Apple M3 Ultra
Apple (Hacker News, MacRumors):
M3 Ultra is built using Apple’s innovative UltraFusion packaging architecture, which links two M3 Max dies over 10,000 high-speed connections that offer low latency and high bandwidth. This allows the system to treat the combined dies as a single, unified chip for massive performance while maintaining Apple’s industry-leading power efficiency. UltraFusion brings together a total of 184 billion transistors to take the industry-leading capabilities of the new Mac Studio to new heights.
[…]
It features up to a 32-core CPU with 24 performance cores and eight efficiency cores, delivering up to 1.5x the performance of M2 Ultra, and up to 1.8x that of M1 Ultra. It also has the largest GPU in any Apple chip, with up to 80 graphics cores that bring up to 2x faster performance than M2 Ultra, and up to 2.6x faster than M1 Ultra.
[…]
The unified memory architecture of M3 Ultra integrates the most high-bandwidth, low-latency memory ever available in a personal computer. Starting at 96GB, it can be configured up to 512GB, or over half a terabyte.
Previously:
Update (2025-03-14): Joe Rossignol:
Apple said the M3 Ultra chip is the “highest-performing chip it has ever created,” and the unverified benchmark result seems to confirm that. In the single result, the 32-core M3 Ultra chip achieved a multi-core CPU score of 27,749, which makes it around 8% faster than the 16-core M4 Max chip that previously held the performance record. The result also reveals that the M3 Ultra chip is up to 30% faster than the 24-core M2 Ultra chip.
As expected, the M4 Max chip tops the M3 Ultra chip in terms of single-core CPU performance by nearly 20%, according to the result.
In one Geekbench 6 result for the new Mac Studio, the M3 Ultra with an 80-core GPU achieved a Metal score of 259,668, up from 222,582 for the M2 Ultra chip with a 76-core GPU in the previous-generation Mac Studio. If that single result is accurate, then the M3 Ultra offers up to 16% faster graphics performance than the M2 Ultra.
The M3 Ultra chip is very much modeled on the M3 Max and offers twice as many CPU and GPU cores, Neural Engine cores, and the like. But look closer: The M3 Ultra offers a maximum of 512GB of RAM, four times the maximum 128GB of the M3 Max. And the M3 Ultra supports the faster Thunderbolt 5 specification, not the older Thunderbolt 4 of the rest of the family.
That means the M3 Ultra is an in-between chip, mostly based on the older M3 generation, but with a few high-end additions that push it up toward the M4 in terms of capability.
[…]
Still, the M3 Ultra’s arrival is weird enough to make it feel like we’re all missing part of the bigger picture. If there’s no other shoe to drop, why did the M2 Ultra-based Mac Pro not get its own M3 Ultra chip bump?
Update (2025-03-17): treblewoe (Threadreader):
When challenged why on earth in 2025, Apple is shipping an M3 Ultra when the M4 has been out for ten months, they spun, “not every chip generation will have an Ultra config.” While technically true due to Tim’s greed, it is conscious deception on their part.
Here’s what really happened. Jumping back to late 2020, Apple shipped the very first M1 silicon in Macs alongside the A14 Bionic from the iPhone 12 upon which it is heavily based. Chip design is an incredibly difficult task, and reuse of elements is key to making variants.
[…]
When the Studios and Mac Pro sold poorly, Apple made the call in the last quarter of ’23 to cancel revising the Studio or Mac Pro with an M3 Ultra – too low volume to be worth it. Tragically this was the precise moment when Apple was realizing local compute was vital.
By mid ’24 when all those high-memory configs were selling like crazy (relatively) due to LLMs, Apple had already made two terrible mistakes: killed the M3 Ultra that could’ve sold right into that demand. And worse, they had deleted the UltraFusion from the M4 entirely.
There will never be an M4 Ultra not due to some policy, but due to one of the worst near-misses in computing history. And given how long it took them to emergency ship the M3 Ultra (which was probably initially for Private Cloud Compute), the M5 might not be interposed either.
Previously:
