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

MacBook Pro with M5 Pro and M5 Max

apple.com

Post reply on HN
Print view View original

781–790 of 1001 posts

Re: MacBook Pro with M5 Pro and M5 Max

#781

Post by seec »

OccamsMirror wrote: post #689
compounding_it wrote: post #336

Earlier quoted context omitted.

yes pretty much this. make useless features use up resources and make basic scrolling slow. the Liquid Glass for example probably is not so great when it comes to resources. Probably works better with latest metal and hardware blocks on the GPU in M5 as opposed to using GPU cores and unified memory on 8gb M1 making latest macOS work not so great. I have the M1 8gb air and it is really slow on Tahoe. It was snappy jus…

I'm so tempted to do this. But having to wipe my MBP is currently too much friction for me. Liquid Glass is really killing my love for Apple products. I'll probably get a Framework and an Android phone for my next device purchases. They really need to just admit it was a bad move and make like Sonic.

The animations and layout of Liquid Glass aren't that bad, but it is really ugly in many ways.

They could have just made some layout improvements without trashing everything visually; that's sad, really.

The contour they put around the icon is really, really bad. How the fuck did they approve that?

Re: MacBook Pro with M5 Pro and M5 Max

#782

Post by freehorse »

danudey wrote: post #334
GeekyBear wrote: post #332

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?

Let’s say TTFT needed the most improvement. At some point, loading the model with enough context size may take tens of seconds in some macs.

Re: MacBook Pro with M5 Pro and M5 Max

#783

Post by sroussey »

fnordpiglet wrote: post #753
jbellis wrote: post #298

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…

I think the key stats are this (for m5 max) M5 128GB RAM with 614GB/s memory transfer This is a huge step over M4 32GB 153GB/s memory transfer For local LLM this make it a replacement for a DGX Spark, which offers a third of the transfer speed and is not something you toss in your backpack as your laptop. It’s practically useful for a lot of local use cases and that I think is the 4x factor (memory xfer) - but the 12…

You are comparing a M5 Max to a base M4.

The M4 Max has 546 GB/s compared to 614GB/s for the M5 Max. Which is like 12% faster not 4x.

Re: MacBook Pro with M5 Pro and M5 Max

#784

Post by sgt »

swyx wrote: post #367
bob1029 wrote: post #244

I feel like Apple pulled an Instant Pot with the M1 MacBook Pro. I still haven't had a single situation where I felt like spending more money would improve my experience. The battery is wearing out a bit, but it started out life with so much runtime that losing a few hours doesn't seem to matter.

> The battery is wearing out a bit, but it started out life with so much runtime that losing a few hours doesn't seem to matter. this is my exact opposite experience. my M3 Max from 2 years ago now has <2hrs battery life at best. wondering if any experts here can help me figure out what is going on? what should i be expecting?

I've got a M2 Pro from 3 years ago and battery is still so good I can go to a whole day of meetings and not even need to bring my charger. Then I can probably work all night as well without plugging it in. Battery time is insane.

Unless of course you're doing something that truly sucks down your battery! If I spin up a few Docker instances doing 100% CPU then obviously battery will go down much quicker.

Re: MacBook Pro with M5 Pro and M5 Max

#785

Post by germinalphrase »

raw_anon_1111 wrote: post #773
garbageman wrote: post #754

Earlier quoted context omitted.

I'm curious, what does the 'determine intent' mean in this case?

An “intent” is something that a person wants to do - set a timer, get directions, etc. A “slot” is the variable part of an intent. For instance “I want directions to 555 MockingBird Lane”. Would trigger a Directions intent that required where you are coming from and where you are going. Of course in that case it would assume your location. Back in the pre LLM days and the way that Siri still works, someone had to man…

Can you recommend any good resources that discuss structure and performance improvement of these types of systems?

Re: MacBook Pro with M5 Pro and M5 Max

#787

Post by sroussey »

jbellis wrote: post #710
giancarlostoro wrote: post #437

Earlier quoted context omitted.

On my 24GB RAM M4 Pro MBP some models run very quickly through LM Studio to Zed, I was able to ask it to write some code. Course my fan starts spinning off like the worlds ending, but its still impressive what I can do 100% locally. I can't imagine on a more serious setup like the Mac Studio.

Your limitation after prefill is memory bandwidth. A maxed out Studio has less than a single 3090 (really).

Yeah, the 3090 has faster memory, but not by a lot.

The 5090 is at 1,792GB/sec and potential M5 Ultra would be 1,230GB/sec and 512GB RAM. Maybe 1TB. Not 32.

Re: MacBook Pro with M5 Pro and M5 Max

#788

Post by spockz »

hx8 wrote: post #717
kobalsky wrote: post #289

Earlier quoted context omitted.

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…

You're too far down the rabbit hole. Anytime they can make M1 incompatible with the latest version of macOS which would most people to upgrade.

Well then you can use CoreBoot (or OpenCore always forget which is which) to run newer versions on older hardware.

Re: MacBook Pro with M5 Pro and M5 Max

#789

Post by freehorse »

neya wrote: post #767
manofmanysmiles wrote: post #114

I love the following section of their copy: > Even More Value for Upgraders > The new 14- and 16-inch MacBook Pro with M5 Pro and M5 Max mark a major leap for pro users. There’s never been a better time for customers to upgrade from a previous generation of MacBook Pro with Apple silicon or an Intel-based Mac. I read as "Whoops we made the M1 Macbook Pro too good, please upgrade!" I think I will get another 2-5 years…

I upgraded to an M3 Pro from an M1 Pro. I sold my M1 Pro at 90% of the original cost (not even exaggerating) on Facebook marketplace AFTER 2 YEARS. I thought the buyer was insane to buy it at that price. But, of course mine had a decent spec and still had the Apple care warranty with very low battery cycle count. After the sale, the buyer told me the truth: The M1 is the best chip Apple ever made and I wouldn't see m…

M3 was a weird generation, as they contained fewer transistors than the previous ones. It is slightly faster in single core tasks, and has a few more cores, but they are very close. But in terms of gpu, m3s are quite nerfed esp because they lowered the memory bandwidth, so on llm performance they are on par. I have both an m3 and an M1 Max, one of them from work, so I have tested them extensively (though the m3 is binned and 14”, the m1 full and 16”). M3 had better TTFT but the M1 had a bit higher tokens/s.

Re: MacBook Pro with M5 Pro and M5 Max

#790

Post by imoverclocked »

me551ah wrote: post #771
bbshfishe wrote: post #666

Earlier quoted context omitted.

Yeah no it didn’t. If you have a fully speced out M3/4 MacBook with enough memory you’re running pretty decent models locally already. But no one is using local models anyway.

If your company can afford fully speced out M3/4 MacBook, then it can also afford cloud AI costs.

Perhaps, but sending everything to the cloud might get them in (very expensive) trouble. Depending on who we are talking about, of course.
Post reply on HN