It is also hard to know how much effort Apple has or will put into their implementation of OpenGL on M1. But the same statement is true of Intel’s Integrated Graphics, and we all know how poorly they perform for SketchUp. It is coupled tightly to the same memory as the CPU, so the need to copy data back and forth between external GPU memory and main memory is eliminated. Apple’s GPU is a unique proprietary design. But benchmarks published so far suggest the benefit isn’t enormous. Also, the Pro and Max have wider paths to memory, which may have an impact on performance. SketchUp is not a memory hog, but having more memory lets you run more things simultaneously before they start competing with each other. The Pro and Max come with different options for memory than the base M1, notably larger amounts and wider access paths. However, renderers should be able to make use of the additional cores, and that should make the Pro or Max faster than the base M1. As has been explained innumerable times on the forum, there is no possibility that SketchUp can be rewritten to be multi-core. Since SketchUp uses only one core, there will be no real difference due to CPU variations between these chips. What differs is the number of cores and the allocation between performance and power-saving versions. The CPU cores in the M1 Pro and Max are the same as in the base M1. So, although we can assume that native code should be faster, there is not much beyond anecdotal to establish how much faster - especially for a specific app such as SketchUp. There are not a lot of apps that have yet been ported to M1 native, instead of running via Rosetta 2. That is of great importance for laptops, perhaps less so for desktops where better cooling is feasible. But there are some general observations we can make.Īll of the Apple Silicon variations use less power and stay cooler than the competition. We will need to wait for real world experience from the early adopters to know better how SketchUp runs. Because Apple silicon puts so many functions on a single chip and ties them all closely together, it is difficult to predict comparative performance based just on specs and non-specific benchmarks.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |