Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Leaked Apple M5 9 core Geekbench scores

browser.geekbench.com

Post reply on HN
Print view View original

231–240 of 565 posts

Re: Leaked Apple M5 9 core Geekbench scores

#231

Post by ascagnel_ »

jackothy wrote: post #220
ascagnel_ wrote: post #122

Earlier quoted context omitted.

> though if Apple suddenly took gaming seriously, like extending Rosetta to act as a translation layer for Windows games a la Proton, I’d gladly throw down for an M5 Ultra when it’s released No joke, if I could run my Steam library on my phone, I'd probably buy a new phone every year (and might need to, given what the thermals and rapid charge/discharge cycles would do to battery longevity). But Apple's current strat…

The problem is they don't "let the developers do the work themselves" If only the platform was open enough that developers had real access, Apple might get away with like you say not providing first party support for gaming.

I mean it more as a counterpoint to what Valve is doing: games are automatically opted-in to working with Proton and pass through a Valve-operated certification program. Because developers aren't part of that loop, you see occasional cases where games issue updates that break compatibility, with users blaming developers based on an outdated, Valve-provided certification.

Re: Leaked Apple M5 9 core Geekbench scores

#232

Post by GeekyBear »

stego-tech wrote: post #72

Damn impressive if true, but I’ll be honest, I still don’t feel the need to replace my M1 iPad Pro or my M1 Pro Macbook Pro. They’re both more than amazing for my use cases - though if Apple suddenly took gaming seriously, like extending Rosetta to act as a translation layer for Windows games a la Proton, I’d gladly throw down for an M5 Ultra when it’s released.

> if Apple suddenly took gaming seriously, like extending Rosetta to act as a translation layer for Windows games a la Proton

Apple already provides the translation layer to convert from DirectX 11 or 12 to Metal that Wine uses on Macs.

https://wccftech.com/apple-game-porting-toolkit-2-supports-a...

Proton does the exact same thing, only it translates DirectX into the Graphics API that Wine on Linux uses.

The new thing is that the M5 versions of the GPU cores picked up a 40% performance boost, on the version that just shipped on the new iPhones.

Re: Leaked Apple M5 9 core Geekbench scores

#233

Post by whatever1 »

criddell wrote: post #123
jyap wrote: post #109

Earlier quoted context omitted.

Regarding market share and your friends and family recommendations, you’re thinking first world. Rest of the world wants and can only afford sub-$500 laptops.

I’ve found that the $1000 Mac laptop is worth about $500 after 3 years and the $500 laptop is worth $50. The price difference over time really isn’t that big and the Mac is going to have a better trackpad and display and longer battery life.

If your $1000 MacBook breaks after a year you need $1000 to repair it.

A 500 laptop is probably more repairable and worst case you pay $500 to get a new one. Not to mention battery replacement etc.

The expected total cost of ownership is very high for a Mac. It’s like owning a Mercedes. Maybe you can afford to buy one, but you cannot afford maintenance.

Re: Leaked Apple M5 9 core Geekbench scores

#234

Post by kstrauser »

aleph_minus_one wrote: post #225
kstrauser wrote: post #167

Earlier quoted context omitted.

>> Compiles code twice as fast as my new Windows desktop >That's because MS's filesystem layer has been garbage since NT was launched decades ago [...] I confess that this kind of excuse drives me batty. End users don't buy CPUs and buy filesystems. They buy entire systems. "Well, it's not really that much faster, it's just that part of the system is junk. The rest is comparable!" That may be, but the end result for…

> End users don't buy CPUs and buy filesystems. They buy entire systems. [...] Possibly true, but still totally irrelevant if the rest of the system can't keep up. At least historically for hardware components of PCs, this was not irrelevant, but the state of things: You basically bought some PC as a starting basis. Because of the high speed of improvements, everybody knew that you would soon replace parts as you dee…

But that still doesn't save away, "oh, it's only slow because the filesystem is so slow". Assuming that's true, that's a very integral part of the system that can't readily be swapped out by most people. You can't say "the system is actually really fast, it's just the OS that's slow", because the end result is just plain "the system is slow."

Re: Leaked Apple M5 9 core Geekbench scores

#235

Post by onionisafruit »

tedk-42 wrote: post #226
porphyra wrote: post #166

Earlier quoted context omitted.

Nice. Lots of people still claim that M1 is super duper fast but now we're at almost twice the performance!

A car that travels 0-60 MPH in 2 seconds is fast A car that does this in 4 seconds is still fast (though twice as slow)

If they both top out at 60 they are equally fast. One just had better acceleration than the other

Re: Leaked Apple M5 9 core Geekbench scores

#236

Post by Reason077 »

bhouston wrote: post #5

Nice. The iPads generally measure around ~8% slower than the MacBooks, I guess for cooling reasons. So we should see approximately a 4400 single core Geekbench score for the MacBook series. This is nice. Single thread MacBook progression on Geekbench: M1: 2350 M2: 2600 M3: 3100 M4: 3850 M5: 4400 (estimated) https://browser.geekbench.com/mac-benchmarks

Here's the comparison between the M4 iPad Pro and the M5:

https://browser.geekbench.com/v6/cpu/compare/14173685?baseli...

About 10% faster for single-core and 16% faster for multi-core compared to the M4. The iPad M5 has the same number of cores and the same clock speed as the M4, but has increased the RAM from 8GB to 12GB.

Re: Leaked Apple M5 9 core Geekbench scores

#237

Post by aleph_minus_one »

astrange wrote: post #216
FootballMuse wrote: post #108

Earlier quoted context omitted.

Passmark is an outdated benchmark not well optimized for ARM. Even so the single thread marks are 3864 (AI365) vs 4550 (M4) OTOH, Geekbench correlates (0.99) with SPEC standards, the industry standard in CPU benchmark and what enterprise companies such as AWS use to judge a CPU performance. https://medium.com/silicon-reimagined/performance-delivered-...

Hmm, why would you need to optimize a benchmark for something? Generally it's the other way round.

> Hmm, why would you need to optimize a benchmark for something? Generally it's the other way round.

It had always been both ways. This is why there exist(ed) quite a lot of people who have/had serious thoughts whether [some benchmark] actually measures the performance of the e.g. CPU or the quality of the compiler.

The "truce" that was adopted concerning these very heated discussions was that a great CPU is of much less value if programmers are incapable of making use of its power.

Examples that evidence the truth of this "truce" [pun intended]:

- Sega Saturn (very hard to make use of its power)

- PlayStation 3 (Cell processor)

- Intel Itanium, which (besides some other problems) needed a super-smart compiler (which never existed) so that programs could make use of its potential

- in the last years: claims that specific AMD GPUs are as fast as or even faster than NVidia GPUs (also: for the same cost) for GPGPU tasks. Possibly true, but CUDA makes it easier to make use of the power of the NVidia GPU.

Re: Leaked Apple M5 9 core Geekbench scores

#238

Post by aleph_minus_one »

kstrauser wrote: post #234
aleph_minus_one wrote: post #225

Earlier quoted context omitted.

> End users don't buy CPUs and buy filesystems. They buy entire systems. [...] Possibly true, but still totally irrelevant if the rest of the system can't keep up. At least historically for hardware components of PCs, this was not irrelevant, but the state of things: You basically bought some PC as a starting basis. Because of the high speed of improvements, everybody knew that you would soon replace parts as you dee…

But that still doesn't save away, "oh, it's only slow because the filesystem is so slow". Assuming that's true, that's a very integral part of the system that can't readily be swapped out by most people. You can't say "the system is actually really fast, it's just the OS that's slow", because the end result is just plain "the system is slow."

> You can't say "the system is actually really fast, it's just the OS that's slow", because the end result is just plain "the system is slow."

If performance is so critical, people do find ways around this. Just to give an arbitrary example since you mention file systems:

Oracle implemented their own filesystem (ASM Cluster File System (ACFS)):

> https://en.wikipedia.org/w/index.php?title=Oracle_Cloud_File...

"ACFS provides direct I/O for Oracle database I/O workloads. ACFS implements indirect I/O however for general purpose files that typically perform small I/O for better response time."

Re: Leaked Apple M5 9 core Geekbench scores

#239

Post by GeekyBear »

rowanG077 wrote: post #229
GeekyBear wrote: post #206

Earlier quoted context omitted.

Asahi isn't in choppy waters. The new project leadership team made the decision to prioritize getting their existing work upstreamed into the Linux kernel, before working on supporting newer SOCs. It's been going well.

I agree with that somewhat. But they still lost their main GPU kernel driver developer AND Also their uber productive user space GPU driver developer. With especially the loss of Alyssa leaving a basically unfillable hole.

Alyssa's work was already upstreamed in May, and the driver is ready to go when the kernel is ready to accept Rust GPU drivers.

> We are pleased to announce that our graphics driver userspace API (uAPI) has been merged into the Linux kernel. This major milestone allows us to finally enable OpenGL, OpenCL and Vulkan support for Apple Silicon in upstream Mesa. This is the only time a graphics driver’s uAPI has been merged into the kernel independent of the driver itself, which was kindly allowed by the kernel graphics subsystem (DRM) maintainers to facilitate upstream Mesa enablement while the required Rust abstractions make their way upstream. We are grateful for this one-off exception, made possible with close collaboration with the kernel community.

https://asahilinux.org/2025/05/progress-report-6-15/

Alyssa didn't abandon the project, she completed it.

Re: Leaked Apple M5 9 core Geekbench scores

#240

Post by astrange »

aleph_minus_one wrote: post #237
astrange wrote: post #216

Earlier quoted context omitted.

Hmm, why would you need to optimize a benchmark for something? Generally it's the other way round.

> Hmm, why would you need to optimize a benchmark for something? Generally it's the other way round. It had always been both ways. This is why there exist(ed) quite a lot of people who have/had serious thoughts whether [some benchmark] actually measures the performance of the e.g. CPU or the quality of the compiler. The "truce" that was adopted concerning these very heated discussions was that a great CPU is of much…

> - Intel Itanium, which (besides some other problems) needed a super-smart compiler (which never existed) so that programs could make use of its potential

Well, no such thing is possible. Memory access and branch prediction patterns are too dynamic for a compiler to be able to schedule basically anything ahead of time.

A JIT with a lot of instrumentation could do somewhat better, but it'd be very expensive.

Post reply on HN