I recommend watching the Mac Mini announcement video. The cooling/thermals on the Mini with M4 Pro will more sufficient.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
M4 Max Chip Up to 25% Faster Than M2 Ultra in First Benchmark Results
- Thread starter MacRumors
- Start date
- Sort by reaction score
I recommend watching the Mac Mini announcement video. The cooling/thermals on the Mini with M4 Pro will more sufficient.
I think the M4 Pro-based Mac mini could be a huge seller, especially if the end user can match it with Thunderbolt 3 or faster external storage device.
If ports is the only reason then why not a Mini and a hub (next time)? Would save a bunch of money.
*the fastest chip for geekbench. Geekbench shouldn't be used for comparing cpus with different (p-)core counts.
This is a lot less impressive when you consider the huge price disparity in this comparison.
On one hand it’s impressive how far ARM SoCs have come. On the other hand, M4 Max is a very high end SoC that ships with world class CPU cores but mid-tier almost budget level desktop performance. Really hoping they put more effort on the GPU side or branch out more SKUs with greater number of GPU cores.
Last edited:
Hubs and docks keep disconnecting. macOS is still quite horrible with that.
M4 Ultra will be bonkers...but I'll settle for a mini m4 pro as an upgrade for my, now very old, 2008 Mac Pro running various OS hacks. But I will miss the heating up the room effect.
So much "wrong-ing" in one sentence.
155K vs 192K GB6 Metal result is 15%?
16 Core GPU?
Ugh.
I tend to upgrade when there is a 2X improvement over my current machine. For M1 users the M4 is a no brainer time to update if you’re already it seeing lots of page outs, high memory pressure, and slow compute times in important tasks.
With only 16GB in my top M1 Pro laptop I’m constantly in high memory pressure and intentionally limiting app use in different ways to avoid page outs. 16GB feels way too limited these days. My 30 day processor average meanwhile is about 25%, but video production is barely possible with any effects hovering around 1 fps rendering. (Started doing video after I bought it. Been stuck waiting for 2x improvement in MacBooks.)
CPU gets lots of attention but RAM often is the main bottleneck. Even for general office use if you like to have tons of browser tabs open, and multiple apps you use across 2 or more external monitors you will need or benefit from a higher end Pro or Max device. For video, 3D, or on-device LLM work 68GB RAM is often a current recommendation or requirement in various apps. Get more if you can.
With only 16GB in my top M1 Pro laptop I’m constantly in high memory pressure and intentionally limiting app use in different ways to avoid page outs. 16GB feels way too limited these days. My 30 day processor average meanwhile is about 25%, but video production is barely possible with any effects hovering around 1 fps rendering. (Started doing video after I bought it. Been stuck waiting for 2x improvement in MacBooks.)
CPU gets lots of attention but RAM often is the main bottleneck. Even for general office use if you like to have tons of browser tabs open, and multiple apps you use across 2 or more external monitors you will need or benefit from a higher end Pro or Max device. For video, 3D, or on-device LLM work 68GB RAM is often a current recommendation or requirement in various apps. Get more if you can.
Last edited:
It comes down to who the machine is aimed at. If you’re buying an ultra presumably time is money like a production work house or you have money to burn. In a work environment it will pay for itself in those six months easy and then you buy the next thing to stay current.
If you’re a regular home user and skint yourself to buy an ultra just because it’s the best then yes that’s a bad purchasing decision.
HobeSoundDarryl
macrumors G5
I make my living on the thing... but doubt I'm taxing the full ULTRA enough to justify it for the 6 or so months that it is King. If I just kick the purchase can to Fall, I can get the next gen chip as PRO or MAX for significantly less money. Perhaps my work would tax that more often but I'll also have upwards of a full year+ before it to no longer the fastest unless Apples gets on an annual ULTRA rollout.
Yes, I know all this. But odds are probably pretty good that the guy who buys an ULTRA is not then also buying a laptop 6 months later. When it is one or the other new Mac this year, Apple can maximize profit per unit sold by selling the ULTRA or MAX vs. having the very same "wallet" kick the can for 6 months to then get as much power if not a little more in the next gen chip for much less money.
Yes, overall Apple will sell a much larger number of MBpros vs. these upper-end desktops but if the intent is to keep selling both, flipping the order has no effect on the launch date on MBpros... but may move some who kick that can to pay up for the more powerful Mac... now king of the power hill until the next year instead of only 6 months. In this concept, Apple maximize profit per unit sold... which is apparently something they like to do... without harming their higher volume Fall releases of MBs which would still launch at the very same time.
I said nothing about deferring. MBpros could still launch in October if ULTRA Macs launched in the spring before instead of the spring after. What this is doing is flipping the schedule- ULTRA for BASE- which seems logically backwards right now if Apple wants to harvest as much profit from each unit sold.
All my opinion of course but as a consumer who once bought ULTRA, I don't do it again unless the schedule flips... as Apple's choice to schedule it as they do makes the next gen MAX always the power M for which to wait... unless one need ULTRA power and can rationalize paying way up for it to have the most power for only those 6 months or so until next-gen MAX arrives.
My M1 Max MacBookPro was bought first as the Ultra MacStudio did not exist.
The MacStudio (128GB ram and 8TB SSD) was acquired along with two Studio Displays. They replaced the trash can Mac and two prior generation 27” Apple monitors.
So a fully loaded M4 Max MacBook Pro with the same memory and SSD has more power in the specs than my Studio.
Real test numbers in a few weeks will clarify the performance reality.
The MacStudio (128GB ram and 8TB SSD) was acquired along with two Studio Displays. They replaced the trash can Mac and two prior generation 27” Apple monitors.
So a fully loaded M4 Max MacBook Pro with the same memory and SSD has more power in the specs than my Studio.
Real test numbers in a few weeks will clarify the performance reality.
Written like someone who is worried over status, and not by someone who works day to day with their machine.
By far the big expense for any business that employs people are the labor costs.
The price differences between these machines are trivial (for professional/business use, certainly hobbyists/students beg to differ.)
Any equipment bought for your livelihood should be on 48 or 60 month depreciation schedules anyway.
The 3rd result has a single core score of 4060. As an audio pro that number is very exciting; finally we might be running really demanding VIs at high quality (U-he Diva!) in realtime, no freezing tracks required.
Upgrading from a 2015 MBP w/ 2.5 GHz Quad-Core Intel Core i7, 16gb RAM.
The first Geekbench 6 benchmark results for the high-end M4 Max chip with a 16-core CPU surfaced today, and they show that the chip is up to 25% faster than the high-end M2 Ultra chip with a 24-core CPU in terms of peak multi-core CPU performance.
M4 Pro chip benchmark results already surfaced in the Geekbench 6 database on Thursday. Based on the results available so far, the M4 Max appears to be up to 20% faster than the M4 Pro in terms of peak multi-core CPU performance.
Here is a comparison of the results (averaged):
M4 Max is now the fastest-ever Apple silicon chip in the Geekbench 6 database, surpassing the M2 Ultra that Apple released in the Mac Studio and Mac Pro in June 2023.
- MacBook Pro with M4 Max (16-core CPU): 26,675 multi-core score (highest result), 26,445 multi-core score (3 results)
- Mac mini with M4 Pro (14-core CPU): 22,094 multi-core score (11 results)
- Mac Studio with M2 Ultra (24-core CPU): 21,351 (More than 600 results)
As we mentioned in our previous reporting, you can now purchase a Mac mini with a 14-core M4 Pro for $1,599 in the U.S. and get similar to faster peak performance than a Mac Studio with the 24-core M2 Ultra, a configuration that starts at $3,999. And if you want up to 25% faster performance than the M2 Ultra, the 16-inch MacBook Pro with the 16-core M4 Max starts at the same $3,999 price as the Mac Studio.
Article Link: M4 Max Chip Up to 25% Faster Than M2 Ultra in First Benchmark Results
so excited for my 16gb m4 Max, 64gb, 2TB to arrive next week! Converging all my laptops into this one. Now just need to see if parallels is the best thing to get or VMware for windows 11 work stuff. Student Discount and Trade in got me this deal for roughly 2K total.
Bring on SOLITAIRE and SUDOKU
Bring on SOLITAIRE and SUDOKU
If they happen, I think it will be a bespoke chip. The Ultrafusion interconnects are missing in the m3s and m4s as they are now. It really makes me wonder about the m4 ultras/extremes/whatever they decide to call them. Are they coming out at all? Or will they just jump to m5 for the studio and pros?
Don't cherry pick the lowest M3 Max GPU Metal score then it's about 13%. Scores fluctuate for whatever reasons that could be due to plugged vs unplugged, thermal throttling, etc. Compare highest score from both.
Mac16,5 - Geekbench
Benchmark results for a Mac16,5 with an Apple M4 Max processor.
browser.geekbench.com
MacBook Pro (16-inch, Nov 2023) - Geekbench
Benchmark results for a MacBook Pro (16-inch, Nov 2023) with an Apple M3 Max processor.
browser.geekbench.com
Remember the M1 and M2 Pro and Max had the same number of CPU cores and same performance. It was mainly the GPU that differentiated them.
Wrong. Apple compared with M1 because it’s getting close for those users to upgrade, such as myself. Computer upgrade cycles are not every year or two like phones. (Hell, I upgrade my iPhone every 4 years!) I bought the base M1 mini almost 4 years ago and I’m really interested in the M4 Pro mini, but still kind of hoping and holding out for a large iMac - sometime before or at WWDC. The M4 Pro and Max could definitely, finally, make that iMac possible.
The Ultra doesn’t become unusable just because a faster, cheaper chip comes out. If it’s not worth buying at the time, then don’t buy it. Most people can’t afford to wait around and need to get work done now and can’t wait 6 months.
Never did and never will understand this argument. I have a base M1 with 8GB of RAM and it still works just fine, and have no reason to upgrade, other than wanting to replace an old (2012) mini that acts as my web/file server with it and get a new M4 Pro mini.