The Dell Pro Precision 5 16s Intel is a 16-inch mobile workstation built for users who want more screen space while keeping the system reasonably portable. Our review unit pairs the Core Ultra X9 388H with Intel Arc Pro B390 graphics, 64GB of LPCAMM2-8533 memory, and a 1TB Gen5 SED SSD. The 2560 x 1600 IPS display gives plenty of room for larger spreadsheets, development tools, and creative applications, while the added numeric keypad makes better use of the wider chassis. It also manages to pack all of this in while delivering excellent battery life.
The 16s is built for engineers, developers, analysts, creators, and other professionals who need workstation-level performance in a system that is still fairly easy to carry around. Intel vPro Enterprise, Wi-Fi 7, a fingerprint reader, smart card support, TPM 2.0, and Dell’s management tools also make it a good fit for managed business environments. Our 64GB LPCAMM2 configuration has plenty of memory for heavier multitasking, professional applications, and local AI workloads. Dell also offers the same chassis with an AMD Ryzen AI 9 HX PRO 475, so potential buyers have another platform option with a very similar overall design.
The Arc Pro B390 gives the 16s considerably more graphics capability than most business laptops with integrated graphics, along with 12 Xe cores and certified drivers for professional applications. The Core Ultra X9 388H also includes a 50 TOPS NPU, giving the system dedicated hardware for supported local AI workloads without relying entirely on the CPU or GPU. The larger 16-inch chassis gives the cooling system more room to work as well, and, as you will see in our testing below, the 16s posted the strongest multicore results we have seen from a B390-based system.
Our review configuration is priced at $5,929.25 as a single-unit purchase on Dell.com, with the jump from 16GB to 64GB of LPCAMM2 memory accounting for $1,600 of that total. Commercial pricing can vary considerably depending on configuration, support agreements, volume, and account-level discounts, so the listed web price is better treated as a reference point than a typical fleet purchase price. The system is available now through the Dell Pro Precision 5 Series 16S product page.
Dell Pro Precision 5 16s Specifications
| Specification | Dell Pro Precision 5 16s (PW516260) |
| Processor | Intel Core Ultra X9 388H vPro (Series 3, 16 cores/16 threads, up to 5.1GHz, 50 TOPS NPU) |
| Graphics | Intel Arc Pro B390 (12 Xe cores, integrated) |
| Memory | 64GB LPCAMM2, 8533 MT/s, dual-channel |
| Storage | 1TB SED PCIe NVMe 2280 SSD, Gen5 x4 |
| Display | 16-inch QHD+/WQXGA (2560 x 1600), non-touch, 500 nits, IPS, 100% sRGB, anti-glare, low blue light |
| Camera | 8MP HDR RGB + IR with User Presence Detection |
| Wireless | Intel Wi-Fi 7 BE211, 2×2 |
| Keyboard | English US backlit with numeric keypad and Copilot key |
| Security | Fingerprint reader, smart card reader, ControlVault 3+, TPM 2.0 |
| Battery | 3-cell, 70Wh Long Lifecycle, ExpressCharge and ExpressCharge Boost |
| Power | 100W USB-C adapter |
| Operating System | Windows 11 Pro (Copilot+ PC) |
| Chassis | Aluminum top cover, palm rest, and bottom cover |
| Systems Management | Intel vPro Enabled |
| Certifications | ENERGY STAR, EPEAT Gold with Climate+, TCO Certified |
| Warranty | 36 months Basic Onsite Service after Remote Diagnosis |
| Price | $5,929.25 as tested (Dell.com single-unit) |
Build and Design
The Dell Pro Precision 5 16s uses aluminum across the top cover, palm rest, and bottom cover, giving the 16-inch chassis a solid feel while keeping the design fairly slim for a mobile workstation. It measures 14.12 x 9.98 inches, reaches 0.79 inches at its thickest point, and starts at 4.20 lb. Compared with the 14-inch Pro Precision 5 14s, which starts at 3.12 lb, the added size and roughly 1.1 lb of extra weight are easy to notice in a bag, but the larger chassis gives users considerably more screen and keyboard space. The AMD version of the 16s uses the same chassis, so the exterior design, dimensions, keyboard, port layout, and service access are essentially identical between the two platforms.
The 16-inch display is a 2560 x 1600 IPS panel with a 16:10 aspect ratio, 500-nit brightness, 100% sRGB coverage, an anti-glare finish, and low-blue-light support. That combination works nicely for a workstation this size, especially when working with larger spreadsheets, development environments, timelines, or applications with several tool panels open at once. The extra vertical room from the 16:10 panel is particularly useful beside a conventional 16:9 display, and the 500-nit rating gives the screen enough brightness for offices with stronger overhead lighting. Our configuration is non-touch.
Dell uses a wide hinge across the rear of the chassis, with most of the mechanism tucked behind the display rather than occupying space along the keyboard deck. The lid itself is fairly thin, which helps keep the 16s from looking bulky when viewed from the side, and the aluminum construction gives the display assembly more rigidity than a typical plastic business notebook. Compared with the 14s, the overall design is very similar, but the wider base gives the keyboard, touchpad, and display more room without dramatically increasing chassis thickness.
For audio, it has two internal 2W speakers for 4W of total output, with the speaker modules visible at the lower corners when the bottom cover is removed. Their placement keeps speaker grilles away from the keyboard deck and directs audio through openings along the lower chassis. The setup is primarily geared toward conferencing, calls, and everyday media use, with keyboard controls available for quick volume adjustment. Dell also pairs the speakers with dual digital-array microphones for conferencing.
The larger keyboard deck is put to good use with a full numeric keypad, which is a useful addition for spreadsheet work, financial applications, engineering software, and other number-heavy workloads. Our unit has a backlit keyboard with a dedicated Copilot key, while the power button at the upper-right corner also houses the fingerprint reader. Dell shifts the large clickpad slightly toward the left so it lines up more closely with the main typing area instead of centering it across the entire width of the notebook. There is still plenty of palm-rest space around it, even with the wider keyboard and number pad.
Port selection is the strongest practical argument for working from this machine without a dock. Dell fits nine ports and slots across the two sides, which is a lot for a chassis that measures 0.79 inches at its thickest point, and almost none of it is legacy that has been quietly dropped on competitive systems. The left side carries HDMI 2.1, one USB-A port, two Thunderbolt 4 USB-C ports with DisplayPort and Power Delivery, and the smart card reader.
The right side adds the 3.5mm headset jack, a second USB-A port, Gigabit Ethernet on a drop-jaw hinge that lets a full-size RJ45 fit a chassis this thin, and a wedge lock slot. The small blank ahead of the headset jack is an eSIM cover rather than a card slot.
The effect is a full desk setup that needs nothing extra. Displays can hang off HDMI 2.1 or either Thunderbolt port; wired networking does not need a dongle or a USB adapter; and the two USB-A ports are split, one per side, for the peripherals that never moved to USB-C. The smart card reader is one that most competitors have given up, and it remains a requirement in government, healthcare, and finance deployments where CAC or PIV login is mandatory. The only layout quibble is that both USB-C ports sit on the left, so charging is left-side-only.
Above the display, Dell includes an 8MP HDR RGB and IR camera with dual-array microphones, user presence detection, and a physical privacy shutter. The 8MP sensor provides considerably more image resolution than the basic 1080p webcams still common in business notebooks, which can help retain facial detail during calls and conferencing. IR support provides Windows Hello facial authentication, while presence detection can work with supported Windows features to wake or lock the notebook as the user approaches or leaves. The physical shutter is built directly into the camera housing and can be closed with a small slider above the display. Dell lists several camera configurations for the 16s, with the 8MP HDR RGB+IR setup used in our review unit sitting at the top of that range.
In addition to the fingerprint reader and IR camera, security hardware with our configuration also includes a smart card reader, TPM 2.0, and Dell ControlVault 3+. There is no NFC reader on this configuration, which is one difference from the 14s unit we tested. Intel vPro Enterprise adds another layer for organizations using remote administration and hardware-assisted security features, which gives IT departments several authentication and management options without requiring external hardware.
Opening the bottom cover shows several major components accessible instead of fixing everything permanently to the motherboard. The 70Wh long-lifecycle battery in our configuration stretches across much of the lower half of the chassis and is customer-replaceable, while the LPCAMM2 memory module, M.2 SSD, WLAN card, fan, and several other components can also be replaced. Below, you can also see a single large cooling fan and heat pipe covering the processor area, with the LPCAMM2 module and storage positioned nearby for direct access once the cover is removed. Dell officially classifies the battery, memory, SSD, WLAN card, fan, speakers, and fingerprint reader as customer-replaceable components, which gives IT departments a lot more flexibility when it comes to repairs and upgrades over the life of the notebook.
Dell Pro Precision 5 16s Performance
Our review unit runs the Core Ultra X9 388H with Arc Pro B390 graphics, 64GB of LPCAMM2 at 8533 MT/s, and a 1TB Gen5 SED SSD on Windows 11 Pro, tested on the Balanced power profile.
For comparables, we included the AMD version of the same machine, the Dell Pro Precision 5 16s AMD (Ryzen AI 9 HX PRO 475, Radeon 890M, 64GB), the step-up Dell Pro Precision 7 16 (Core Ultra 9 386H, NVIDIA RTX PRO 3000 Blackwell 12GB, 64GB, dual-drive RAID 0), and the Lenovo ThinkPad P14s Gen 7 (Core Ultra 7 366H, RTX PRO 1000, 64GB) as the external workstation reference. A few comp cells are marked N/A where a system could not complete a workload. The 7 16 Geekbench 6 run carried an invalid flag from tamper detection, which we reviewed and treated as a false positive, and its Cinebench CPU results were confirmed by a repeat run. We also tested the Dell Pro Precision 5 14s Intel, which shares this unit’s CPU and GPU; its results land within a few percent across the suite, so we left it off the tables and call out the places where the chassis difference shows up.
PCMark 10
PCMark 10 measures general system performance across everyday work such as web browsing, video conferencing, spreadsheets, writing, photo editing, and rendering. The overall score is supported by Essentials, Productivity, and Digital Content Creation subscores that show where a system’s strengths sit. Higher scores are better.
| PCMark 10 | Dell Pro Precision 5 16s Intel | Dell Pro Precision 5 16s AMD | Dell Pro Precision 7 16 | Lenovo ThinkPad P14s Gen 7 |
| Overall Score | 9,994 | 8,627 | 8,901 | 9,083 |
| Essentials | 12,106 | 10,744 | 9,671 | 10,686 |
| Productivity | 15,505 | 14,574 | 16,381 | 16,501 |
| Digital Content Creation | 14,433 | 11,127 | 12,081 | 11,534 |
At 9,994 overall, the 5 16s Intel leads this group by a comfortable margin, 12% over the far more expensive 7 16 and 16% over its AMD twin, though it stops just six points short of the 10,000 mark its 14-inch sibling crossed. Its Digital Content Creation score of 14,433 is the best of the four, while the 7 16 and ThinkPad claw back ground in the Productivity subscore.
PCMark 10 Modern Office Battery
The PCMark 10 Modern Office battery test repeatedly runs common office tasks until the battery reaches the test’s cutoff point, in Balanced mode at 50% display brightness. This is a rundown of the whole system rather than a synthetic idle drain, so it tracks closely with what a full day of productivity work does to the battery. Longer runtimes are better.
| Modern Office Battery | Dell Pro Precision 5 16s Intel | Dell Pro Precision 5 16s AMD | Dell Pro Precision 7 16 | Lenovo ThinkPad P14s Gen 7 |
| Runtime (higher is better) | 24 hours 43 minutes | 15 hours 5 minutes | 11 hours 43 minutes | 15 hours 52 minutes |
This is the longest battery run of the entire Pro Precision fleet. At 24 hours and 43 minutes, the 5 16s Intel more than doubles the dGPU-equipped 7 16, adds nine hours and 38 minutes over the AMD version of the same chassis, runs nearly nine hours past the ThinkPad, and even outlasts its own 14-inch twin by 53 minutes despite the larger, higher-resolution panel, a credit to the same 70Wh pack paired with the 16-inch chassis thermals. For all-day-and-then-some field work, this is the unit in the family to pick.
Geekbench 6
Geekbench 6 measures processor performance using a mix of common tasks, with separate scores for single-core and multi-core workloads, plus GPU compute scores through OpenCL and Vulkan. Higher scores are better. The 7 16 CPU run was flagged as invalid by the benchmark’s tamper detection; we reviewed it and treated the flag as a false positive, so its scores are included below.
| Geekbench 6 | Dell Pro Precision 5 16s Intel (Arc Pro B390) | Dell Pro Precision 5 16s AMD (Radeon 890M) | Dell Pro Precision 7 16 (RTX PRO 3000) | Lenovo ThinkPad P14s Gen 7 (RTX PRO 1000) |
| CPU Single-Core | 2,969 | 2,854 | 2,915 | 2,808 |
| CPU Multi-Core | 17,401 | 14,169 | 17,280 | 16,319 |
| GPU OpenCL | 57,998 | 37,520 | 131,492 | 87,537 |
| GPU Vulkan | 63,015 | 53,969 | 104,344 | 73,204 |
The X9 388H leads the CPU runs on both metrics, edging the Core Ultra 9 386H in the 7 16 by 2% single-core and under 1% multi-core, and the Arc Pro B390’s 63,015 Vulkan score is the best we have recorded from this iGPU, edging past even its own OpenCL result. The RTX PRO 3000 resets the scale on GPU compute, more than doubling the review unit in OpenCL.
Geekbench 7
Geekbench 7 joins the suite alongside Geekbench 6 as comparison data builds. Its CPU scores are calibrated against a baseline of 2,500, set by the AMD Ryzen 7700, while GPU scores are calibrated against a baseline of 100,000, set by the NVIDIA GeForce RTX 4060. Higher scores are better, and double the score indicates double the performance. Because Geekbench 7 uses new workloads and new baselines, its scores are not comparable to Geekbench 6 results.
| Geekbench 7 | Dell Pro Precision 5 16s Intel (Arc Pro B390) | Dell Pro Precision 5 16s AMD (Radeon 890M) | Dell Pro Precision 7 16 (RTX PRO 3000) | Lenovo ThinkPad P14s Gen 7 (RTX PRO 1000) |
| CPU Single-Core | 2,733 | 2,659 | 2,652 | 2,533 |
| CPU Multi-Core | 18,965 | 16,641 | 18,648 | 17,651 |
| GPU OpenCL | 54,247 | 31,874 | 96,919 | 75,340 |
| GPU Vulkan | 47,436 | * | 104,229 | 70,067 |
| GPU CUDA | N/A | N/A | 174,559 | 114,469 |
*The AMD unit’s Vulkan run failed workload validation in Geekbench 7 and was assigned a zero score for the affected subtest; we have discarded the result and will update the table when a clean run is available. Separately, on the dual-GPU 7 16, Geekbench’s Vulkan device selector runs the workload on the opposite GPU from the one selected; the scores above are attributed to the device that did the work.
Here is the quiet upset of the review: at 18,965 multi-core, the 5 16s Intel beats every system in the group, including the Core Ultra 9-equipped 7 16, and its 2,733 single-core leads as well. On the GPU side, the Arc Pro B390’s OpenCL score lands at just over half of Geekbench 7’s RTX 4060 baseline, while the RTX PRO 3000 clears the baseline in Vulkan and beats it by three-quarters in CUDA.
Cinebench 2026
Cinebench 2026 is the current release in the Cinebench line and the only version we report. It tests CPU and GPU performance using Maxon’s Redshift render engine. It is built on the latest Cinema 4D 2026 code and is designed to show whether a machine is stable under high CPU load, whether a notebook’s cooling can sustain longer render tasks, and how it handles demanding real-world 3D work. Because code and compiler changes accelerated scene rendering, Cinebench 2026 scores use an adjusted range and should not be compared to scores from previous Cinebench versions. Its GPU test supports current NVIDIA and AMD hardware but does not yet run on Intel integrated graphics, and the 7 16’s CPU results were confirmed by a repeat run.
| Cinebench 2026 | Dell Pro Precision 5 16s Intel | Dell Pro Precision 5 16s AMD | Dell Pro Precision 7 16 | Lenovo ThinkPad P14s Gen 7 |
| CPU Single Thread | 535 | 471 | 518 | 502 |
| CPU Multiple Threads | 4,618 | 4,767 | 3,873 | 4,492 |
| GPU | N/A | 5,667 | 51,517 | 34,437 |
The review unit takes a single thread at 535, the best Cinebench 2026 single-thread result we have recorded from a laptop, while the AMD twin’s 24 threads edge it in the multi-thread test by 3%. On GPU, the RTX PRO 3000’s 51,517 is the number that justifies the 7 16’s existence for supported renderers, half as quick as the RTX PRO 1000 in the ThinkPad.
7-Zip Compression
The built-in 7-Zip benchmark measures how quickly the processor can compress and decompress data using multiple threads, run with a 128MB dictionary across ten passes. Decompression tends to scale with thread count while compression leans on memory latency, so the two halves often tell different stories. Higher GIPS scores are better.
| 7-Zip 24.09 (GIPS) | Dell Pro Precision 5 16s Intel | Dell Pro Precision 5 16s AMD | Dell Pro Precision 7 16 | Lenovo ThinkPad P14s Gen 7 |
| Compressing | 96.281 | 85.627 | 96.202 | 90.364 |
| Decompressing | 97.246 | 117.863 | 95.896 | 90.526 |
| Total Rating | 96.764 | 101.745 | 96.049 | 90.445 |
The AMD unit’s 24-thread decompression hands it the total rating. Still, the review unit wins compression outright, and its 96.764 total edges the 7 16 while beating its own 14-inch twin by 4%, one of the clearest examples of what the bigger chassis buys from identical silicon.
y-cruncher
y-cruncher measures how quickly the processor can calculate large numbers of digits of Pi, placing a heavy load on the CPU and memory subsystem. At the same time, the BBP runs to extract hexadecimal digits of Pi. Results are in seconds, so lower times are better. The AMD unit could not complete the 5-billion- and 10-billion-digit runs because its memory reservation for the integrated GPU reduces the available pool below what those sizes require.
| y-cruncher (seconds, lower is better) | Dell Pro Precision 5 16s Intel | Dell Pro Precision 5 16s AMD | Dell Pro Precision 7 16 | Lenovo ThinkPad P14s Gen 7 |
| Pi 1B | 26.422 | 22.840 | 31.557 | 26.685 |
| Pi 2.5B | 80.463 | 64.242 | 101.524 | 76.787 |
| Pi 5B | 183.822 | N/A | 226.255 | 172.994 |
| Pi 10B | 407.658 | N/A | 500.875 | 392.083 |
| Pi BBP 1B | 1.635 | 1.100 | 1.639 | 1.621 |
| Pi BBP 10B | 19.036 | 12.277 | 21.512 | 18.200 |
| Pi BBP 100B | 234.797 | 140.284 | 291.628 | 219.969 |
The AMD unit’s SMT threads dominate every run it completed, and the ThinkPad stays slightly ahead of the review unit at the largest Pi sizes. The surprise is the 7 16 trailing the whole group at every size, 23% behind the review unit at 10 billion digits, a result consistent with its lower memory throughput in the SPEC runs and worth revisiting alongside its Cinebench reruns.
Blender
The Blender benchmark measures rendering performance using three different 3D scenes: Monster, Junkshop, and Classroom. Results are reported in samples per minute, so higher scores are better, and we test on both the CPU and GPU. Scores are not comparable across Blender versions, so we have trimmed the older releases from the suite and report the current Blender 5.2 results here. The GPU figures are each system’s fastest renderer, meaning the discrete card on the two NVIDIA systems and the integrated GPU on the 5 16s pair.
| Blender 5.2 (samples/min) | Dell Pro Precision 5 16s Intel (Arc Pro B390) | Dell Pro Precision 5 16s AMD (Radeon 890M) | Dell Pro Precision 7 16 (RTX PRO 3000) | Lenovo ThinkPad P14s Gen 7 (RTX PRO 1000) |
| GPU | ||||
| Monster | 572.84 | 123.00 | 1,555.53 | 922.95 |
| Junkshop | 468.85 | 101.21 | 1,257.01 | 795.65 |
| Classroom | 420.48 | 82.84 | 1,000.62 | 625.62 |
| CPU | ||||
| Monster | 138.88 | 130.86 | 137.49 | 131.08 |
| Junkshop | 94.97 | 99.74 | 101.34 | 97.76 |
| Classroom | 65.39 | 74.18 | 69.26 | 68.45 |
The RTX PRO 3000 renders Monster at 1,555.53 samples per minute, 69 percent ahead of the RTX PRO 1000 in the ThinkPad and nearly triple the Arc Pro B390, which is what the 7 16’s dGPU premium buys in a supported renderer. Among the integrated GPUs, the review unit leads the Radeon 890M by more than four times across all three scenes, and it lands at 37 percent of the RTX PRO 3000, a respectable showing for a system with no discrete graphics. CPU rendering separates the four by only a few percent, with the review unit taking Monster outright.
LuxMark
LuxMark measures GPU compute performance by rendering complex scenes through OpenCL, based on LuxCoreRender. We run the Food and Hall scenes on all available OpenCL devices in each system, so single-GPU systems are scored on that GPU while the dual-GPU systems render on the discrete and integrated GPUs together, as noted in the column headers. Higher scores are better.
| LuxMark v4 | Dell Pro Precision 5 16s Intel (Arc Pro B390) | Dell Pro Precision 5 16s AMD (Radeon 890M) | Dell Pro Precision 7 16 (RTX PRO 3000 + iGPU) | Lenovo ThinkPad P14s Gen 7 (RTX PRO 1000 + iGPU) |
| Hall | 3,505 | 2,058 | 19,300 | 11,342 |
| Food | 1,713 | 1,034 | 7,624 | 4,103 |
The review unit led the single-GPU systems by 70% in Hall, essentially matching its 14-inch twin, while the 7 16’s combined RTX PRO 3000 and iGPU output is in a different class at 19,300.
V-Ray
Chaos V-Ray measures ray-traced rendering throughput, reported in vpaths, where higher is better. We run the CUDA-compatible engine on every system so results remain comparable to notebooks with only integrated graphics; on systems without a discrete GPU, that path executes on the integrated graphics, even though V-Ray reports the processor name in its device field. On dedicated NVIDIA hardware, we also capture V-Ray’s RTX engine, which engages the card’s ray tracing cores and is reported separately.
| V-Ray GPU (vpaths) | Dell Pro Precision 5 16s Intel (Arc Pro B390) | Dell Pro Precision 5 16s AMD (Radeon 890M) | Dell Pro Precision 7 16 (RTX PRO 3000) | Lenovo ThinkPad P14s Gen 7 (RTX PRO 1000) |
| CUDA Engine | 946 | 1,039 | 2,469 | 1,568 |
| RTX Engine | N/A | N/A | 3,891 | 2,589 |
The Radeon 890M takes the integrated matchup here by 10%, one of the few GPU tests where it beats the Arc Pro B390. The 7 16’s RTX PRO 3000 is the clear headliner, and its 3,891 vpaths on the RTX engine are 50% up on the ThinkPad’s RTX PRO 1000 in the same mode.
3DMark CPU Profile
The 3DMark CPU Profile benchmark measures CPU performance at fixed thread counts, from a single thread up to the maximum available, showing how performance scales as more cores are engaged. Higher scores are better.
| 3DMark CPU Profile | Dell Pro Precision 5 16s Intel | Dell Pro Precision 5 16s AMD | Dell Pro Precision 7 16 | Lenovo ThinkPad P14s Gen 7 |
| Max Threads | 10,748 | 9,267 | 10,557 | 10,500 |
| 8 Threads | 6,818 | 6,573 | 5,996 | 6,550 |
| 4 Threads | 4,483 | 4,238 | 4,261 | 4,219 |
| 1 Thread | 1,210 | 1,178 | 1,185 | 1,165 |
The review unit sweeps every thread count, and its 10,748 max threads score is the best of any Pro Precision unit we have tested, 4.6% up on its 14-inch twin from the same silicon.
3DMark Storage and Blackmagic Disk Speed Test
3DMark Storage measures how an SSD performs during gaming-related tasks such as loading games, installing software, saving progress, and moving game files. Blackmagic Disk Speed Test measures an SSD’s sequential read and write speeds using large media files. Higher is better in both.
| Storage | Dell Pro Precision 5 16s Intel | Dell Pro Precision 5 16s AMD | Dell Pro Precision 7 16 | Lenovo ThinkPad P14s Gen 7 |
| 3DMark Storage | 2,804 | 2,325 | 2,189 | 3,094 |
| Blackmagic Write (MB/s) | 7,660.2 | 5,070.5 | 7,387.8 | 8,262.3 |
| Blackmagic Read (MB/s) | 8,418.1 | 5,071.2 | 6,776.2 | 8,511.5 |
The review unit’s SK hynix PCB01 posted solid Gen5 numbers, though this is one place the 14-inch twin’s Micron 4600 pulls ahead, reading nearly 9.8 GB/s against 8.4 here and scoring 3,093 in 3DMark Storage versus 2,804. Drive assignments vary within the family: the AMD twin’s SanDisk SN7100S lands near 5 GB/s in both directions, and the 7 16’s dual-drive RAID 0 volume wins nothing in these traces, trailing even in reads. Buyers with storage-sensitive workflows should watch which SSD their configuration ships with.
Blackmagic RAW Speed Test
The Blackmagic RAW Speed Test measures how many frames per second a system can decode Blackmagic RAW video on the CPU and on the GPU. We quote the 8K results at 12:1 compression, and higher is better.
| Blackmagic RAW Speed Test | Dell Pro Precision 5 16s Intel (Arc Pro B390) | Dell Pro Precision 5 16s AMD (Radeon 890M) | Dell Pro Precision 7 16 (RTX PRO 3000) | Lenovo ThinkPad P14s Gen 7 (RTX PRO 1000) |
| 8K 12:1 CPU (fps) | 77 | 74 | 77 | 77 |
| 8K 12:1 GPU (fps) | 87 | 49 | 137 | 96 |
CPU decode is effectively flat across the group at 74 to 77 fps. On the GPU, the 7 16 RTX PRO 3000 runs away with it at 137 fps through CUDA, while the review unit’s 87 fps over OpenCL comfortably clears real-time 8K playback, and the Radeon 890M falls short at 49.
Topaz Video AI
The Topaz Video AI benchmark measures AI video upscaling and frame-interpolation performance in frames per second across the application’s enhancement models, run here at 1080p input, where higher is better. The 16X Slowmo Aion model failed to complete on both Intel-graphics Precisions, this unit and the 7 16, while the AMD twin ran it without issue.
| Topaz Video AI (fps) | Dell Pro Precision 5 16s Intel (Arc Pro B390) | Dell Pro Precision 5 16s AMD (Radeon 890M) | Dell Pro Precision 7 16 (RTX PRO 3000) | Lenovo ThinkPad P14s Gen 7 (RTX PRO 1000) |
| Artemis 1X / 2X / 4X | 6.24 / 5.21 / 1.90 | 3.72 / 2.21 / 0.78 | 12.38 / 8.94 / 3.15 | 5.17 / 3.19 / 1.12 |
| Iris 1X / 2X / 4X | 5.29 / 3.19 / 0.96 | 4.91 / 2.72 / 0.92 | 13.06 / 7.49 / 2.32 | 5.91 / 3.12 / 1.01 |
| Proteus 1X / 2X / 4X | 6.45 / 6.09 / 2.47 | 3.99 / 2.70 / 1.18 | 12.06 / 8.60 / 2.66 | 4.78 / 3.14 / 1.05 |
| Gaia 1X / 2X / 4X | 3.30 / 2.26 / 1.51 | 1.87 / 1.34 / 0.96 | 3.73 / 2.66 / N/A | 1.62 / 1.13 / 0.79 |
| Nyx 1X / 2X | 1.56 / 1.57 | 1.87 / 1.53 | 3.59 / 3.08 | 2.40 / 2.09 |
| Hyperion HDR 1X | 3.23 | 11.48 | 15.58 | 14.56 |
| 4X Slowmo Apollo / APFast | 8.48 / 22.00 | 6.08 / 17.59 | 18.04 / 30.60 | 10.39 / 29.94 |
| 16X Slowmo Aion | DNF | 9.05 | DNF | N/A |
The 7 16’s RTX PRO 3000 leads the upscaling models across the board, roughly doubling the review unit in Artemis and Proteus, which is the expected order of things. More interesting is the middle of the table: the Arc Pro B390 beats the RTX PRO 1000 in Artemis, Proteus, and Gaia while losing the interpolation models and Hyperion HDR, the same split we saw on the 14s. The Aion failure on Intel graphics repeats here, one for the driver team.
UL Procyon AI Text Generation
The Procyon AI Text Generation Benchmark streamlines LLM performance testing by providing a concise, consistent evaluation method. It enables repeated testing across four local models, Phi, Mistral, Llama3, and Llama2, while minimizing the complexity of large models and the number of variables. Developed with AI hardware leaders, it optimizes the use of local AI accelerators to deliver more reliable, efficient performance assessments. All four systems ran the models through ONNX Runtime with DirectML on their GPUs.
| Procyon AI Text Generation | Dell Pro Precision 5 16s Intel (Arc Pro B390) | Dell Pro Precision 5 16s AMD (Radeon 890M) | Dell Pro Precision 7 16 (RTX PRO 3000) | Lenovo ThinkPad P14s Gen 7 (RTX PRO 1000) |
| Phi | 893 | 434 | 2,244 | 1,618 |
| Mistral | 646 | 403 | 2,062 | 1,397 |
| Llama3 | 669 | 357 | 1,859 | 1,252 |
| Llama2 | 750 | 390 | 1,990 | DNF |
The 7 16 owns this table, with its 12GB card also completing Llama2 at 1,990, where the 8GB RTX PRO 1000 could not load the model. Among the integrated GPUs, the Arc Pro B390 roughly doubles the Radeon 890M in Phi and Llama2, and the review unit’s 64GB memory pool finishes everything; the quiet advantage shared-memory platforms hold for local AI.
UL Procyon AI Computer Vision
The Procyon AI Computer Vision Benchmark measures AI inference performance across CPUs, GPUs, and dedicated accelerators using a range of neural networks, evaluating tasks such as image classification, object detection, segmentation, and super-resolution with models including MobileNet V3, Inception V4, YOLO V3, DeepLab V3, Real ESRGAN, and ResNet 50. The WinML runs use float32 on CPU and GPU, giving a like-for-like view across vendors. We also run the newer Computer Vision 2 suite through each vendor’s native path: OpenVINO in int8 on the Intel NPUs and fp16 on their iGPUs, TensorRT in fp16 on the RTX PRO 3000, and Ryzen AI on the AMD NPU; those results are listed separately since precision and runtime differ by platform. Higher scores are better.
| Procyon AI Computer Vision (WinML) | Dell Pro Precision 5 16s Intel | Dell Pro Precision 5 16s AMD | Dell Pro Precision 7 16 | Lenovo ThinkPad P14s Gen 7 |
| CPU | 143 | 114 | 122 | 134 |
| GPU | 410 | 245 | 553 | 426 |
| Procyon AI Computer Vision 2 (native runtimes) | Dell Pro Precision 5 16s Intel | Dell Pro Precision 5 16s AMD | Dell Pro Precision 7 16 |
| NPU (int8) | 1,630 | 1,189 | 1,614 |
| iGPU (fp16) | 1,529 | N/A | 837 |
| dGPU (TensorRT fp16) | N/A | N/A | 3,470 |
In the vendor-neutral WinML view, the review unit posts the best CPU score and lands within 4% of the ThinkPad’s dGPU on graphics. On the native paths, the Intel AI Boost NPUs in the review unit and the 7 16 are within 1% of each other, both more than a third ahead of the Ryzen AI NPU, and the review unit’s iGPU nearly matches its NPU. Note the 7 16’s iGPU score of 837 comes from the smaller integrated Arc in the Core Ultra 9 386H, not the B390 class part in the 5-series units; its TensorRT result on the RTX PRO 3000 towers over everything at 3,470.
UL Procyon AI Image Generation
The Procyon AI Image Generation Benchmark provides a consistent method for measuring AI inference performance from low-power NPUs to high-end GPUs, with three tests: Stable Diffusion XL FP16 for high-end GPUs, Stable Diffusion 1.5 FP16 for moderately powerful GPUs, and Stable Diffusion 1.5 INT8 for low-power devices. The benchmark uses the optimal inference path for each platform: OpenVINO on Intel systems, the AMD-optimized DirectML pipeline on Radeon GPUs, and TensorRT on NVIDIA GPUs. The INT8 test uses Intel’s quantized SD 1.5 model where supported; the AMD pipeline does not offer a comparable quantized run. The 7 16 runs this suite through TensorRT on its RTX PRO 3000. New with this round, we also ran the INT8 workload on the Intel NPU.
| Procyon AI Image Generation | Dell Pro Precision 5 16s Intel (Arc Pro B390) | Dell Pro Precision 5 16s AMD (Radeon 890M) | Dell Pro Precision 7 16 (RTX PRO 3000) | Lenovo ThinkPad P14s Gen 7 (RTX PRO 1000) |
| SD 1.5 FP16 | 638 | 316 | 1,424 | 943 |
| SD 1.5 INT8 | 7,778 | 2,855 | 17,287 | 12,403 |
| SDXL FP16 | 738 | 200 | 1,273 | 765 |
| SD 1.5 INT8 NPU | 2,881 | 2,870 | N/A | N/A |
At 50.8 seconds per SDXL image, the review unit sits within 4% of the RTX PRO 1000, mirroring the 14s result, while the 7 16’s TensorRT run turns in 1,273 at 29.4 seconds per image for those who need volume. The NPU run at 2,881 delivers about 37% of the iGPU’s INT8 throughput while leaving the GPU free. Scores are not comparable between the FP16 and INT8 rows since the workloads use different batch sizes and step counts.
SPECviewperf 15
SPECviewperf 15 measures graphics performance using viewsets derived from professional applications in CAD, 3D modeling, rendering, engineering, and medical visualization, replayed here at 1080p. Higher scores are better, although performance can vary considerably between applications and graphics architectures. The 7 16 ran SPECviewperf at its native 4K resolution, which is not comparable to the FHD runs, so it sits this table out; the enscape-01 viewset also produced an unusually low score on the AMD twin that is worth a driver revisit.
| SPECviewperf 15 (FHD) | Dell Pro Precision 5 16s Intel (Arc Pro B390) | Dell Pro Precision 5 16s AMD (Radeon 890M) | Lenovo ThinkPad P14s Gen 7 (RTX PRO 1000) |
| 3dsmax-08 | 21.46 | 26.23 | 28.20 |
| blender-01 | 23.15 | 23.17 | 40.69 |
| catia-07 | 26.35 | 22.70 | 43.36 |
| creo-04 | 68.14 | 49.42 | 107.59 |
| energy-04 | 38.86 | 29.69 | 45.79 |
| enscape-01 | 15.24 | 8.45 | 25.45 |
| maya-07 | 85.61 | 53.48 | 112.84 |
| medical-04 | 70.91 | 73.31 | 86.38 |
| snx-05 | 80.24 | 61.33 | 103.61 |
| solidworks-08 | 32.99 | 36.40 | 53.63 |
| unreal_engine-01 | 42.15 | 27.38 | 49.24 |
The review unit wins seven of the eleven viewsets against the Radeon 890M, with the certified CAD and engineering traces (creo, maya, snx, energy) showing the Arc Pro driver stack at its best. Unlike its 14-inch twin, it completed catia-07 cleanly at 26.35. The RTX PRO 1000 sweeps the table. For the record, the 7 16 native-4K run posted 114.70 in creo-04 and 63.99 in solidworks-08 even at four times the pixels, so its FHD equivalents would land higher still.
SPECworkstation 4
SPECworkstation 4 measures workstation performance across CPU, graphics, storage, AI, product design, engineering, financial services, and other professional workloads, using real applications grouped into seven industry verticals. Higher scores are better, and N/A means the system did not complete every workload required for that category. The standalone 7-Zip workload inside SPECworkstation failed on both the AMD twin and the 7 16, which suppresses their CPU subsystem and Productivity and Development scores; a rerun of the AMD twin reproduced the same failure, and our separate 7-Zip section above covers that ground.
| SPECworkstation 4 | Dell Pro Precision 5 16s Intel | Dell Pro Precision 5 16s AMD | Dell Pro Precision 7 16 | Lenovo ThinkPad P14s Gen 7 |
| Hardware Subsystems | ||||
| Graphics | 2.67 | 2.65 | 6.54 | 4.51 |
| Accelerator | 2.25 | 2.31 | 3.88 | 3.26 |
| Storage | 1.80 | 1.00 | 1.30 | 1.67 |
| CPU | 1.34 | N/A | N/A | 1.35 |
| Industry Verticals | ||||
| AI & Machine Learning | 1.47 | 1.45 | 1.76 | 1.65 |
| Energy | 1.55 | 1.38 | 1.87 | 1.69 |
| Financial Services | 0.94 | 1.29 | 1.03 | 0.96 |
| Life Sciences | 1.84 | 1.47 | 2.03 | 1.82 |
| Media & Entertainment | 1.60 | 1.48 | 1.87 | 1.81 |
| Product Design | 1.75 | 1.41 | 1.87 | 1.89 |
| Productivity & Development | 1.35 | N/A | N/A | 1.34 |
The review unit posts the best Storage subsystem score of the four systems. The 7 16’s Graphics subsystem score of 6.54 shows the dGPU gap at its widest, while the review unit and its AMD twin split the verticals, with AMD ahead in Financial Services and the review unit well ahead in Life Sciences and Product Design. Against the ThinkPad, the review unit trades verticals rather than trailing across the board, taking Life Sciences and splitting the productivity scores.
Conclusion
The Dell Pro Precision 5 16s Intel ended up delivering the strongest CPU performance we have seen from any B390-based system, along with the longest battery life across the Pro Precision lineup we tested. Its 18,965 Geekbench 7 multicore score even beat the Core Ultra 9-equipped Pro Precision 7 16, and 7-Zip compression topped the group. PCMark 10 was equally strong at 9,994, six points shy of the 10,000 mark and 25 behind the 14-inch Intel model. Battery life was even more impressive at 24 hours and 43 minutes, more than double the Pro Precision 7 16, nearly nine hours beyond the ThinkPad P14s Gen 7, and 53 minutes longer than the 14s Intel despite its larger QHD+ display.
The Arc Pro B390 also performed well across the professional graphics and AI tests, including strong certified ISV results, good Topaz upscaling performance, SDXL image generation within 4% of the RTX PRO 1000, and an NPU image generation score of 2,881.
Choosing between the Pro Precision models depends heavily on the workload. The AMD 16s has an advantage in heavily threaded rendering and y-cruncher, while the Intel model performed better in single-threaded work, graphics, AI, storage, and battery testing. Buyers who need substantially more GPU performance can move up to the Pro Precision 7 16, which was roughly 2.7 times faster in Blender and much faster in V-Ray, local AI text generation, and BRAW processing, but that configuration costs roughly $1,900 to $2,300 more and lasts less than half as long on battery. Compared with the Intel 14s, the 16s delivers very similar overall performance while adding the larger QHD+ display, numeric keypad, slightly better sustained CPU results, and another 53 minutes of battery life, although it is larger and heavier, and our review unit shipped with a slower SSD. The 16s also gives IT teams good service access, with a customer-replaceable battery along with replaceable LPCAMM2 memory, SSD, WLAN card, fan, and other internal components.
For configuration options and current pricing, visit the Dell Pro Precision 5 Series 16S product page.
Amazon