Future Macs could adopt Intel's new, high-performance discrete graphics chips
Intel has confirmed it intends to create its own discrete graphics processing chips by 2020, a move that opens up the possibility of Apple using discrete Intel GPUs across its laptop and desktop Mac lines.
Initially revealed by CEO Brian Krzanich during an analyst event in early June, Intel plans to offer its discrete GPU in just a few years, reports MarketWatch. An official Intel Twitter account confirmed the news, first by noting the 2020 target date, then by retweeting the story.
Intel's intention is to provide the discrete GPUs to enterprise and consumer markets. For enterprise, Intel wants to provide its discrete GPUs for use in data centers, powering machine learning and AI in a similar way existing GPU technology from AMD and Nvidia are used.
In the consumer market, Intel will be directly competing against Nvidia and AMD, possibly by offering graphics cards for desktop computers or supplying notebook producers with GPUs. Intel does already provide integrated graphics, which is a feature of its processor lines, but a discrete card typically provides more performance than the integrated GPU, making Intel's current graphics offering appear less desirable.
Intel has seemingly already acknowledged the lower performance of its own integrated graphics, after revealing G-series processors in January that combined Intel CPUs with an onboard AMD GPU, providing the equivalent of discrete graphics performance on the same board as the processor.
An Intel G-Series processor, combining an Intel Core processor with an AMD Radeon RX Vega M GPU
Due to the significant effort required to create a new GPU architecture that can compete with AMD and Nvidia, Intel is likely to have worked on the project for some time already if it is to meet its 2020 target. In November 2017, Intel hired Raja Koduri, formerly the head of AMD's graphics arm and credited with improving the Radeon brand, to head up its graphics and compute projects.
While Nvidia and AMD's dominance in the GPU market over the last two decades may be hard for Intel to crack, the processor producer could be in a good position to create discrete GPUs that work better with its own CPUs than its rivals. This synergy could make an Intel CPU and GPU combination an attractive prospect for notebook producers, including Apple.
Current 15-inch MacBook Pro models include both Intel-based integrated graphics and AMD Radeon discrete graphics, which are used depending on the application's performance needs. The introduction of a discrete Intel GPU could give Apple more options for what to include in a future MacBook Pro, especially if there is an extra benefit in the Intel GPU and CPU combination other than performance.
For other Mac products, Intel's integrated graphics are offered alongside AMD Radeon discrete graphics, such as the integrated Intel Iris Plus Graphics 640 in the entry-level iMac against the discrete AMD Radeon Pro GPUs in the higher models, while the iMac Pro offers AMD's Vega GPUs. If acceptable to Apple, there is the prospect of Intel offered for both integrated and discrete graphics across the board.
Intel's attempt to join the discrete GPU market could also apply pressure on AMD and Nvidia, especially considering Intel's size and experience. The sudden appearance of a viable competitor may force the incumbent industry leaders to make bigger moves forward in performance, if only to make it harder for the relative newcomer to find a consumer audience.
Initially revealed by CEO Brian Krzanich during an analyst event in early June, Intel plans to offer its discrete GPU in just a few years, reports MarketWatch. An official Intel Twitter account confirmed the news, first by noting the 2020 target date, then by retweeting the story.
Intel's intention is to provide the discrete GPUs to enterprise and consumer markets. For enterprise, Intel wants to provide its discrete GPUs for use in data centers, powering machine learning and AI in a similar way existing GPU technology from AMD and Nvidia are used.
In the consumer market, Intel will be directly competing against Nvidia and AMD, possibly by offering graphics cards for desktop computers or supplying notebook producers with GPUs. Intel does already provide integrated graphics, which is a feature of its processor lines, but a discrete card typically provides more performance than the integrated GPU, making Intel's current graphics offering appear less desirable.
Intel has seemingly already acknowledged the lower performance of its own integrated graphics, after revealing G-series processors in January that combined Intel CPUs with an onboard AMD GPU, providing the equivalent of discrete graphics performance on the same board as the processor.
An Intel G-Series processor, combining an Intel Core processor with an AMD Radeon RX Vega M GPU
Due to the significant effort required to create a new GPU architecture that can compete with AMD and Nvidia, Intel is likely to have worked on the project for some time already if it is to meet its 2020 target. In November 2017, Intel hired Raja Koduri, formerly the head of AMD's graphics arm and credited with improving the Radeon brand, to head up its graphics and compute projects.
While Nvidia and AMD's dominance in the GPU market over the last two decades may be hard for Intel to crack, the processor producer could be in a good position to create discrete GPUs that work better with its own CPUs than its rivals. This synergy could make an Intel CPU and GPU combination an attractive prospect for notebook producers, including Apple.
Current 15-inch MacBook Pro models include both Intel-based integrated graphics and AMD Radeon discrete graphics, which are used depending on the application's performance needs. The introduction of a discrete Intel GPU could give Apple more options for what to include in a future MacBook Pro, especially if there is an extra benefit in the Intel GPU and CPU combination other than performance.
For other Mac products, Intel's integrated graphics are offered alongside AMD Radeon discrete graphics, such as the integrated Intel Iris Plus Graphics 640 in the entry-level iMac against the discrete AMD Radeon Pro GPUs in the higher models, while the iMac Pro offers AMD's Vega GPUs. If acceptable to Apple, there is the prospect of Intel offered for both integrated and discrete graphics across the board.
Intel's attempt to join the discrete GPU market could also apply pressure on AMD and Nvidia, especially considering Intel's size and experience. The sudden appearance of a viable competitor may force the incumbent industry leaders to make bigger moves forward in performance, if only to make it harder for the relative newcomer to find a consumer audience.
Comments
That being said, it would be interesting to see that if Apple releases a Mac mini and/or MacBook Air with their own CPU, if it also has their own GPU inside it as well.
As far as I'm aware they've only used their internally designed GPU in one A-series chip, tjhe A11 Bionic. I don't think its in any older chips and if it's in other chips, like T-series for the Touch Bar display or S-sereis for Apple Watch display then I'm not aware of it.
I can't see Apple using an Intel discrete GPU, for many reasons:
• Intel has a long history of designing seriously underperforming integrated GPUs;
• Intel hasn't designed discrete GPUs for 15 years, and they weren't competitive back then;
• Apple has supposedly been actively trying to reduce its reliance on Intel parts;
• Apple likes to develop hardware that works hand-in-glove with its software. The Apple GPU in the A11 Bionic is purpose-built for Metal, Apple's graphics API. With Apple depreciating OpenGL in macOS Mojave in favor of Metal, it's more likely Apple would design its own built-for-Metal GPU than use Intel's built-for-DirectX GPU;
• If Apple indeed moves to an ARM-based Mac using the A-series processor, of course it's going to use the Apple GPU that's now part of the A-series chip.
The IBM-built system is comprised of 4,608 nodes, each one housing two Power9 CPUs and six NVIDIA Tesla V100 GPUs. The nodes are hooked together with a Mellanox dual-rail EDR InfiniBand network, delivering 200 Gbps to each server.
Thus means 24k Nvidia GPUs in one computer. This is small change compare to putting them in normal computers but most supercomputers are using GPUs instead of CPUs so maybe Intel is pushing to play with the big boys. Don’t see it happening with their current GPU design, however.
That said, I think Apple’s GPU/APU focus would be on mobile/portable Metal and image/video transcoding rather than compute numbers and gaming frame-rates.
Apple’s family 4 GPUs are currently delivering around 110Gflops and a Geekbench compute score of 5K per core. The PowerVR architecture it’s based on is limited to 16 cores (memory bandwidth?) and GPU performance is highly scalable so even a cautious upgrade could see current tech matching RX Vega M. Upgrade the memory architecture (HBM2), increase the power budget and double/quadruple core count and you’re looking at an GTX/Vega killer.
Apple discreet graphics+T-series for MBP 15” business case would be; no additional power or cost budget, release control, metal optimisation, strengthen ARM cores for iOS ‘simulator’ + OS core/1st party app execution, precursor to ARM transition.
The question should be; why wouldn’t they go this way?