Intel Core i9-10910: Rumored Apple Exclusive Fails to Impress on Geekbench

(Image credit: Shutterstock)

The Intel Core i9-10910, which is believed to be an Apple-exclusive chip, has shown its face on Geekbench 5 again. Without confirmation from Intel, of course, we can't be positive of the accuracy of the results, spotted by @Leakbench. But this allows us to do an early pseudo-assessment of the CPU's performance. 

The Core i9-10910 in question, as well as the current Core i9-10900, belong to the 10th Generation Comet Lake-S family. The 14nm processors find their home inside the new LGA1200 CPU socket. They're equipped with 10 CPU cores, 20 threads and 20MB of L3 cache. One theory as to the Core i9-10910's existence is that Apple wants  a custom chip for its iMac.

For context, the Core i9-10900 has a 2.8 GHz base clock and a 5.2 GHz TVB (Thermal Velocity Boost) clock. Today's Geekbench 5 submission paints the Core i9-10910 with a 3.6 GHz base clock and a boost clock that almost hits 5 GHz. The most attractive aspect of the Core i9-10900 is the fact that the processor is rated for 65W, so it isn't overly demanding on cooling or power delivery. The finer details of the Core i9-10910 are still unknown at this point, so we don't know if there is a change in the TDP (thermal design power) or not.

Core i9-10900 vs. rumored Core i9-10910 (Image credit: Primate Labs Inc.)

The two Geekbench 5 results depict exactly what the specifications already hinted. The Core i9-10910 excelled in single-threaded workloads but ultimately lost to the Core i9-10900 when multi-threaded workloads are involved.

Despite the 28.6% higher base clock, the Core i9-10910 only delivered up to 6.9% higher single-core performance than the Core i9-10900 in Geekbench 5. The tables turned when it comes to multi-core performance though. The Core i9-10900 outperformed the Core i9-10910 by 9.6%.

Assuming these results are correct, it shouldn't be hard to pick between a Core i9-10910 or Core i9-10900. You'd just have to identify the type of workloads you commonly execute on your system. The Core i9-10910 purportedly has better single-core performance, and the Core i9-10900 is expected to offer superior multi-core performance.

Zhiye Liu
News Editor and Memory Reviewer

Zhiye Liu is a news editor and memory reviewer at Tom’s Hardware. Although he loves everything that’s hardware, he has a soft spot for CPUs, GPUs, and RAM.

  • Provideroffacts
    This article is so half-baked that I’m surprised it passed the editor’s desk. The author is comparing the performance of a i9-10900 in a Hackintosh (presumably with ample cooling) to a i9-10910 that’s thermally constrained inside the iMac chassis. Obviously, the i9-10900 with sufficient cooling (and possibly relaxed power limits / BCLK) is going to win out in multi-core performance.
    Reply
  • margrave
    You're right. But ... most everything here is half-baked, eh?
    Reply
  • watzupken
    It just shows that is thermally constrained and that is hardly surprising.
    Reply
  • AlistairAB
    Yeah they are the exact same CPU. This article is a bit confused. You can change the power limit in a 10900, the writer is not using 65W results, of course there is no way the 10900 would score higher than the 109100.

    Set your 10900 to 200W? There you go.
    Reply
  • Zenlop
    It is not. The articles focus is not the performance itself but the CPU choice. Why would Apple choose a 95w when they could have picked a 65w CPU? People are starting to get fed-up with Apple's obsolescence practices. Imagine the ssd going bad. Now you need to take the whole computer to a recovery center (to get the data back) instead of an ssd card. This is garbage.
    Reply
  • TerryLaze
    Zenlop said:
    Now you need to take the whole computer to a recovery center (to get the data back) instead of an ssd card. This is garbage.
    Also the apple center is going to charge you as much as the full price of the system for that, maybe even more.

    But on topic we don't know if this CPU was even running inside an mac,it could be a developer platform or whatever testsystem,it could also be that it was a mac and the technician just changed it to 95 to test or just to run the bench while in the final product it will run at lower TDP.
    At this point everything is speculation.
    Reply
  • JordonB
    According to Apple fans, the realty about the slower performance and overheated throttling typical of many Macs is of no consequence. They think Apples look good (they do) and care little for things like privacy, performance and sharing data with the NSA. This is not conspiracy theory as the Apple server/user customer glitch proved. With Sequoia, Apple servers must digitally give the go ahead for every non apple program to start. Without comms your stuck with only limited use. The excessive pricing of simple repairs and constant lawsuits against individual tech companies that work on Apple devices show that profit is all that Apple cares about. Much like some people in powerful positions, their fans care not how much their rights are abused and the users are scrwed. Stop drinking the apple juice and try a different flavor.
    Reply