Geekbench

“They tuned (the last version of their) benchmark specifically to give better numbers on hardware that tends to throttle under performance load.”

“heavily tuned to exactly the code Apple silicon does well”

“Geekbench is known to be favoring specialized instruction sets like AVX”

“geekbench corresponding very closely to SPEC as well”

“geekbench results cannot be compared across platforms”

“the benchmarks are quite consistent, even across platforms, and track SPEC results fairly accurately”

“Geekbench is Apple shill (designed for Apple on purpose”

“Geekbench is the only consumer usable cross-platform benchmark”

“I dont think Geekbench is a good cross-platform benchmark, given what it measures”

“I always got the sense that it had a sort of pro-Apple bias that doesn't seem to be reflected in real world applications.”

“The Geekbench multi-threaded results do not provide any useful information about how a computer will behave for such applications.”

“cheating on Geekbench is more difficult because of the necessity of runtime binary patching”

“I have no reason whatsoever to trust that the Geekbench implementers are capable to optimize the inter-thread communication well enough”

“Geekbench can be gamed by CPU vendors.”

“It's just a bad practice among any kind of benchmark, basic or not.”

“Personally I think Geekbench is a pretty damn good consumer testing benchmarks.”

“The CUDA addition (a vendor-specific API) is definitely a mistake.”

“Including workloads that are almost entirely single-threaded does not tell you anything about the "multi-core" performance of the processor.”

“Geekbench is extremely good at showing general CPU performance, especially ST.”

“Geekbench 5 and 6 both correlated with SPEC scores at nearly 1:1.”

“Synthetic benchmarks are easy to use and easy to understand but borderline useless.”

“I believe it's an excellent proxy for general performance of a CPU”

“The Geekbench 5 approach of pretending Amdahl's Law doesn't exist is sometimes a valid benchmarking strategy”

“Geekbench 5 is famously much better than Geekbench 6 for benchmarking these workstation-class CPUs”

“geekbench is notorius for being absolutely shit at showing actual processing power”

“Both that and the SPEC compiler shenanigans are cheating by changing the test”

“I just don't believe these geekbench numbers represent real world performance numbers”

“Geekbench is a lot better than Cinebench IMO, because it's more "real world”

“it has a bunch of workloads, unlike many other benchmarks like cinebench for example”

“It has nothing to do with Geekbench 6 being a surprisingly bad benchmark, especially for big CPUs.”

“The scaling is atrocious, even for tasks that should be extremely parallelizable.”

“Geekbench 6 Multi-Core is fundamentally a single-task benchmark.”

“because all we have is vague marketing language, people use it as the latter”

“The problem is that it is labelled as a general multicore CPU metric, and is used as such, including for 18-core vs 96-core comparisons.”

“GeekBench probably made the right choice to optimize for more realistic real-world workloads”

“You can only really trust it for inter-generational comparisons within a single architecture.”

“Geekbench results do not reflect at all the performance of such applications”

“This makes the score an unreliable narrator of performance”

“I like Geekbench's approach of combining general and specialized tests to reward CPUs for being well-rounded”

“just puts a list of all benchmarks, even when the CPU is the same”