Live data from Hacker News

MacBook Pro with M5 Pro and M5 Max

apple.com

341–350 of 1001 posts

Re: MacBook Pro with M5 Pro and M5 Max

#341

Earlier quoted context omitted.

My 32gb m1 max was probably the best purchase I've made. Still plenty of headroom in performance left in this beast. Wonder what reason they'll use to end software support in the future. Bet it'll be some security hardware they make up for the sake of forcing upgrades.

my tinfoil hat theory is that they make small features depend on new hardware. for example, let's say the new os depends on m5's exclusive thumbnail generator accelerator, and let's say it improves speed by a 20%. now, your M1 notebook than on previous OSes uses standard gpu acceleration for thumbnails will not have this specialized hardware acceleration, it will have software fallback that will be 90% slower. you wo…

It's not tinfoil, that's just how publicly traded companies work - increasing the share value

Re: MacBook Pro with M5 Pro and M5 Max

#343
post #322

Earlier quoted context omitted.

Seems very reasonable to me

A bit strange to use time to first token instead of throughput. Latency to the first token is not like a web page where first paint already has useful things to show. The first token is "The ", and you'll be very happy it's there in 50ms instead of 200ms... but then what you really want to know is how quickly you'll get the rest of the sentence (throughput)

I assume it’s time to first output token so it’s basically throughput. How fast can it output 8001 tokens

Re: MacBook Pro with M5 Pro and M5 Max

#344
post #322

Earlier quoted context omitted.

Seems very reasonable to me

A bit strange to use time to first token instead of throughput. Latency to the first token is not like a web page where first paint already has useful things to show. The first token is "The ", and you'll be very happy it's there in 50ms instead of 200ms... but then what you really want to know is how quickly you'll get the rest of the sentence (throughput)

As far as benchmarketing goes they clearly went with prefill because it's much easier for apple to improve prefill numbers (flops-dominated) than decode (bandwidth-dominated, at least for local inference); M5 unified memory bandwidth is only about 10% better than the M4.

Re: MacBook Pro with M5 Pro and M5 Max

#345

Earlier quoted context omitted.

The memory bandwith on M4 Max is 546 GB/s, M5 Max is 614GB/s, so not a huge jump. The new tensor cores, sorry, "Neural Accelerator" only really help with prompt preprocessing aka prefill, and not with token generation. Token generation is memory bound. Hopefully the Ultra version (if it exists) has a bigger jump in memory bandwidth and maximum RAM.

Do any frameworks manage to use the neural engine cores for that? Most stuff ends up running Metal -> GPU I thought

It's referring to the neural cores(for matrix mul) in the GPU itself, not the NPU.

https://creativestrategies.com/research/m5-apple-silicon-its...

Re: MacBook Pro with M5 Pro and M5 Max

#346
post #310

Earlier quoted context omitted.

So Firefox was using 5gb? There's your problem.

It’s a losing battle me trying to tell my wife to close her Firefox tabs, haha, but yes, Firefox does use a lot of ram when you have 500 tabs. Maybe I’ll get her a 64GB MacBook Pro for the premium web browsing experience she so desires!

Tabs as bookmarks, people keep falling into this trap, my wife included

Re: MacBook Pro with M5 Pro and M5 Max

#347
post #320

Earlier quoted context omitted.

Which majority? I certainly only use Macs when being project assigned, then there are plenty of developers out there whose job has nothing to do with what Apple offers. Also while Metal is a very cool API, I rather play with Vulkan, CUDA and DirectX, as do the large majority of game developers.

Honestly though, gamedevs really are among the biggest Windows stalwarts due to SDK's and older 3d software. Only groups of developers more tied to Windows that I can think of are probably embedded people tied due to weird hardware SDK's and Windows Active Directory dependent enterprise people. Outside of that almost everyone hip seems to want a Mac.

80% of the desktop market has to have their applications developed by someone, at least until software replicators replace them.

Everyone hip alright, or at least those that would dream to earn a salary big enough to afford Apple taxes.

Remember there are world regions where developers barely make 1 000 euros per month.

Re: MacBook Pro with M5 Pro and M5 Max

#348

Earlier quoted context omitted.

In previous generations, throughout was excellent for an integrated GPU, but the time to first token was lacking.

So throughput was already good but TTFT was the metric that needed more improvement?

Good is relative but first token was clearly the biggest limitation.

Re: MacBook Pro with M5 Pro and M5 Max

#349

I just bought a M5 Macbook Pro 2 weeks ago. Thinking of returning it and getting a M5 Pro with the same configuration but only $200 more. How should I compare M5 vs M5 Pro?

You'll get slightly more performance and ever so slightly less battery life. I'd do it

Re: MacBook Pro with M5 Pro and M5 Max

#350

I chased down what the "4x faster at AI tasks" was measuring: > Testing conducted by Apple in January 2026 using preproduction 13-inch and 15-inch MacBook Air systems with Apple M5, 10-core CPU, 10-core GPU, 32GB of unified memory, and 4TB SSD, and production 13-inch and 15-inch MacBook Air systems with Apple M4, 10-core CPU, 10-core GPU, 32GB of unified memory, and 2TB SSD. Time to first token measured with an 8K-to…

Seems very reasonable to me

I would consider it reasonable if this was 4x TTFT and Throughput, but it seems like it's only for TTFT.
Post reply on HN