We have seen the Fusion technology Apple has used in the past for creating the “Ultra” versions of its M1, M2, and M3 chips. It skipped the M4 Ultra to focus on re-engineering Apple Silicon for the M5 line. Apple says this is the biggest chip design update since the original M-series debuted.

In this article, we will dive into details that others may miss and highlight the relevance of these chips to the AEC markets.

M5 Pro and Max

The M5 Pro and M5 Max are interesting because now the fusion technology is deployed between the Pro and Max, whereas previously it was deployed between the Max and Ultra. More than that, the division is between CPU and GPU, where before an entire Max chip was doubled using Apple Fusion to create the Ultra version.

Apple’s M5 Pro and M5 Max introduce a totally new Apple Silicon architectural strategy and industry-leading chip performance.

This time, the CPU for the Pro and Max is essentially the same, just binned with one core each, but the GPU die is doubled to get the Max variant from the Pro.

• The M5 Pro – 
• 5/6 – super cores
• 10/12 performance cores (18 core total max)
• 16/20 GPU cores

Or the:

• M5 Max –
• 6 – super cores
• 12 – performance cores (18 cores total)
• 32/40 GPU cores

Apple explicitly says in its press release that the “M5 Max pairs the 18-core CPU with an up-to-40-core GPU.” So the CPU and GPUs are definitely their own dies, and the GPU dies come in two sizes, and Apple may have engineered them so they are cuttable at the wafer level to control how many of each-sized GPU die they need.

Fusion Architecture Details

The new Fusion Architecture connects two dies with advanced IP blocks and maintains high bandwidth and low latency using advanced packaging. Exactly which sub-components on which die isn’t fully clear.

M5 Pro chip diagram (not official), based on Apple’s descriptions of the Apple Fusion Architecture. (Image: screen grab from Gary Explains, with Architosh-added text)

To make this efficient at the wafer level, the CPU and GPU dies would likely want to be the same die size, unless Apple engineered its chiplet so that TSMC produced the CPU and GPU dies on separate wafer runs. But how Apple achieves this at the manufacturing level isn’t really our concern, and we may learn about it eventually.

Super Cores and P Cores Only

Another interesting aspect of the M5 Pro and M5 Max chips is that they feature only performance and super cores. There are no efficiency cores for the Pro and Max. This isn’t a full departure from ARM’s big.LITTLE architecture philosophy as the super cores are larger than the performance cores, but it signals that Apple is gunning to compete with chip competitors trying to nip at its heels, especially at multicore CPU performance.

MORE: End of an Era: How Silicon Will Decide BIM’s Future

For instance, ARM competitor Qualcomm, with its Oryon architecture, only has “performance” and “prime” cores and explicitly omits traditional efficiency (or E) cores. Qualcomm’s thinking is that performance cores are efficient enough at low voltages to handle background tasks without needing purpose-built efficiency cores. Apple seems to have fully embraced this strategy as well, at least for the M5 Pro and M5 Max.

The GPU also features Neural Accelerators integrated into each GPU core. This is a game-changer for GPU performance and AI performance alike, and it shows in Apple’s AI benchmarks.

Node and Performance

The M5 Pro and M5 Max are manufactured on TSMC’s third-generation 3-nm process (N3P), which yields only a 4% improvement in transistor density but up to 30% faster multithreaded CPU performance, which is notable. GPU performance is also substantially faster than the M4 series. Even without the efficiency cores, the new node is helping Apple get up to 24 hours of battery life.

Another notable improvement is the memory speed-up. Bandwidth is now up to 307GB/s for the M5 Pro and 614GB/s for the M5 Max, with maximum memory of 64GB and 128GB, respectively.

MORE: Intel Debuts Panther Lake—Intel 18A: For CAD and BIM? 

For graphics performance, the new chips feature enhanced shader cores with second-generation dynamic caching and hardware-accelerated mesh shading. Also, Apple’s third-generation ray-tracing engine is onboard these new chips, which will improve rendering performance across numerous AEC applications.

Massive Single-Core

Apple leads the world in single-core performance, and the new super cores are the world’s fastest yet. This is partly driven by increased front-end bandwidth (wider front end, new cache hierarchy, and new branch prediction technology. Apple remains hyper-focused on industry-leading IPC (instructions per clock), and the ultra-wide fetch and decoder can process more instructions simultaneously before they even reach the execution units.

Vectorworks Architect is showcased on Apple’s M5 Pro and M5 Max webpage. The application is a model of modern code development for OS-centric optimization across platforms, but especially macOS. Maxon’s Redshift rendering engine is also noted and featured in this image. These kinds of tools are highly single-threaded and dependent for main design modeling tasks.

At the same time, the 30% improvement in multicore CPU performance is also dramatic, and we will see how the M5 Max fairs against the leading Apple Silicon chip on multithreaded performance, the M3 Ultra.

AI Performance

But beyond Fusion, Apple’s strategy of placing a Neural Accelerator into every single GPU core is another major architectural change in Apple Silicon, yielding big results. Apple says the M5 Pro and M5 Max offer over 4x peak GPU compute for AI compared to the previous M4 generation of chips. This is massive.

Additionally, the Neural Engine itself is faster than previous versions, and these chips also boast the latest Media Engine for video workflows and an industry-first Memory Integrity Enforcement and Thunderbolt 5. Learn more here.

Architosh Analysis and Commentary

Apple Silicon has traditionally been strongest in IPC and single-core performance, an area we have emphasized is absolutely critical to the CAD industry because CAD and 3D software are intrinsically resistant to parallelization. The entire M5 chip family continues to place emphasis on this industry leadership, which is a boon for performance-hungry BIM and CAD professionals using macOS native CAD and BIM platforms, from AutoCAD to Vectorworks and everything in between. 

It is notable that the M5 base model chip also features the same architecture core improvements, with four Super cores being paired with six Efficiency Cores. The base M5 chip maxes out its unified memory at 32 GB, which is the threshold you need to run Windows in Parallels in a professional AEC environment. This is the one regret we have, as a 48 GB option would have given the market a more affordable machine for contractors. They only need to run tools like Revit, SketchUp, and Bluebeam, along with web-based tools like Procore, but they are generally not authoring in tools like Revit. 

For architects who need access to Windows apps like Revit but prefer working on the Mac, the MacBook Pro with M5 Pro at 48 GB is an ideal starting point for a powerful mobile workstation with industry-leading snappiness across all native applications. 

Beyond all this, the new shader cores, third-gen ray-tracing engine, and Neural Accelerators in each GPU core all benefit visualization tools and workflows that architects and designers need.

The post Apple M5 Pro and M5 Max introduce new Fusion Architecture appeared first on Architosh.

The article, however, will debut first this week in the next Xpresso-4X newsletter arriving this Thursday, 18 December 2025. [Editor’s note: This article is now live]

Why Silicon Decides

The features excerpt reads:

“As computing enters the voltage-limited era, the shift away from x86 toward ARM—already reshaping datacenters, laptops, and cloud platforms—carries consequences for architectural computing. This feature explores how those silicon realities are redefining BIM’s future, and why adaptation is no longer an option.”

It is important to note that the notion that silicon drives software, not the other way around, is not new. It has largely always been the case.

As microprocessors gain transistors and computational performance, roughly every 10x improvement, a new type of computer (or computing era) emerges, with periods of overlap.

At every 10x leap in microprocessor power (performance), a new type of computing has emerged. If we look at this chart, we are now in the era of AI Compute. In the latter half of the first decade, we were in the era of smartphone and tablet compute (cellular devices) and smoothly entered the era of cloud compute (datacenter). Yet, today’s dominant BIM and CAD solutions were all architected during the Workstation and Desktop eras.

Yet, the tools we use today in CAD and BIM industries are originally from the Workstation and Desktop eras. Today’s BIM 2.0 solutions are challenging desktop-era tools by leveraging unique capabilities enabled by the cloud (datacenter) compute paradigm.

But the AI computer era is different. It is categorically scaling (innovation, investment, and speculation) at unprecedented speeds, not seen since the Dot-Com boom. Because of this scale, it is placing unprecedented emphasis on energy efficiency due to its exponential electricity use.

Heterogeneous Silicon

Mobility has been around for decades, but smartphones and tablets have exponentially increased the number of mobile computer devices. As a result, ARM processors today outnumber x86 processors by nearly 30:1. And ARM chips were designed from the beginning with heterogeneous compute in mind, including the big.LITTLE architectural paradigm, whereas x86 chips from Intel and AMD have been playing catch-up with ARM.

In the early PC era, Intel x86 chips outnumbered RISC processors from companies like Sun Microsystems, HP, IBM, and others by a ratio very similar to 30:1. The vast volume helped Intel at both economies of scale and Wright’s Law (net positive effects from experience gained from production). Now the shoe is on the other foot, and it is AMD and Intel that are battling those same benefits now accruing to ARM and its many licensees.

Neural CAD Engines

While x86 chips have fallen behind Apple and Qualcomm badly in performance per watt (and absolute performance in cases like single-core), AMD and Intel are formidable chip design powerhouses. Yet, the battle now becomes substantially more complex in the AI era, where system-on-a-chip (SoC) designs offer key benefits such as unified system memory and high throughput between CPU, GPU, and NPU (neural) cores. Intel and AMD are emulating these designs.

This view of the geometry-oriented AI Foundational Model (neural CAD engine) can create designs spontaneously from a text prompt. It is an entirely new machine learning approach to generating CAD objects, in contrast to classical parametric CAD engines that have been in use for 40 years.

Meanwhile, the CAD and BIM industry is increasingly leveraging NPUs and GPUs for new types of workloads based on AI inference and training. Autodesk’s Neural CAD engines were shown at AU25 in Nashville this past fall.

The special feature will cover in detail how Intel ran into physics at 10nm back in the middle of the last decade and how x86 contributes to the physics problem. It also highlights how Intel’s new 18A process node dramatically attacks these physics problems and why Intel may actually potentially lead again in some performance areas.

Viewers can read the special feature on Architosh on December 24th, or earlier by signing up for the Xpresso-4X newsletter. The newsletter goes out this week on Thursday, so sign up now to not miss out.

The post Preview: End of an Era: How Silicon Will Decide BIM’s Future appeared first on Architosh.

Revu 21.6

The June 2025 update adds a powerful Integrations Directory, connecting Bluebeam users with other tools and third-party ecosystems. These connections include such things as plugins and add-ons for tools such as Procore Documents, Microsoft SharePoint, Autodesk Construction Cloud (ACC), and Vectorworks. These tools are accessible across subscription plans.

Alongside this major update, Revu 21.6 now runs 30 percent faster on ARM-based devices, including Microsoft Surface (these would be ARM-chip-based Windows laptops) and Apple’s M-series Mac computers.

Bluebeam Revu 21.6 adds a new Integrations Directory, Revu support for ARM processor-based computers ,and Markups on Capture.

The mention of Mac computers should not be mistaken for native macOS support for Revu. The support of Revu on Mac is via Parallels software, a Windows hypervisor application for running Windows OS on Apple Mac computers.

Also new in version 21.6 is the ability to draw directly on field photos via the new Markups on Capture functionality. There are also Tool Chest upgrades with multi-step drag-and-drop and customizable punch keys for faster workflows.

“These updates reflect our commitment to solving real-world challenges faced by AEC professionals,” said Jason Bonifay, CTO at Bluebeam.

To learn more, visit here: https://www.bluebeam.com/whats-new/

The post AIA25: Bluebeam supercharges performance, new integrations appeared first on Architosh.

SPECworkstation 4.0 adds significant changes from version 3.1 and includes all key aspects of modern workstation performance workflows, including new AI and ML (machine learning) workflows.

SPECworkstation 4.0

The SPECworkstation 4.0 benchmark allows testers to gain a solid understanding of real-world CPU, graphics, accelerator, and disk performance for workstation computers, including any system that can run the benchmark, such as notebook computers.

SPECworkstation 4.0 is the latest benchmark for comprehensive workstation system performance testing.

The benchmark is directed to the needs of workstation users such as architects, engineers, scientists, and software developers who rely on workstation-class hardware for daily needs. It includes real-world leading applications used by professionals including Blender, Handbrake, LLVM and more. The benchmark provides comprehensive testing measures across several different industry verticals, each focused on specific use cases. These include:

• AI & Machine Learning
• Energy
• Financial Services
• Life Sciences
• Media & Entertainment (DCC)
• Product Design (MCAD)
• Productivity & Development

“A key value of this release lies in its ability to provide AI professionals with insights into how different hardware configurations handle diverse AI workloads,” said SPECwpc Chair Chandra Sakthivel. “This is essential for optimizing workstation setups for AI/ML tasks, driving the development of more efficient models, and ultimately advancing the capabilities of AI technology. As AI models evolve, a standardized benchmark that can assess performance across various dimensions – such as precision, throughput, and latency – will be invaluable for industry professionals and researchers.”

Key New Features of SPECworkstation 4.0

Designed to more accurately reflect today’s workloads, the new benchmark includes all of the following types of tests and measures:

• Support for AI/ML – The benchmark includes a dedicated set of tests focusing on AI and ML workloads, including data science tasks and ONNX runtime-based inference tests, reflecting the growing importance of AI/ML in workstation environments.
• New workloads

And…

• Autodesk Inventor: Measures key performance metrics in Inventor, a popular software package in the architecture, engineering, and construction (AEC) segment.
• LLVM-Clang: Measure code compilation performance using the LLVM compiler and toolchain.
• Data Science: Represents a data scientist’s workflow, including data scrubbing, ETL (extract, transform, load), and classical ML operations using tools such as Numpy, Pandas, Scikit-learn, and XGBoost.
• Hidden Line Removal: Measures time taken to remove occluded edges from wireframe models.
• MFEM: Uses finite element methods to perform dynamic adaptive mesh refinement (AMR).
• ONNX Inference: Benchmarks AI/ML inference latency and throughput using the ONNX runtime, a popular framework for evaluating machine learning models.
• Updated CPU Workloads: Updates for popular workloads like Blender, Handbrake, NAMD, and Octave improve relevance, accuracy, and compatibility for performance measurements.
• Updated Graphics Workloads: Tests adapted from the SPECviewperf2020 v3.1:  measure the performance of professional graphics cards using application-based traces.

There is also…

• New Accelerator subsystem – This release introduces the Accelerator subsystem, recognizing the rapid innovation in the computing space. With accelerators playing a critical role in speeding up tasks such as AI/ML processing, video transcoding, and other compute-intensive operations, measuring and understanding their performance in the workstation environment is essential.

SPECworkstation 4.0 boasts a completely redesigned UI that simplifies the benchmarking process, making it easier to understand and use. And a command line interface (CLI) enables easier automation.

To learn more and obtain the new benchmark, go here. It is free for the user community and USD 5,000 for sellers of computer-related products and services (both hardware systems and software systems, and services). SPEC/GWPG members receive benchmark licenses as a membership benefit.

About SPEC

SPEC is a non-profit organization that establishes, maintains and endorses standardized benchmarks and tools to evaluate performance for the newest generation of computing systems. Its membership comprises more than 120 leading computer hardware and software vendors, educational institutions, research organizations, and government agencies worldwide.

The post Major SPECworkstation 4.0 update—adds A/ML workloads appeared first on Architosh.

NVIDIA ARM

Nvidia’s client PC platform roadmap is reported to be ambitious, with multiple designs in development, including both Nvidia-only-based designs and designs in partnership with Taiwan’s MediaTek, an existing ARM chip provider.

Nvidia already has experience developing an ARM CPU in its Grace CPUs, which are based on the ARM architecture. These chips combine with Nvidia Hopper architecture-based GPUs to create the Nvidia Grace Hopper Superchip. It is theoretically possible that this formulation in a smaller size could be created for the Windows on ARM market, but more likely, Nvidia will deliver an SoC that combines a GPU, CPU, and AI chip into a single piece of silicon with shared integrated memory, similar to Apple’s M-series chips.

Nvidia’s Grace CPU is an ARM-architecture wonder, seen here paired with Nvidia Hopper architecture GPUs.

Scheduled for September of 2025, the first ARM SoC for the Windows market from Nvidia will likely target the high-end consumer market and be fit for purpose in laptops and small form-factor (SFF) desktops and workstation computers running Windows on ARM operating systems.

ARM Options

An SoC on ARM is not uncharted territory for Nvidia. They first delivered such a chip with their Tegra SoC. Nvidia may also surprise the market with an ARM SoC that pairs with a discreet Nvidia GPU and goes head-to-head with the top-performing systems from AMD and Intel for gaming PCs. Such systems would also appeal to professional users who need high-performance graphics plus the fastest CPUs they can obtain.

This is where the AEC market and CAD markets, in general, come into play. Leading workstation makers like BOXX Technologies deliver extremely high-performing professional workstations with both certified professional GPUs and top gaming-oriented GPU options. An ARM chip that can compete with Apple’s leadership in single-threaded performance combined with Nvidia’s top GPU technology would awaken the possible in the general Windows PC market.

MORE: A Big Data Difference — Vectorworks Sets Itself Apart with Data

The current fastest single-core threaded chip is the Apple M4 series, bursting through the 4,000 barrier in Geekbench 6. It is the first chip to crack that barrier. Like Apple and AMD, Nvidia is one of Taiwan Semiconductor Manufacturing Company’s (TSMC) largest customers and may have first dips on TSMC’s leading process nodes right there with Apple. In fact, Nvidia may deliver their first ARM SoC on TSMC’s new 2nm node which is expected to ship product in the second half of 2025. The September timing works well with TSMC on this node which is running ahead of schedule.

ARM Competition

Currently Qualcomm has an exclusive deal with Microsoft to ship SoCs for Windows on ARM that runs out in 2024. Starting next year anybody can deliver an ARM-based computer for the Windows on ARM operating system.

Qualcomm’s Snapdragon X Elite exceeded the performance of the Apple M2, and AMD and Intel’s top chips at the time of its release. The chart above shows single-threaded (ST) performance for the CPU test for Cinebench by Maxon.

Qualcomm is working on the sequel to its Snapdragon X Elite SoC which is expected to ship in mid-2025. This chip will compete with Apple’s M4 and future M5 and MediaTek’s debut ARM chip for the Windows on ARM market, expected in late 2025 or 2026. While AMD and Intel are both rumored to be possibly working on ARM SoCs for the Windows on ARM market, no specifics about such efforts exist.

Apple Response

With Nvidia likely expected to deliver the higher performing competition to Apple’s ARM chips, it remains to be seen how Apple will respond to this vigorous competition. One side note is that Apple may develop it own discreet GPU card for its Mac Pro workstation computer, which today runs an Apple M2 Ultra SoC.

That chip features a 24-core CPU and 60-core GPU and 32-core Neural Engine and 800GB/s memory bandwidth, which exceeds AMD’s EPYC server processors which peak at 460.8GB/seconds. And for further comparison, the Intel Core Ultra chip (Lunar Lake) offers a memory bandwidth of 136.5GB/second. Lunar Lake was developed to bust the myth that X86 can’t be as efficient as ARM chips. Lunar Lake was the first processor design that was essentially fabricated by TSMC rather than its own fabs.

Nvidia’s new ARM chips may offer PC gamers incredible performance at the CPU level combined with best-in-class GPU performance via discreet GPU cards. For Apple to match this kind of performance in its Mac Pro workstation, it must develop it own discreet GPU cards. There has been some indication Apple may be working on this, including a patent application described here on AppleInsider. On the flipside, Apple’s progress with its M-series chips may warrant them forgoing discreet GPUs indefinitely as it is only of value to the Mac Pro user, which is a very small subset of its overall Mac customers.

For Nvidia to exceed Apple and Qualcomm’s ARM performance won’t necessarily be easy but an advantage Nvidia can ultimately deliver would be CUDA and RTX support. If these technologies were onboard Nvidia ARM SoCs this could be a game-changer, especially for professional laptops that seek to be thin and light and yet offer industry-leading performance and superior performance per watt. Currently, Nvidia’s RTX 4090 laptop GPU is the fastest performing mobile GPU, but it is paired in top configurations with Intel’s Ultra Core CPU which is easily bested by Apple’s M4.

But imagine an Nvidia ARM chip that matches or exceeds Apple’s best M-chips and delivers thousands of CUDA cores and RTX capabilities.

Windows on ARM and CAD Industry

Finally, there is definitely some attention being paid to the Windows on ARM movement in the CAD industry. Specifically, both Nemetschek Group BIM companies Graphisoft and Vectorworks have made note about Windows on ARM. Graphisoft specifically mentioned at a press event his past October that they are investigating supporting it, while in a similar event by AEC software company Bluebeam, it was also reported that they are looking at Windows on ARM as well. Meanwhile, the first CAD program written for Windows on ARM of any note is the recent release of ILEXSOFT’s HighDesign R8.2. The well-executed 2D and graphics-oriented CAD program for AEC is natively built for both Windows on ARM and macOS’s Apple Silicon.

If the Windows on ARM movement takes on incredible momentum—and Nvidia entering this competitive space may certainly help that—we can definitely see the Windows app ecosystem go through a major transition as Microsoft itself is pushing for the success of Windows on ARM. Why are they doing that? Because Microsoft realizes that Intel and AMD can’t deliver competitive chip offerings to out muster Apple in the laptop market. And the laptop market is very important.

In the overall PC market, more than two thirds of all PCs are laptops. In fact, it is greater than that in most quarters these days. In Q1, 2024, Canalys reported that total PC shipments equaled 57.2 million with desktops taking up only 12.1 million compared to notebooks at 45.1 million units. Apple’s global share of the notebook market is about 12 percent, where overall Mac share in the US has exceeded 30 percent. To Microsoft, Windows on ARM is meant to stop Apple’s consistent climb in market share.

The post Nvidia aims for high-end ARM chip in September 2025 appeared first on Architosh.

“The introduction of Allplan 2025, Vectorworks 2025, and Archicad 28 underscores our dedication to empowering architects, engineers, and designers with cutting-edge digital solutions. These products are not only about improving workflows but also about driving innovation and sustainability in the AEC/O industry,” said Yves Padrines, CEO of the Nemetschek Group, on the parallel launch of the three flagship products. “We are providing our customers with a true choice. Our AI-powered tools and enhanced collaboration features will help professionals design more efficiently and sustainably, making a lasting impact on the future of construction and design.”

Allplan 2025

Allplan, the popular German AEC industry BIM solution used worldwide, announced cutting-edge new AI-driven features and automation in ALLPLAN 2025. Allplan also announced a new ALLPLAN Civil product that delivers a comprehensive set of civil and structural modeling capabilities in a product formerly known as ALLPLAN Bridge for the more generalized civil and transportation infrastructure market. This Civil product offers the market a complete design-to-build workflow for all types of transportation infrastructure sector projects.

All three Nemetschek Group BIM solutions now available worldwide.

New features in ALLPLAN 2025 boast real-time collaboration capabilities, new connected workflows to structural engineering, and advanced detailing tools. AI Visualizer technologies are also included as part of ALLPLAN 2025.

Vectorworks 2025

The latest version of the popular Vectorworks BIM/CAD software product line offers cutting-edge and industry-first features in a highly intuitive interface. Delivering solutions for architecture and urban design, interiors professionals, landscape design professionals and entertainment industry professions across a range of targeted solutions, users of Vectorworks 2025 products gain new Onscreen View control capabilities useful to 3D model navigation, sophisticated and industry-first Object Level visibility controls that can be associated and driven by data logic, plus a brand new Vectorworks Cloud Document Reviewer which sets the stage for 3D model viewing and further integrations with sister company Bluebeam.

MORE: Behind the Scenes of Vectorworks 2025: Leadership Talks Innovation with Architosh

Vectorworks 2025 further solidifies its leading position in the entertainment industry with its Vectorworks Spotlight 2025 and Vision 2025 products alongside BraceWorks and ConnectCAD for professionals in film, tv, theater, music entertainment and events design.

Archicad 28

Archicad 28 is the latest version of the popular global BIM solution and the latest product brings further enhancements to collaborative workflows, integrations with Bluebeam Sessions, advanced automation, unique Design Options capabilities, improved workflows with Revit files and much more, including excellent new Key Noting tools.

MORE: Budapest Part 2: Details from the Graphisoft Launch Event Press Conference

Archicad 28 also complements new BIMcloud and BIMx features—the latter of which delivers BCF support for field to office workflows, new view by layers capabilities and object mode view isolation features turning BIMx into one of the industry’s best BIM model viewer applications in the market. Additionally, BIMx supports Apple’s Vision Pro headset. A key focus in Archicad 28 is sustainability, helping architects design better-performing buildings when it comes to energy consumption. Archicad was also the first solution within the Nemetschek Group to integrate AI Visualizer technology into its rendering workflow options.

Driving Innovations

The Nemetschek Group feels bolstered and proud of its three main BIM solutions and their technical leadership in driving digital transformation in the AEC/O industry. ALLPLAN 2025, Vectorworks 2025, and Archicad 28 are now available worldwide in multiple languages.

To learn more about the Nemetschek Group visit them here.

The post Available Worldwide — Allplan 2025, Vectorworks 2025 and Archicad 28 appeared first on Architosh.

BOXX Workstations

BOXX workstations are a trusted brand for custom and pre-configured workstations for AEC, DCC, and M&E industries, as well as scientific and medical computing. The company makes workstations, rendering systems, and servers for these high-performance computing industries.

APEXX S-Class workstations now feature the new Intel Core Ultra 7 and Ultra 9 CPUs, delivering excellent performance for CAD, BIM, 3D design, motion media, and other professional software applications. As is often the case, BOXX Technologies is the first to the workstation market with these new Intel processors.

BOXX Technologies was the first workstation maker to ship the new Intel Core Ultra series CPUs in professional workstations in the United States.

“Intel Core Ultra processors unlock new levels of performance with features like AI optimization, increased PCIe lanes, and maximized peripheral connectivity,” said BOXX founder and Chief Technical Officer Tim Lawrence. “Whether you’re 3D modeling, video editing, engineering, designing, or utilizing AI enhancements, BOXX APEXX S-class systems powered by Intel Core Ultra processors will accelerate your creative applications and render engines through their most demanding tasks.”

Intel Ultra Core

The new chips feature Intel’s 24-hybrid core architecture and leading industry technologies designed to improve software efficiency. Such features include:

• Multiple CPU cores
• 24 total CPU PCIe lanes
• New AI capabilities in silicon

BOXX Technologies has coupled these Intel chips with advanced liquid cooling and 192 GB of memory. The smaller APEXX S3 tower is configurable with NVIDIA RTX or NVIDIA GeForce RTX GPUs, while the larger APEXX S4 can fit up to two NVIDIA GPUs. Both systems feature DDR5 memory with Thunderbolt 5 and WiFi 7 enable fast connections.

These systems are purpose-built for pro apps like SOLIDWORKS, Autodesk 3ds Max, Autodesk Maya, Autodesk Revit, Maxon Cinema 4D, Chaos V-Ray and more.

“The new Intel® Core™ Ultra processors available now inside our APEXX S-class workstations maximize application performance for architects, engineers, 3D designers, and motion media creators,” said Kirk Schell, CEO, BOXX Technologies. “When you couple these best-in-class CPUs with BOXX innovative engineering like overclocking, liquid cooling, and expandability, creators have all the ingredients necessary to realize their vision, accelerate their workflow and outpace their competition, while experiencing genuine workstation ROI.” 

For further information and pricing on the APEXX S3 and APEXX S4 workstations, visit BOXX online here.

The post Product Watch: BOXX APEXX S3 and S4 with Intel Core Ultra CPUs appeared first on Architosh.

Now, Nvidia is aiming to enter the ARM for the personal computer market, and the AI chip leader is highly competitive and offers interesting advantages.

NVIDIA vs Apple vs Qualcomm

When it comes to performance per watt, Apple has ruled the world since the 2020 introduction of Apple Silicon (with its M1 SoC). But it hasn’t just been performance per watt; Apple has also captured absolute performance. Data released on the Internet show that the new Apple M4 chip line-up outperforms all other rivals in sheer computational performance.

Take a look at this data table with Geekbench 6 scores.

Note how the M4 even beats out AMD and Intel’s top-of-the-line brand-new desktop-class PC chips. And the M4 Max is the first chip in the world to break through the 4,000 barrier for Geekbench 6.

NVIDIA’s Grace CPU is a custom ARM-based chip. In the image below we see the ARM-based Grace CPU on the left and the Hopper GPU on the right, in a complete package.

Just when Qualcomm delivered some exciting ARM for Windows news with its Snapdragon X Elite and suggested it could take on Apple’s M-series as well as AMD and Intel, all three competitors stepped up with their latest and greatest and delivered greater performance.

So what will Nvidia do in this race?

Heated Competition

It was only a few years ago when the PC market had but just two players (on Windows) and one player (on Mac) and it was just Intel versus AMD. Now the veteran leaders in the CPU market are not just locked into a very heated battle for supremacy but have been smoked by Apple’s ARM-based M-series chips. With Qualcomm committed to this competition we now are about to add both Nvidia and MediaTek. That makes for six SoC semi providers for the PC market. What a remarkable turn of events.

Nvidia Offering

First of all, Nvidia is not new to ARM. The company’s Grace-Hopper CPU is a custom ARM processor. And years ago Nvidia developed the Tegra SoC with an ARM Cortex core for the Nintendo Switch. And remember that Nvidia wanted to buy ARM but was stopped by anti-trust regulators.

Now NVDIA has teamed up with MediaTek (a Taiwan-based ARM chip maker) to deliver a new generation of ARM-powered NVIDIA chips.

The new NVIDIA ARM chip is likely to be produced en masse in Q1-2025. Here is what we know about it.

• Nvidia’s ARM SoC will be a “high-end” chip (aiming to take away Apple’s leadership)
• Will go into production in 2025
• Targeted at high-performance Windows on ARM laptops and gaming handhelds
• 8-16 cores (likely mixed between efficiency and performance cores
• Manufactured on 3nm TSMC process node (expect the same node as Apple’s M4)
• LPDDR6 Memory
• RTX GPU
• Nvidia AI engine onboard

According to reports from DigiTimes and Tom’s Hardware, Nvidia’s plans involve both a co-engineered chip with MediaTek and a fully custom in-house design. The latter clearly coming later than the former.

Architosh Analysis and Commentary

At the moment Qualcomm is the exclusive chip maker for Windows 11 on ARM and certified Microsoft Co-Pilot AI PCs. But that deal ends in 2024, whereupon other chip makers could deliver ARM-based ships for Microsoft Co-Pilot AI PCs. Our estimate is Nvidia will enter that same space to compete head-on with Qualcomm in the ARM Windows world and deliver game-changing AI processing capabilities to rival Apple’s Macs with Apple Intelligence and best both AMD and Intel with their on-chip AI capabilities. 

At the same time, if there is one chip company that can deliver a SoC that can best Apple’s leadership, especially when it comes to graphics, it is definitely going to be Nvidia. Meanwhile, AMD is also rumored to be designing an ARM-based PC chip.

The performance crown has been on Apple’s head for four years since the M1. Apple’s M4 just thrashed AMD and Intel’s best. The diminutive Mac mini with M4 Max likely delivers the great compute performance per dollar of any computer in the world. (note: we will formally investigate that hunch in a separate article). While Qualcomm ended up with Nuvia’s ARM chip expertise (led by former Apple chip engineers), a competition with Nvidia will surely light a competitive fire under Apple. With such heated competition, computer users are the big winners here. 2025 is shaping up to be truly exciting! 

The post Nvidia to enter PC market with high-end ARM chip appeared first on Architosh.

These new Ryzen 9000 series chips deliver excellent performance upgrades for CAD and 3D users across industries and for AI workloads.

New BOXX Workstations

“From the desk side to the data center, BOXX offers the widest range of workstations powered by state-of-the-art AMD Ryzen 9000 Series Processors,” said BOXX CEO Kirk Schell. “So architects, engineers, 3D designers, and motion media professionals can not only create, render, and multitask faster than ever before, they can do so on a highly reliable system ideally suited to their particular workflow.”

The new BOXX AMD Ryzen 9000 series powered workstations.

Zen 5 Architecture

The new AMD Ryzen 9000 series processors are based on the Zen 5 CPU architecture and offer up to 35 percent faster performance compared to the previous Ryzen generation. The BOXX systems outfitted with 9000 series processors run at CPU frequencies of up to 5.7 GHz and provide 16-core/32-thread performance.

APEXX A3

The compact APEXX A3 subcompact tower features up to two professional-grade GPU cards, 192 GB of 5600 MHz DDR5 memory, PCIe 5.0, 2.5 GB ethernet, USB 4, and support for SSDs, plus a 750-watt power supply. The system is ideal for architects, engineers, and 3D artists working with professional software like Autodesk Revit, Maya, Cinema 4D, V-Ray, Adobe Creative Cloud, and much more.

APEXX A4

Similar to the APEXX A3, the A4 delivers the same components but in a more spacious form factor, capable of supporting large-scale top-of-the-line NVIDIA GeForce GPUs. It also features a 1200-watt power supply and can be mounted in a rack.

The RAXX A1G is a rack-mounted exclusive workstation designed for enterprises and a superior alternative to VDI (virtual desktop infrastructure). It features the same AMD Ryzen 9000 series processors yet in a 1U chassis with a 1300-watt power supply.

Finally, the FLEXX A1G platform is meant for deployment in data centers for remote work access.

Liquid Cooling Technology

BOXX’s highly configurable professional workstations have long offered industrial-strength and reliable, closed-loop liquid cooling systems. Architosh Labs has one BOXX workstation it uses for reviews testing and we can attest to BOXX’s reliability.

BOXX Workstations is a leading workstation provider that serves the AEC industry, among other CAD and 3D-intensive industries. The workstation maker provides in-depth guidance on what kind of machine to configure based on the specific applications a user will be using.

“At BOXX, we’ve earned our reputation of being first-to-market with game-changing technology for any workflow,” said Schell. “Our innovative integration of the AMD Ryzen 9000 Series inside the APEXX, RAXX, and FLEXX A-class workstations demonstrates our commitment to providing creative professionals and enterprise organizations with the best, most reliable tools purpose-built to optimize their productivity and maximize profits.”

For further information and pricing on APEXX A-class workstations, RAXX, and FLEXX, contact BOXX at 1-877-877-2699. Learn more about AMD Ryzen processors, BOXX solutions, finance options, and how to contact worldwide BOXX resellers by visiting www.boxx.com. 

Architosh Analysis and Commentary

The AMD Ryzen 9000 series CPUs follow up on the Ryzen 7000 series from nearly two years ago. The Ryzen 9600X is a 6-core chip, the 9700X is an 8-core chip, the 9900X is a 12-core chip, and finally, the 9950X is a 16-core chip. On Geekbench, performance is up 35 percent—which is noteworthy—but on many other benchmarks, from Speedometer to Cinebench, the performance uptick is half that. Some top Geekbench single-core scores are:

• Ryzen 9600X = 3157 
• Ryzen 9700X = 3303
• Ryzen 9900X = 3507
• Ryzen 9950X = 3460+

There are other scores that go up and down, but these are some of the higher scores in the search results. To put that in context, Apple’s much-valued M3 Max chip is 3128. So these are big scores and exceed the highest confirmed average score for a CPU, the Intel Core i9-13900KS, at 3140. 

The post New BOXX Workstations Feature AMD Ryzen 9000 Series Processors appeared first on Architosh.

Radeon PRO W7900 Dual Slot

The W7900 Dual Slot graphics card differs from the regular Radeon PRO W7900, which is a triple slot unit. The W7000 series cards are the world’s first pro workstation graphics built with chiplet technology, something AMD has innovated in partnership with Taiwan Semiconductor Manufacturing Company (TSMC). Built on TSMC’s 5nm GCD with 6nm MCD combo chiplet strategy, this GPU is a multi-chipset module (MCM) implementation versus what the industry would refer to as a monolithic implementation. For those that are interested they can learn more about this here.

This AMD Radeon PRO W7900 graphics card is AMD’s most powerful performer and offers higher TFLOPs per dollar than NVIDA’s Ada generation.

Key stats for the new AMD Radeon PRO W900 Dual Slot Card include:

• GPU Architecture = AMD RDNA 3
• Process = 5nm and 6nm, multi-chipset design
• Transistor Count = 57.7 billion
• 122.62 TFLOPs (FP166 / peak half precision)
• Compute Units = 96
• Ray Accelerators = 96
• AI Accelerators = 192
• Stream Processors = 6144
• 48 GB memory (GDDR6), ECC = yes
• Memory Interface = 384 bit
• Peak Memory Bandwidth = 864 GB/s

The graphics card also features 96 MB AMD Infinity Cache Technology. Display outputs include 3x DisplayPort 2.1 and 1x Enhanced Mini DisplayPort 2.1. HDR support up to 12K.

Advanced ray tracing with its built-in ray accelerators (all 96 of them).

Software OS API support includes Microsoft Windows and Linux operating systems. DirectX 12, OpenGL 4.6, OpenCL 2.1 and Vulkan 1.3, support. Supported technologies include AMD Viewport Boost, AMD Remote Workstation, AMD Radeon Media Engine, AMD Software PRO Edition, AMD Radeon VR Ready Creator, and AMD Radeon ProRender.

Architosh Analysis and Commentary

The AMD Radeon PRO W7900 Dual Slot is the most powerful professional AMD workstation GPU you can obtain. It offers 61.3 TFLOPs single. The triple-slot version retails for USD 3,999. That works out to USD 65.23 per TFLOP. The NVIDIA Ada RTX 6000 graphics card retails for USD 6,800. It delivers 91.1 TFLOPs single precision performance, more than the Radeon PRO W7900. However, on a cost-per-performance basis, it works out to USD 74.64 per TFLOP. So, on a performance basis, the AMD is more economical as NVIDIA’s top-of-the-line professional GPU is about 14 percent more expensive on a cost-per-TFLOP basis.

AMD has brought its industry-leading innovations in chiplet technology to its professional graphics GPUs.

For further comparison, the just-cited Ada RTX 6000 also features 48GB of memory. The memory is naturally ECC, as well. It also features 4x DisplayPort technology but version 1.4. In terms of ray tracing, the Ada GPU contains 142 RT cores compared to the 96 ray accelerators for the AMD card; this is 47 percent more ray tracing cores for 70 percent more cost. To learn more about the new AMD Radeon PRO W7900 graphics card, go here. 

Our latest review of an AMD graphics card is here.

The post A Look at AMD’s Radeon PRO W7900 Dual Slot GPU appeared first on Architosh.

New GPUs

The new RTX A400 GPU is a single-slot card that delivers 24 Tensor Cores for AI processing, helping professionals work faster with cutting-edge AI technology such as AI chatbots and copilots. For visualization professionals, the GPUs feature second-generation RT (ray tracing) cores for photorealistic, physically-based rendering and visualization workflows in AEC and design-manufacturing industries.

The RTX A400 GPU features 4GB of GPU memory, third-generation Tensor Cores, and seventh-generation encode (NVENC) and fifth-generation decode (NVDEC) engines for video processing workflows.

The new NVIDIA RTX A1000 GPU shown here, with 8 GB of memory.

The A1000 GPU features 8 GB of GPU memory and everything in the RTX A400. Both GPUs are designed to expand the reach of RTX technology. To learn more, visit here and here.

Architosh Analysis and Commentary

We wonder who these GPUs are actually for in the markets we cover here at Architosh. At just 4 GB of video memory, we wonder if the RTX A400 is a sound investment.

MORE: Product Review: NVIDIA RTX A2000 GPU for Workstations

AEC professionals looking for AI, GPU compute, and advanced rendering workflow assistance should consider the Ampere Architecture generation RTX A2000 RTX as a baseline GPU, but perhaps the 8 GB A1000 is the new baseline.

The post New NVIDIA RTX A400 and A1000 GPUs Announced—RTX for Entry Level appeared first on Architosh.

RTX 500 and 1000

Whether you are an architect, engineer, 3D content creator, or AI researcher, the new NVIDIA RTX 500 and 1000 Ada Generation laptop GPUs will soon be available in new, thinner, and highly portable mobile workstations.

NVIDIA’s Ada Lovelace generation GPU chip powers the new RTX 500 and RTX 1000 mobile GPUs.

The Ada Lovelace architecture-based GPU lineup now includes the RTX 2000, 3000, 3500, 4000, and 5000 laptop GPUs.  The new RTX 500 and 1000 include neural processing units (NPUs), a component of the CPU, and an NVIDIA RTX GPU, with its onboard tensor cores for AI processing. The NPU offloads light AI tasks while the GPU delivers an additional 682 TOPS of AI performance.

The Power of AI

Mobile workstation professionals across a range of industries can benefit from the enhanced Ada Lovelace architecture and AI processing benefits. Video editors can remove background noise with AI. 3D rendering professionals can use AI to accelerate high-fidelity images and animations.

Architects can leverage a 14x performance increase in generative AI performance from models like Stable Diffusion, which is now at the core of tools like Graphisoft’s Archicad 27, to iterate visualized concepts in architecture rapidly.

The RTX 500 and 1000 GPUs deliver the following general results:

• Third-generation RT Cores: Up to 2x the ray tracing performance of the previous generation for high-fidelity, photorealistic rendering.
• Fourth-generation Tensor Cores: Up to 2x the throughput of the previous generation, accelerating deep learning training, inferencing and AI-based creative workloads.
• Ada Generation CUDA cores: Up to 30% the single-precision floating point (FP32) throughput compared to the previous generation for significant performance improvements in graphics and compute workloads.
• Dedicated GPU memory: 4GB GPU memory with the RTX 500 GPU and 6GB with the RTX 1000 GPU allows users to run demanding 3D and AI-based applications, as well as tackle larger projects, datasets and multi-app workflows.
• DLSS 3: Delivers a breakthrough in AI-powered graphics, significantly boosting performance by generating additional high-quality frames.
• AV1 encoder: Eighth-generation NVIDIA encoder, aka NVENC, with AV1 support is up to 40% more efficient than H.264, enabling new possibilities for broadcasting, streaming and video calling.

The NVIDIA RTX 500 and 1000 Ada generation laptop GPUs will be available spring of 2024 in new mobile workstations from global computer manufacturers, including partners Dell Technologies, HP, Lenovo, and MSI.

The post NVIDIA launches new RTX 500 and RTX 1000 laptop GPUs appeared first on Architosh.

RTX 2000 Ada Generation GPU

Delivering the same SFF (small form-factor) GPU but powered by the Ada Generation architecture, the new GPU features up to 16 GB of GPU memory, vastly more than the 6 GB found in the RTX A2000. When it comes to performance, the RTX 2000 Ada Generation delivers substantial performance gains over the RTX A2000. 12GB version GPU (a model we did not review). Rendering performance is up to 50 percent faster, while Generative AI and Graphics are around 35-40 percent faster.

The double-slot wide and half-length SFF GPU not only delivers 12.0 TFLOPS (SPP) butalso 27.7 TFLOPS for its RT Cores and 191.9 TFLOPS for its Tensor Performance.

NVIDIA RTX 2000 Ada Generation is the latest professional GPU for workstation users. Built on a much smaller node and at TSMC compared to the A2000 fabricated at Samsung, this latest GPU comes with a lot more processing power in an SFF but at a larger price. (click on image for larger view).

The Ada Lovelace architecture-based CUDA cores deliver exactly 1.5X the speed of the previous generation for single-precision floating-point (FP-32) operations, driving up performance for graphics, simulation, and complex 3D CAD and CAE (computer-aided engineering) workflows.

For rendering, NVIDIA’s third-generation RT cores help deliver 1.7X speed up over the previous generation, while 4th gen tensor cores speed up AI workloads.

Quick Specs:

• 16 GB GDDR6 ECC memory
• 2,816 CUDA Cores
• 4x mini DisplayPort 1.4a
• 70 W max power draw
• PCIe Gen 4×8
• VR Ready = Yes
• Dual Slot / 6.6″ length
• Active thermal fan (onboard)

NVIDIA is charging USD 625 for this latest SFF GPU aimed at DCC and CAD/CAE. Readers can learn more about the details of this latest GPU here.

The post NVIDIA announces RTX 2000 Ada Generation GPU appeared first on Architosh.

40 SUPER Series GPUs

Using the latest iteration of the NVIDIA Ada Lovelace architecture, these newest GeForce RTX cards deliver up to 52 shaders TFLOPS, 121 RT TFLOPS, and 836 TOPS.

PC gamers are always demanding the highest visual quality, and those games leveraging the AI-powered NVIDIA DLSS technology (deep learning super sampling super-resolution) will experience the very best in visual quality. Stunning game worlds are to be found in touch titles as Diable IV, Pax Dei and Horizon Forbidden West.

New NVIDIA GeForce RTX 40 SUPER Series GPUs.

With this DLSS technology, seven out of eight pixels can be AI-generated, boosting overall full ray tracing rendering by up to 4x with better image quality.

“For everyone from gaming enthusiasts to creative professionals, GeForce RTX SUPER GPUs are simply awesome upgrades,” said Matt Wuebbling, vice president of global GeForce marketing at NVIDIA. “GeForce RTX SUPER cards support over 500 RTX games and applications and will have users prepared for the wave of generative AI apps coming to PC.”

Adobe Photoshop is one of those apps that take advantage of AI today. The popular graphics app leverages Tensor Cores to accelerate workflows. Here is a look at the three new GPUs available:

NVIDIA GeForce RTX 4080 SUPER

• 16 GB GDDR6X memory
• 256-bit memory interface width
• 7168 Cuda Cores
• 1.98 GHz base clock
• 52 TFLOPS for Shader Cores (Ada Lovelace architecture)
• 121 TFLOPS for Ray Tracing cores
• 836 AI TOPS for Tensor Core performance

NVIDIA GeForce RTX 4070 Ti SUPER

• 16 GB GDDR6X memory
• 256-bit memory interface width
• 7168 Cuda Cores
• 1.98 GHz base clock
• 44 TFLOPS for Shader Cores (Ada Lovelace architecture)
• 102 TFLOPS for Ray Tracing cores
• 706 AI TOPS for Tensor Core performance

NVIDIA GeForce RTX 4070 SUPER

• 12 GB GDDR6X memory
• 192-bit memory interface width
• 7168 Cuda Cores
• 1.98 GHz base clock
• 36 TFLOPS for Shader Cores (Ada Lovelace architecture)
• 82 TFLOPS for Ray Tracing cores
• 568 AI TOPS for Tensor Core performance

To learn more about the 4080 SUPER, visit here. To learn more about the 4070 SUPER family, visit here.

The post CES24: Nvidia announces GeForce RTX 40 SUPER GPUs appeared first on Architosh.

AMD says the new Radeon PRO W7700 is the most powerful professional workstation graphics card under USD 1,000.

AMD Radeon PRO W7700

“We built and optimized the AMD Radeon PRO W7700 graphics cards to create premium workstation graphics experiences, enabling our customers to boost innovation and productivity,” said Dan Wood, corporate vice president, Radeon product management at AMD.

The all-new AMD Radeon PRO W7700 professional graphics card for demanding workstation users.

The new workstation graphics card is designed to meet the evolving demanding nature of workflows with complex 3D visuals, special effects, and explosive growth in AI applications. The new GPU is designed to deliver exceptional performance.

The new Radeon PRO W7700 features:

• 48 Compute Units and Ray Accelerators
• 96 AI Accelerators
• 28 TFLOPS (peak single precision – FP32)
• 16 GB of GDDR6 memory with ECC
• 256-bit memory bus
• 4X UHBR 13.5 DisplayPort 2.1 (display ports)

With 16 GB of VRAM, the new GPU allows 3D artists and video editors to work with high-resolution models and textures with on-card memory, allowing faster workflows. Plus the new GPU card supports the latest codecs and AV1 encode/decode and AI-enhanced video encoding capabilities.

The Radeon PRO W7700 card leverages the AMD RDNA™ 3 architecture, and AMD says it is 1.7X better price/performance than competitor offerings. The new card boasts 52 percent higher performance in SOLIDWORKS 4, up to 24 percent higher performance in Creo 5, and 37 percent higher performance in CATIA 6.

“The great collaboration between AMD and ACCA continues five years strong. Architectural customers have benefited from the combination of AMD Radeon ProRender in our ACCA Edificius BIM application, thanks to the integration of HIP,” said Guido Cianciulli, founder and CEO, ACCA Software. “We are excited that ACCA Edificius customers will benefit from the new AMD Radeon PRO W7000 series workstation graphics, which provide increased photorealism and improved image rendering power, as well as support for ultra-performance displays.”

Pricing and Availability

The AMD Radeon PRO W7700 workstation graphics card is expected to be available November 13, 2023, from leading etailers/retailers. Product availability in OEM workstations and SI systems is expected to begin in November, 2023. The AMD Radeon PRO W7700 workstation graphics card has an SEP of $999 USD.

To learn more visit here.

The post AMD Radeon PRO W7700 GPU Launched appeared first on Architosh.

The new workstations are ideal for professionals in fields such as AEC, M&E (media and entertainment), healthcare, education, and scientific research.

ThinkStation P8

The new workstation features 5nm Zen 4 architecture Ryzen Threadripper PRO chips. This silicon features up to 96 cores and 192 threads of processing. Designed for unrivaled performance, the new ThinkStation P8 builds on the success of the award-winning P620, the world’s first workstation powered by an AMD Ryzen Threadripper PRO processor.

The workstation case was designed for optimal thermals by Aston Martin and pairs the industry’s strongest CPU and GPU industry architectures (AMD Ryzen and NVIDIA RTX).

The new Lenovo ThinkStation P8 offers industry-leading performance featuring AMD Ryzen Threadripper PRO processors.

“At Lenovo, we understand that our customers need high-quality workstations that can adapt to their changing and diverse needs. That’s why we collaborated with AMD and NVIDIA to create the ThinkStation P8, a workstation that combines power, flexibility and enterprise-grade features,” said Rob Herman, vice president and general manager, Workstation and Client AI Business Unit, Lenovo. “Designed to offer unparalleled performance and scalability, whether to run complex simulations, render stunning visuals, or develop cutting-edge AI applications, the ThinkStation P8 can handle it all. And with Lenovo’s certifications, security and support, you can trust that the ThinkStation P8 will exceed expectations.”

The P8 can be a tower or can optionally be put into a rack environment and offers platinum-rated power supplies for highly demanding expansion configurations. For example, it can support up to three (3) NVIDIA RTX 6000 Ada Generation GPUs to help reduce time to completion in graphics-intensive applications like real-time ray tracing or video rendering, simulation, or computer-aided design.

”We are delighted to power the latest ThinkStation P8 tower workstations with our cutting-edge AMD Ryzen Threadripper PRO 7000 WX-Series processors to bring leading performance and bandwidth to overcome ever more complex workflows,” said Jason Banta, corporate vice president and general manager OEM PCs, AMD. “Architects, engineers, visual artists and AI model developers will relish the power, flexibility and reliability that ThinkStation P8 with AMD delivers.”

“The Lenovo ThinkStation P620 with AMD Threadripper PRO technology has been an absolute game-changer for our 3D animation and development workflows over the last two years,” said Bill Ballew, CTO, from DreamWorks Animation. “We are looking forward to significantly faster iterations due to the increased performance with the new ThinkStation P8 workstation powered by AMD Threadripper PRO 7000 WX Series in this coming year.”

The ThinkStation P8 supports up to 7 PCIe Gen4 SSDs with RAID support or 3x HDDS for mega storage and up to 2TB of DDR5 memory.

The P8 workstation can accommodate up to seven (7) M.2 PCIe Gen 4 SSDs with RAID support, or up to three (3) HDDs for large capacity storage3, up to 2TB4 of DDR5 memory with octa-channel support and seven (7) PCIe slots including six (6) PCI Gen5 that deliver higher speed connectivity, lower latency and more expansion capability, and includes 10 Gigabit Ethernet onboard to virtually eliminate network bottlenecks.

ThinkStation P8 includes built-in hardware monitoring via ThinkStation Diagnostics software and the Lenovo Performance Tuner software that ships with multiple profiles to optimize ISV applications.

Availability

ThinkStation P8 powered by AMD Ryzen Threadripper PRO WX-Series Processors and NVIDIA RTX graphics cards will be available starting Q1 20245. To learn more about the newest ThinkStation P8, download the datasheet, and for more information about Lenovo’s workstation portfolio, visit www.lenovo.com/workstations.

Lenovo workstation solutions are also on display at Autodesk® University, November 13-15, 2023, and can be found at booth #434.

The post Lenovo Unleashes Performance with New ThinkStation P8 appeared first on Architosh.

The new workstations shipping from major workstation makers will feature next-gen AMD Ryzen Threadripper PRO processors paired with NVIDIA RTX Ada Generation GPUs.

Ryzen Threadripper PRO

AMD has new Ryzen Threaderipper PRO 7000 WX-Series CPUs delivering significant performance improvements across the board for the demands of present and future workloads. The new CPUs deliver up to 95 cores (192 threads) per CPU and industry-leading maximum memory bandwidth in a single socket CPU.

These much-anticipated CPUs delivered unparalleled multi-core performance from their Zen 4 Architecture, and they include enterprise-level security and modern management features. In addition, AMD is also introducing the AMD Ryzen Threadripper 7000 Series processors with the 7980X CPU, offering the most computing power possible in an enthusiast desktop computer with exceptional multithreaded performance.

The latest AMD Ryzen Threadripper PRO CPU paired with NVIDIA Ada RTX cards offer the ultimate workstations for AI, DCC, CAD and scientific computing.

“AMD has a long history of leading innovation in the workstation and high-performance computing segments, and this new line up of Ryzen Threadripper 7000 Series processors marks another step on that path,” said Jack Huynh, senior vice president and GM, Computing and Graphics Business Group at AMD. “These processors are a testament to our unwavering commitment to delivering world-class innovation and performance, and we’re confident that they will unlock incredible creative potential in our users while continuing to raise the bar on energy efficiency in the most capable platform on the planet.”

With the new Ryzen Threadripper PRO processors with up to 96 cores, users can see up to 2x faster performance in rendering with tools like Chaos’ V-Ray with similar boosts in performance across a range of simulation, generative design, and software compilation tasks.

Training AI Models

The new NVIDIA RTX GPU and AMD CPU-powered workstations make industry-leading workstations for training AI models, local fine-tuning, and offloading data center and cloud resources for AI development tasks.

With up to 48GB of memory in a single RTX card, these latest workstations offer an effective way for organizations to lessen the load and costs of data center computing by offloading AI training tasks.

Ada Generation GPUs include the RTX 4000 SFF, RTX 4000, RTX 4500, RTX 5000, and RTX 6000. Built on the NVIDIA Ada Lovelace architecture, they can feature up to 142 third-generation RT cores, 568 fourth-generation Tensor Cores, and 18,176 latest-generation CUDA cores.

From architecture and manufacturing to media and entertainment and healthcare, professionals across industries will be able to use the new workstations to tackle challenging AI computing workloads — along with 3D rendering, product visualization, simulation, and scientific computing tasks.

Learning More

Readers can learn more here: https://blogs.nvidia.com/blog/2023/10/19/ai-workstations/

The post AMD and Nvidia Partner on New Workstations Tech appeared first on Architosh.

Ada Generation GPUs

“Few workloads are as challenging as generative AI and digitalization applications, which require a full-stack approach to computing,” said Bob Pette, vice president of professional visualization at NVIDIA. “Professionals can now tackle these on a desktop with the latest NVIDIA-powered RTX workstations, enabling them to build vast, digitalized worlds in the new age of generative AI.”

The new NVIDIA RTX 5000 Ada generation GPU delivers accelerated workflows for digital content creation (DCC) and CAD professionals, especially those using AI-powered applications.

The very latest RTX workstations now can offer up to four NVIDIA RTX 6000 Ada GPUs, each fully equipped with 48GB of memory. A single workstation can deliver 5,828 TFLOPS of computing power for demanding AI workloads and up to 192GB of GPU memory. Here’s a look at the three new Ada additions in more detail:

NVIDIA RTX 5000

This GPU is 2.1x faster than the predecessor NVIDIA RTX A5500 GPU at rendering, 3.3x faster at Omniverse, and 1.9x faster at graphics overall.

• 32GB of GDDR6 memory
• 576GB/s memory bandwidth
• 12,800 CUDA Cores
• 400 Tensor Cores (4th generation)
• 100 RT (ray tracing) Cores
• 65.3 TFLOPS Single-Precision Performance
• 4x DisplayPort 1.4
• Dual Slot Form Factor

NVIDIA RTX 4500

This GPU is 1.5x faster than the predecessor NVIDIA RTX A4500 GPU at rendering, 2.7x faster at Omniverse, 1.6x faster at graphics, and 1.5 faster at CAD.

• 24GB of GDDR6 memory
• 432GB/s memory bandwidth
• 7,680 CUDA Cores
• 240 Tensor Cores (4th generation)
• 60 RT (ray tracing) Cores
• 39.6 TFLOPS Single-Precision Performance
• 4x DisplayPort 1.4
• Dual Slot Form Factor

NVIDIA RTX 4000

This GPU is 1.7x faster than the predecessor NVIDIA RTX A4000 GPU at rendering and 1.5x faster at graphics in general.

• 20GB GDDR6 memory
• 360GB/s memory bandwidth
• 6,144 CUDA Cores
• 192 Tensor Cores
• 48 RT (ray tracing) Cores
• 26.7 TFLOPS Single-Precision Performance
• 4x DisplayPort 1.4
• Single Slot Form Factor

All three new Ada generation RTX GPUs are VR-Ready, and all three do not support NVIDIA NVLink. Graphics APIs supported include DirectX 12, Shader Model 6.7, OpenGL 4.6, and Vulkan 1.3. Compute APIs include CUDA 12.2, OpenCL 3.0, and DirectCompute.

Click on the following links to learn more. NVIDIA RTX 5000, RTX 4500, and RTX 4000.

To read our last RTX GPU product review, go here.

The post NVIDIA’s New RTX Ada Generation GPUs for Workstations Pack Power appeared first on Architosh.

Workstation GPU’s

AMD has designed its new Radeon graphics cards to handle mainstream professional workloads across a wide range of industries and businesses. These include architecture, construction, design, engineering, entertainment, manufacturing, and media.

AMD Radeon PRO 7000 series graphics cards. Screenshot from AMD promotional video.

Senior vice president and general manager of AMD’s Graphics Business Unit, Scott Herkelman outlines their capabilities:

The new Radeon graphics cards also feature breakthrough AMD RDNA 3 technology. As a result, they are optimized to deliver exceptional performance, outstanding stability, and remarkable efficiency. They also feature 8GB of high-speed GDDR6 memory to support data-intensive tasks and enable ray-traced renderings with incredible detail and realism.

Key features

• AMD Radiance Display Engine with DisplayPort 2.1. With 12-bit HDR color support and over 68 billion colors, it supports next-generation displays and multi-monitor configuration options. This also creates ultra-immersive visual environments.
• AMD RDNA 3 architecture. This features redesigned compute units, unifying ray-tracing, and AI accelerators. Also includes second-generation AMD Infinity Cache technology and second-generation ray-tracing technology. It additionally offers optimizations for AEC, D&M, and M&E workflows for 3D modeling, animation, rendering, video editing, and general multitasking.
• Dedicated AI acceleration. New artificial intelligence instructions and increased AI throughput deliver over twice the performance of the previous generation RDNA 2 architecture.
• Optimized driver performance and application certification. AMD continues to work with leading professional software application vendors on a comprehensive application certification program. This is to ensure that AMD Radeon PRO graphics cards can survive demanding 24/7 environments and meet the required technical standards.

Pricing and Availability

The AMD Radeon PRO W7600 and AMD Radeon PRO W7500 will be available in the third quarter of 2023. They will retail for around USD 599 and USD 429, respectively. These new Radeon graphics cards are also expected to appear in OEM workstations and SI systems later this year. Please click here for further details.

About Advanced Micro Devices

AMD has driven innovation in high-performance computing, graphics, and visualization technologies for more than half a century. As a result, billions of people, leading Fortune 500 businesses, and cutting-edge scientific research institutions around the world use AMD technology. AMD employees are focused on building leadership, high-performance and adaptive products that push the boundaries of what is possible. For more information about AMD, please click here.

The post AMD Intros two new high-performance Radeon PRO graphics cards appeared first on Architosh.

The new computer will be available with the new NVIDIA L40S—a new universal GPU for data centers based on the NVIDIA Ada Lovelace architecture.

RAXX P1G

The RAXX P1G platform is purpose-built to power generative AI, as well as media and entertainment workflows.

“Demanding AI workloads require powerful, specific, purpose-built solutions,” said Tim Lawrence, BOXX VP of Engineering. “With RAXX P1G, organizations can harness the performance of NVIDIA L40S GPUs for large language model inference and training, as well as sophisticated graphics and video applications.”

The new BOXX Technologies RAXX P1G rack-mounted server is ideal for AI and M&E workflows.

The new computer features fourth-generation Tensor Cores and the NVIDIA Transformer Engine to accelerate AI training and inference workloads while offering support for FP8. This enables state-of-the-art performance for Al-enabled graphics, video, and image generation. Additionally, third-generation NVIDIA RTX and DLSS 3 technology allow for high-fidelity creative workflows with 48GB of GPU memory available.

Powering Omniverse

The new RAXX P1G can deliver the Omniverse with excellent accelerated performance, delivering raytraced and path-traced rendering, physically accurate simulation, and photorealistic 3D synthetic data generation. The GPU also utilizes the FP32 for scientific data analysis and simulation in life sciences, physics, higher education, and financial services.

BOXX RAXX P1G, ideal for deploying AI solutions on a large scale, is available with up to four NVIDIA L40S GPUs and is powered by a single AMD EPYC™ 7003 processor (up to 64 cores). The 1U rack-mounted system also features advanced air cooling, 2TB of memory, and rear I/O access to two 1 GbE LAN ports, one IPMI port, and two USB 3.0 ports.

Built for M&E Workflows

With its 64-core AMD EPYC 7003 processor chip, the RAXX P1G is built for more than just AI applications, it can handle extreme multitasking in feature film editing, VR deployment and power M&E applications like DaVinci Resolve, Cinema 4D, Autodesk Maya, as well as rendering solutions like Blender, Redshift, V-Ray, Arnold, and Octane Render.

“The evolution of AI is requiring organizations to deploy compute resources on a larger scale,” said Bob Pette, vice president of professional visualization at NVIDIA. “As the premier GPU for multi-modal generative AI, the NVIDIA L40S inside the BOXX RAXX P1G will deliver end-to-end acceleration for inference, training, graphics, and video workflows to power the next generation of AI-enabled applications.”

For further information and pricing on RAXX P1G, contact a BOXX sales consultant in the US at 1-877-877-2699. Learn more about BOXX systems, finance options, and how to contact worldwide resellers, by visiting www.boxx.com. 

The post BOXX Upgrades AI Workstation Module with new NVIDIA L40S GPUs appeared first on Architosh.