[xbenjii]

I'm confused. They're claiming "Apple’s M4 Max is the first production CPU to pass 4000 Single-Core score in Geekbench 6." yet I can see hundreds of other test results for single core performance above 4000 in the last 2 years?

[ceejayoz]

Are those production results?

https://browser.geekbench.com/v6/cpu/1962935 says it was running at 13.54 GHz. https://browser.geekbench.com/v6/cpu/4913899 looks... questionable.

[zeroonetwothree]

https://browser.geekbench.com/v6/cpu/7531877 seems fine?

[ceejayoz]

It seems to be a pretty large outlier. https://browser.geekbench.com/processors/intel-core-i7-13700...

[wtallis]

7614 MT/s on the RAM is a pretty large overclock for desktop DDR5.

[ThatMedicIsASpy]

There are 8000MT/s CUDIMMs for the new Intel Chips now...

[wtallis]

They've been announced, within the past two weeks, and as far as I can tell aren't actually available for purchase from retailers yet: the only thing I've seen actually purchasable is Crucial's 6400MT/s CUDIMMs, and Newegg has an out-of-stock listing for a G.Skill kit rated for 9600MT/s.

The linked Geekbench result from August running at 7614 MT/s clearly wasn't using CUDIMMs; it was a highly-overclocked system running the memory almost 20% faster than the typical overclocked memory speeds available from reasonably-priced modules.

[m00x]

Geekbench is run pre-release by the manufacturers.

[xbenjii]

Yeah that's fair lol

[t-sauer]

As far as I can tell those are all scroes from overclocked CPUs.

[zeroonetwothree]

https://browser.geekbench.com/v6/cpu/7531877 doesn't seem to be

[t-sauer]

That result is completely different from pretty much every other 13700k result and it is definitely not reflective of how a 13700k performs out of the box.

[zamadatix]

Geekbench doesn't really give accurate information (or enough of it) in the summary report to make that kind of conclusion for an individual result. The one bit of information it does reliably give, memory frequency, says the CPU's memory controller was OC'd to 7600 MT/s from the stock 5600 MT/s so it feels safe to say that number with 42% more performance than the entry in the processor chart also had some other tweaks going on (if not actual frequency OCs/static frequency locks then exotic cooling or the like). The main processor chart https://browser.geekbench.com/processor-benchmarks which will give you a solid idea of where stock CPUs rank - if a result has double digit differences from that number assume it's not a stock result.

E.g. this is one of the top single core benchmark result for any Intel CPU https://browser.geekbench.com/v6/cpu/5568973 and it claims the maximum frequency was stock as well (actually 300 MHz less than thermal velocity boost limits if you count those).

[api]

Could those be overclockers? I often see strange results on there that looks like either overclockers or prototypes. Maybe they mean this is the fastest general purpose single core you can buy that is that fast off the shelf with no tinkering.

[Refusing23]

AMDs upcoming flagship desktop CPU (9800 X3D) reaches about 3300 points on singlecore (the previous X3D hit 2700ish)

[grecy]

Are you saying a product that has not been released yet will be faster than a product that is?

And that a desktop part is going to outperform a laptop part?

[IshKebab]

I think he was backing up Apple's claim.

[Dylan16807]

No, neither of those.