ghitchens

Members
  • Posts

    2
  • Joined

  • Last visited

Reputation

2 Neutral

1 Follower

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. @Ryan-M Thanks so much! A very cogent, detailed, and helpful answer. I wondered if there might be other libraries besides Qt that were bottlenecks. I'm really glad to hear it's continuing to be worked through. The lack of hardware-ray-trace support even in the M1 MAX GPU is a good point, and I hadn't considered that feature was so much of what people are asking about when they ask for M1 support. My experience is that the performance of Chief Architect is quite reasonable under Rosetta 2 on M1 MacBook Air, and with no fan! I expect the M1 Pro/Max to be even better, and once a native arm64 build arrives, it bet it will an exceptionally responsive platform for general use of Chief Architect, even if hardware-ray-tracing won't be supported.
  2. Thanks, @scottharris and others at Chief Architect for a great product and for supporting the Mac! Mac support is a primary reason I use Chief, since I use a Mac for other design and professional work as well. I wanted to provide a bit more detail on Apple Silicon and why I suspect Chief Architect has not yet supported it with a native arm64 (Apple Silicon) version to date. For my credentials, I have spent my career in tech industry, have worked extensively on software tools, and am very familiar with CPUs, GPUs, architectures, and the strengths and drawbacks of the Apple platform for various purposes (my current company does development work and Mac and PCs are some of our target devices). If a member of the development staff at Chief Architect has more detailed information or a clarification on the issues below, that would be awesome, but since no explanation has yet been given, and no clear roadmap discussed, I have had to infer what might be happening based on my knowledge and experience. Regarding GPU performance, I don't think "integrated" vs "discrete" graphics paints an accurate picture of the tradeoffs involved when discussing the M1 and new M1 Pro/Max chips. Overall, those chips deliver true "dedicated-gpu" class performance, and in most cases beat common dedicated GPUs found laptops non-specialized desktops. It isn't fair to say that performance of these GPUs suffers because they are "integrated". Apple's previous "integrated" M1 GPU (released last year in the Mini, Air, and 13" MacBook Pro) delivers GPU performance far beyond other (Intel, AMD) integrated GPUs, and generally matches the dedicated radeon 560 GPU in 3dmark performance. @scottharris's macbook pro mentioned above, uses the Radeon 555x, which is lower spec than the 560, so it is likely my macbook air with M1 has higher GPU performance than his 15" laptop. You can see some independent benchmarks here so that you know I'm not making this up: https://www.anandtech.com/show/16252/mac-mini-apple-m1-tested/3 Apple just announced new MacBook Pro notebooks with the M1 Pro and M1 MAX, with up to 32 core "integrated" GPU. Note that some people are estimating these GPUs to be competitive with the Nvidia RTX 2070/2080 series GPUs https://appleinsider.com/articles/21/10/19/m1-pro-and-m1-max-gpu-performance-versus-nvidia-and-amd. This certainly implies that they are at or near the top of the charts of laptop GPU performance, even when compared with expensive dedicated "gaming" laptops. Chief Architect is still built for amd64 (AMD/Intel) architecture only and must use Apple's Rosetta2 layer for translation, resulting in performance degradation compared to a "native" arm64 (Apple Silicon) build. Regardless, I use it on an M1 MacBook Air with external 5k monitor, and performance is acceptable, although it is clearly suffering from lack of a native version. It still is often reasonably competitive with my MacBook Pro Core i9 with dedicated Radeon Pro 5500 GPU, and without fan noise on the M1. Given that the performance of Apple's latest Apple Silicon with "integrated" GPU likely is a strong reason for Chief Architect to support Apple Silicon natively, and that most Mac apps have already done so (it's just a recompile in Xcode for many apps), I think the question remains as to why a native version has not yet been released. I have a couple suspicions, but the most likely one is this: Chief Architect uses Qt for a UI toolkit. Qt did not support arm64 / Apple Silicon until Qt 6.2 released late September. Given that the underlying UI library that Chief is built on didn't support Apple Silicon until now, I expect that might be one of the things that has held up a native apple silicon version of Chief Architect so far. https://www.qt.io/blog/qt-6.2-lts-released Now that there is fully released Apple Silicon support in the underlying UI toolkit, and the kind of GPU performance in the new MacBook Pros that puts many "dedicated GPU" PCs to shame, I suspect (hope) that someone at Chief Architect is putting together a release of a native arm64 (Apple Silicon) version that takes full advantage of this amazing hardware. Again, if someone at Chief Architect wants to comment on the roadmap, or explain where the above is wrong, it would be awesome. Also, if you're looking for a tester for a pre-release version, I've got a few versions of the M1 (including the M1 Max 32 core arriving in a week or so) here to test on, and would love to contribute, since I have other configs here, including windows/RTX 3080 to compare to.