 7 processor 165H
  • Integrated Intel Arc™ graphics
  • Integrated Intel AI Boost
  • Discrete NVIDIA RTX™ 500 Ada graphics
  • 83-Whr battery

HP ZBook Power 15.6-inch G10

Mobile Workstation PC

  • Intel vPro with Intel Core i7-13800H processor
  • Integrated Intel Iris® Xe graphics
  • No NPU component
  • Discrete NVIDIA RTX A500 graphics
  • 83-Whr battery

To evaluate system performance from many angles, we used these industry-standard benchmarks and AI tools:

  • 3DMark® Fire Strike Extreme
  • 3DMark Time Spy Extreme
  • Cinebench 2024
  • Geekbench AI
  • ResNet-50
  • SPECworkstation® 3.1
  • Stable Diffusion

The results we report reflect the specific configurations we tested. Any difference in the configurations—as well as screen brightness, network traffic, and software additions—can affect these results. For a deeper dive into our testing parameters and procedures, see the science behind the report.

Boost productivity

Whether your day is filled with problem solving, 3D modeling and rendering, or using AI or machine learning (ML) applications, system performance is key to your success.

The SPECworkstation 3.1 benchmark measures CPU, graphics, I/O, and memory bandwidth for diverse professional applications.5 Each SPECworkstation 3.1 workload uses different tools to measure performance of industry-specific workflows. The Product Development workload uses Rodinia, WPCcfd, CalcuilX, CATIA, Creo, NX, Solidworks, and Showcase tools. The Financial Services workload uses a Monte Carlo simulation algorithm as well as Black-Scholes and Binomial options pricing model algorithms.6

Chart of SPECworkstation 3.1 benchmark results. Higher is better. Put the pedal to the metal for everyday work. Product Development. ZBook Power G11 has a 3.47 overall score and ZBook Power G10 has a 2.68 overall score. 29.4 percent higher. Financial Services. ZBook Power G11 has a 3.28 overall score and ZBook Power G10 has a 2.93 overall score. 11.9 percent higher. Graphics. ZBook Power G11 has a 3.58 overall score and ZBook Power G10 has a 3.20 overall score. 11.8 percent higher. GPU Compute. ZBook Power G11 has a 3.25 overall score and ZBook Power G10 has a 2.53 overall score. 28.4 percent higher.
SPECworkstation 3.1 overall scores. Higher is better. Source: Principled Technologies.

Speed content creation and graphics-intensive efforts

The level of system performance that business, creative, and technical professionals require depends on their specific workflows.

The 3DMark Fire Strike Extreme benchmark is a DirectX 11 test that stresses both CPU and GPU.7 The 3DMark Time Spy Extreme benchmark is a DirectX 12 test that stresses graphics cards and multi-core processors.8 Better real-time graphics performance can translate to faster 3D graphic rendering and smoother viewing experiences.

Chart of 3DMark benchmark results. Higher is better. Add horsepower for graphics-intensive workloads. Fire Strike Extreme. ZBook Power G11 has a 6,651 score and ZBook Power G10 has a 4,833 score. 37.6 percent higher. Time Spy Extreme. ZBook Power G11 has a 2,851 score and ZBook Power G10 has a 1,970 score. 44.7 percent higher.
3DMark results. Higher is better. Source: Principled Technologies.

The Cinebench 2024 benchmark uses the Redshift rendering engine to evaluate CPU and GPU capabilities while running a Cinema 4D scene.9

Chart of Cinebench 2024 CPU results. Higher is better. Pull ahead in content creation projects. ZBook Power G11 has a 928 score and ZBook Power G10 has a 905 score. 2.5 percent higher.
Cinebench 2024 CPU multi-core scores. Higher is better. Source: Principled Technologies.

Set the pace with AI and ML technologies

AI and ML technologies are changing how business, creative, and technical professionals handle challenges. But this advancement can come at a price. AI and ML workloads are data- and processor-intensive activities. Making sure you have a powerful processor under the hood can help you cross the finish line faster.

The Geekbench AI benchmark measures CPU, GPU, and NPU capabilities for machine learning applications.10 The Single Precision score reports float32 precision, the Half Precision score reports float16 precision, and the Quantized score reports int8 precision.11 In our testing, we used the Open Neural Network Exchange (ONNX) AI framework and DirectML AI backend for machine learning on Windows.

Chart of Geekbench AI results. Higher is better. Wait less for valuable insights. Single Precision score. ZBook Power G11 has a 11,175 score and ZBook Power G10 has a 6,779 score. 64.8 percent higher. Half Precision score. ZBook Power G11 has a 19,718 score and ZBook Power G10 has a 13,343 score. 47.7 percent higher. Quantized score. ZBook Power G11 has a 8,611 score and ZBook Power G10 has a 4,520 score. 90.5 percent higher.
Geekbench AI ONNX DirectML GPU results. Higher is better. Source: Principled Technologies.

The MLPerf ResNet-50 benchmark measures how many samples per second a system can process using the ResNet model.12 ResNet models can help users organize and categorize objects, people, and places in images.

Stable Diffusion is a generative AI model that produces photorealistic images from text prompts.13 For our tests, we measured how long it took for each mobile workstation to generate an image based on identical Stable Diffusion text prompts.

Chart of ResNet-50 results. Higher is better.  Take a victory lap with image classification tasks. ZBook Power G11 has 4,601.94 samples per second and ZBook Power G10 has 3,481.82 samples per second. 32.1 percent more.
MLPerf ResNet-50 benchmark results. Higher is better. Source: Principled Technologies.
Chart of Stable Diffusion results. Lower is better.  Burn rubber while generating ideas. ZBook Power G11 generated an image in 3:28 seconds and ZBook Power G10 generated an image in 5:19 seconds. 34.8 percent less time.
Stable Diffusion text-prompt to image-generation results. Time in minutes and seconds. Less time is better. Source: Principled Technologies.

Conclusion

With the right team in your corner and the right tools at your fingertips, you can win the productivity race and cross the finish line faster. In head-to-head AI, 3D rendering, and content creation performance comparisons, an HP ZBook Power 16-inch G11 Mobile Workstation PC powered by an Intel vPro with Intel Core Ultra 7 processor 165H raced circles around its 15.6-inch predecessor. We found upgrading to the newest Intel Core Ultra processor-powered ZBook Power can help you and your team take a victory lap with image classification tasks, burn rubber while creating an image from a text prompt, and speed 3D graphics rendering tasks.

  1. Intel, “Intel® Core Ultra Processors Product Brief,” accessed November 4, 2024, https://www.intel.com/content/www/us/en/products/docs/embedded-processors/core-ultra/product-brief.html.
  2. Intel, “Intel Core Ultra processors,” accessed November 13, 2024, https://www.intel.com/content/www/us/en/newsroom/resources/core-ultra-1.html#gs.i3yo1i.
  3. HP, “HP ZBook Power 16 G11 Mobile Workstation PC,” accessed November 13, 2024, https://www.hp.com/us-en/shop/pdp/hp-zbook-power-g11-mobile-workstation-pc-customizable-9a670av-mb#techSpecs.
  4. Intel, “Intel® Core Ultra 7 Processor 165H,” accessed November 5, 2024, https://www.intel.com/content/www/us/en/products/sku/236851/intel-core-ultra-7-processor-165h-24m-cache-up-to-5-00-ghz/specifications.html.
  5. SPEC GWPG, “SPECworkstation® 3.1” accessed November 5, 2024, https://gwpg.spec.org/benchmarks/benchmark/specworkstation-3_1/.
  6. SPEC GWPG, “SPECworkstation® 3.1.”
  7. UL Procyon, “Overview of 3DMark Fire Strike Extreme benchmark,” accessed November 5, 2024, https://support.benchmarks.ul.com/support/solutions/articles/44002136196-overview-of-3dmark-fire-strike-extreme-benchmark.
  8. UL Procyon, “Overview of 3DMark Time Spy Extreme benchmark,” accessed November 5, 2024, https://support.benchmarks.ul.com/support/solutions/articles/44002136104-overview-of-3dmark-time-spy-extreme-benchmark.
  9. Maxon, “Cinebench,” accessed November 5, 2024, https://www.maxon.net/en/cinebench.
  10. Geekbench AI, “Introducing Geekbench AI,” accessed November 5, 2024, https://www.geekbench.com/ai/.
  11. Geekbench, “Geekbench AI workloads,” accessed November 5, 2024, https://www.geekbench.com/doc/geekbench-ai-workloads.pdf.
  12. ML Commons, “MLPerf Inference: Datacenter Benchmark Suite Results,” accessed November 5, 2024, https://mlcommons.org/benchmarks/inference-datacenter/.
  13. AWS, “What is Stable Diffusion?” accessed November 5, 2024, https://aws.amazon.com/what-is/stable-diffusion/.

This project was commissioned by HP.

December 2024

Principled Technologies is a registered trademark of Principled Technologies, Inc.

All other product names are the trademarks of their respective owners.

Forgot your password?