• Starter Course: The Top Five Must-Reads
• Special Feature: The M1 Mac mini vs iMac Pro vs Everyone
• emTech: Emerging Technologies -- Apple's USD 3,000 rumored VR headset; Tech Soft 3D comments on Apple Silicon and Metal and more; BIM news for Revit. 
• The Briefing: Biggest CAD Industry News Last Month

The Top Five Must-Reads

The M1 Mac mini vs iMac Pro vs Everyone

IT WASN'T THAT LONG AGO THAT AMD stole Intel's single-core performance crown with its Zen 3 chip architecture in the new Ryzen 5000 series CPUs. Now, less than a year later, Apple has come out with a chip (M1) that delivers a stunning performance of its own, and at least by one benchmark is the new reigning single-core performance champion. 

But just how powerful is the M1 as compared to rival hardware options out there? And how well might this chip perform in CAD and 3D application environments? These were the questions I sought to answer ever since Architosh obtained a new (late 2020) Mac mini with an M1 processor. 

I already knew Apple Silicon was a possible game-changer. And I had heard stunning praise from CAD industry software developers. But could a sub-1000 dollar machine really be a serviceable CAD/3D machine for professionals? After many benchmarks and real-world app tests (under Rosetta 2.0 no less), the short answer is "yes!"   

Introduction: Tests and Machines

To back that "yes" up, I want to note that single-core performance is of the highest importance in the CAD world, particularly in AEC. 

That is why the workstation folks over at BOXX aim architecture and engineering customers to high-frequency hardware solutions. For example, when speaking of Rhino, they say, "because modeling is its primary function, Rhino is a frequency-bound application, a key factor to consider when determining your Rhino 3D hardware requirements."




"FP performance is important to both 2D and 3D in Vectorworks. It should help out in basic operations such as transformation, tessellation, projection of geometric entities in Vectorworks, as well as inside the Parasolid modeling kernel."  --- Dr. Biplab Sarkar, CEO, Vectorworks, Inc. 




This is not to disregard multi-core processing. The M1 delivers very respectable multi-core performance, as we will see below. However, its single-core performance—thanks to Apple's Firestorm cores—is quite astounding, besting AMD's Ryzen 5000 series top-flight chip on Geekbench 5 scores. 

The next most important thing to focus on is the M1's GPU performance because today's BIM solutions are more dovetailed with leading photo-realistic and interactive rendering programs. I wanted to focus on two aspects to see if the M1 could meet the minimum hardware requirements for Enscape (if that program ever comes to Mac) and meet the minimum specifications for standard VR headsets. 

This benchmark focuses on single-core and multi-core CPU performance, using a CPU-bound renderer (Cinema 4D technology). The M1 Mac mini easily defeated my three-year-old iMac Pro in single-core tests, but the 2017 iMac Pro with an 8-core Xeon chip was approximately 26 percent faster at multi-core rendering. However, on single-core, the M1 was dramatically faster. (see image above and below.)

Recently, I have come across some information from Nuvia, a young semiconductor design company headed by Gerald Williams, Apple's former Chief CPU Architect, for nearly a decade. Williams led Apple's A-series chip architecture designs. When he founded Nuvia in 2019, he took two other senior Apple chip design leaders with him, in Manu Gulati and John Bruno. I mention this not because I think Apple will suffer without Williams—not at all—but rather because these guys really know what they are doing. And they are big believers in Geekbench 5. 

More than that, they have mathematically determined how to calculate very accurate approximations of SPEC CPU2006 and 2017 scores vis-a-vis Geekbench 5 scores. This an activity for a future article. 

The Nuvia folks have written that Geekbench 5 scores are a good benchmark not only across platforms but also valid for server CPU load tests of a "comprehensive general nature." In a nutshell, then, we should put excellent stock in the following Geekbench 5 numbers on the M1. 

Again, we can see the M1 excels at single-core CPU performance, with a Geekbench 5 score 59 percent faster than the iMac Pro's Xeon chip and 38 percent faster than today's fastest single-core Intel Mac, the 27-inch mid-2020 iMac Retina, which uses the top 15th ranked single-core chip. 

As a point of reference, Geekbench Browser lists top chip scores (reflecting averages of at least five unique results) with the AMD Ryzen 9 (5950X) 16-core chip obtaining an average score of 1685, slightly lower than the average of my M1 Mac test results. However, an iMacPro1.1 system is listed on the site with a score of 2278, using the same chip. We assume that Apple has built test units using AMD processors, but we can't be sure. 

This respected test-suite is M1 native, giving us the ability to test M1's graphics performance. We ran tests that could work on both Metal on the M1 and Metal on the iMac Pro. We looked at Driver Overhead, Texturing (fill rate), and ran the modern-game like Aztec Ruins test. In these tests, we offer various scores along with the two Macs and will explain why below. 
 

GFXBench Aztec Ruins 1080 (normal tier). 
 

This benchmark tests a modern graphics API across a game-like environment with real-time global illumination and deferred rendering combined with physically-based shaders. This test is highly applicable to not just games but "game-engine" based rendering tools. Some game engines support both forward rendering and deferred rendering paths, like Unity, which has partnered with Autodesk on many fronts, including its Revit application. 

GFXBench Aztec Ruins, 1080p Test. Winner: M1 Mac mini defeats iMac Pro 2017.  (Image: Architosh) 

We can see above that today's M1 is about 30 percent faster than the discreet AMD Radeon Pro Vega 56 graphics in the 2017 iMac Pro at this particular test. However, a top reference score from a machine with the NVIDIA GeForce GTX 1080 Ti is nearly 3x faster. That result had a DirectX 11 driver. The M1 faired much better against the same card tested with OpenGL. In that case, the discreet GPU's performance fell to 1.8x faster. Interestingly, M1 is faster than an NVIDIA GeForce GTX 1050, approximately 1.5x times more rapid than this Pascal architecture-based discreet GPU from the fall of 2016. 

We can also see in the image below that the M1's texel fill rate is similar to what the GeForce GTX 680 and AMD R9 280 can do. Both are discreet GPUs that exceed the lesser performing sibling GPUs specified by Enscape as "minimum requirements" for powering that professional real-time renderer used in the AEC industry. The M1's texel fill rate also exceeds that of the GPU performance in the NVIDIA Quadro P2000 (considered a baseline GPU for powering HTC Vive VR headsets). It also exceeds that of AMD Radeon Pro WX 9100 and the AMD D700 used in the 2013 Mac Pro. 

BaseMark GPU

This GPU benchmark is built on top of the Rocksolid proprietary rendering engine coded in C++. The Helsinki-based software company touts the objectivity of this benchmark for multiple operating systems. Rocksolid Engine architecture abstracts resources and rendering, effectively neutralizing the comparison as all tests run with the same workload regardless of the operating system. 

As noted here at Anandtech.com, Apple's CPUs in mobile have dominated in browser-benchmarks. Now having that ARM-based architecture on the Mac platform brings superior web application performance to Apple Silicon-based Macs. This is important because more professional applications happen via browser-based SaaS tools. (see emTech section below and comments by Tech Soft 3D)

As we can see, the M1 Mac mini has blistering web performance, more than 2x the 2017 iMac Pro performance. The M1 is also 55 percent faster than the AMD Ryzen 9-5950X (based on an Anandtech.com score). Apple has highly prioritized web browser performance, as can be seen in even the iPhone 12 Pro (A14 chip), obtaining a 202.2 score (based on Anandtech.com score). As a further point of reference, my Touch-bar, 2016 MacBook Pro with 2.6GHz Intel i7 received a score of 85.8, about 22 percent slower than the iMac Pro.  

Real-World App Tests

Typically Architosh does real-world performance tests with a series of files we have accumulated or built over a decade and a half. These files are used during our product reviews to look at performance from version to version. 

We test using the Timer app (a digital stopwatch with 1/10 a second indicator) and the Apple Developers tool, Quartz Debug, which includes a Frames Per Second (FPS) meter. And we record the whole thing in QuickTime. After clean restarts, only these tiny apps plus the "test app" are running. The apps are loaded into the system in the exact same order: (1) test app, (2) QuickTime, (3) Timer, and finally (4) Quartz Debug.

Our test files are a series of SketchUp and Vectorworks Architect files, two programs we have most been asked (and performed) product reviews for. As a result, we have built up some test files.

Importantly, both of these tools were running under Rosetta 2 because they have not been re-coded for Apple Silicon Macs yet. 

SketchUp Impressions. 

In our SketchUp 2021 tests, the M1 ran reasonably well under Rosetta, but it was outperformed by the native SketchUp for Intel on the iMac Pro 2017. 

In the top row in the image below, our Turning Torso tower model (a quite complex but repeating geometry) runs through four scenes and turns. We calculate achieved Frame Rates per Second using Apple's Quartz Debug app. The iMac Pro achieved 70+ FPS in the first scene transition, compared to the M1 Mac mini, which only reached 40 FPS.  (M1 on upper left, iMac Pro on upper right). 

On the Falkestrasse model, an odd but well-published creation by the architect Coop Himmelblau, the M1 Mac mini achieved greater than 50 FPS scores during its scene transitions. But again, the iMac Pro with its Pro Vega 56 GPU card achieved greater than 80 FPS. (M1 on lower right and iMac Pro on left). 

Both models are simple OpenGL shaded models with transparency (glass), with the Falkestrasse containing some limited texture mapping. We noticed that when the scene passes over the texture-mapping section of the Falkestrasse facade on both machines, the FPS speeds up noticeably.

In our third SketchUp model with intensive photographic landscape props, the M1 Mac mini performed exceptionally well and was on par with the iMac Pro. Overall, we think Rosetta 2 is impacting a penalty on the performance, but even under Rosetta 2, SketchUp on the M1-based Mac was more than serviceable—it was quite good. 

Vectorworks Architect 2021

We have three files for Vectorworks testing. One area that seems to tax Vectorworks performance in regular 2D work is when a file utilizes many external references on a "design layer" in viewports on that layer. The application is incredibly flexible in the creation and use of viewports, internally and externally referenced. That's a good thing. However, many of them on a visible layer (including iterations of the same viewport) cause a noticeable impact on performance. 

In our test, we save different scenes with different combinations of viewports visible. Going from scene-to-scene and using the pan tool to move the screen around represents real-life user-interaction scenarios where screen drawing is taxed. 

The popularity of real-time ray-tracing is surging in AEC these days, and some leading players are not on the Mac. One of those is Enscape. When I first met this company at a trade show a few years ago, they were not very impressed with the Mac platform hardware-wise. Real-time rendering takes powerful GPUs and at that time Apple was lacking them. 

It is notable than that in several GFXBench GPU tests, the M1 out-performed bigger sibling dGPU cards that are today listed as "minimum required" GPUs for Enscape to function. We think texture fill-rate is likely a critical piece of Enscape's requirements. Therefore, it is noteworthy to point out that the M1 exceeded the baseline GeForce GTX 660 and AMD R9 270 by exceeding or matching their bigger sibling GPUs. This was also the case in the Driver Overhead test. 

Curated content: Emerging Technologies and their potential impact on CAD-based industries.

Apple Entering the VR-AR Headset Market – The New!

A lot of folks have been waiting for this news for a long time. You see, Apple's Macs have poor VR headset support, and it's just unacceptable. 

Dan Riccio Stepped Down to Focus on AR/VR

Over at Ars Technica, they are covering a story today about Apple's hardware chief stepping down to take-over full-time leadership on Apple's long-rumored VR/AR headset. 

Biggest CAD Industry News Last Month