Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
The CPU performance improvement wasn't really dramatic compared to M2 and therefore, it's fishy. At this point, I would wait for actual results especially since M3 Max won't be available till late November.
It'll be fun when the chips are released and you realize how wrong you are. This score is perfectly believable and in line with the M3 Max having 50% additional performance cores, faster performance and efficiency cores, and clock speed improvements all around.
 
Eh, I'll admit I was a bit of a jerk there. I appreciate all the info 🙏

I am more frustrated by the ambiguous charts apple has been sharing that has everyone scrambling for info.
Marketing strategy. This way Apple gets a rolling promotion campaign for free for the new machines instead of a single sugar hit


what will be interesting is how the M3 Pro compares with its predecessor, given lower memory bandwidth and swapping going on with performance and efficiency cores.
 
Seems the single core scores are going to be close between the individuals in the chipset:

View attachment 2305901
That’s been the pattern so far. Across the board, from the lowest to the highest of a generation, single core scores are fairly close. Such that, if executing a low intensity single threaded process, there’s little practical difference between a base and an Ultra. That also means that one generations single core base score is more performant than the previous generation’s Ultra score.

That’s something neither AMD nor Intel will ever ship. :)
 
It makes the M2 Ultra Mac Pro looks silly.

I do really hope they have something more exciting planned for the Mac Pro, like what was rumored for the M2 Extreme or whatever they wanna call 4 of the SoC's fused together.

Now that would be something a bit more powerful.
The number of folks that truly NEED a desktop is going to continue shrinking further and further. Folks that need desktops won’t need them for the performance, they’ll need them for the ports or slots, things the laptops just don’t have or don’t have in that quantity. As a result, the sales of ANY desktop is going to likely be far less than a million a year. The question is whether or not they would find value in producing something like an “Extreme” when the ROI just isn’t there.

Because, whenever they release the Ultra, whether it’s 12% faster or 30% faster, it’s not like anyone’s going to be able to buy a faster Apple Silicon Mac anywhere else.
 
I guess that takes some of the sting out of the high prices. But you’d expect this kind of improvement in the next generation of PC processors and GPUs yeah? Chip shortages aside…
 
Wow. Just step back and ponder how crazy it is to be able to buy a laptop with almost 100 billion transistors that also gets 22-hour battery life and is two-thirds of an inch thin and weighs under 5 pounds. The continual scaling and shrinking of technology is amazing to watch.
I remember the memes when the G5 Powerbook was promissed and never shipped, but even compared to the G4 Powerbooks the new MBPs are and order of magnitude smaller for the gains made
 
That's nearly the same multicore score as Intel's 13900K, which is insane. Extremely interested to see real-world benchmarks like compile times.
Agreed, impressive. It’s a preliminary score and a small sample though. As you state I’m looking forward to the more cpu app generated benchmarks as well. Power/performance is the true killer stat that will emerge. What kills Apple as usual though is the price I just installed an i9 13900k tonight which i bought on prime day for $520. 15th gen is 3-4 months away. You can build a comparable desktop loaded with storage, ram and amazing graphics for less than 1/3rd a comparably loaded Mac. Depends on your use case and technical skills but in the end it’s software availability that really hurts Apple too. They can solve that - they’re worth &#3@&%%ing 3 trillion dollars as a company but for whatever reason they won’t. I’ve been waiting 20 years with feigned commitments and uplifting launches and have owned a Mac since 1988 and my SE.
 
The result seems fishy.
ok as far as they go but…

If you look at the opencl scores for the m3 max and the ones for the m2 ultra, you’ll see that the m2 ultra scores a little more than 1/3 higher: 128608 for the m2 ultra and 93579 for the m3 max so don’t bin your m2 Studio yet, especially for heavy duty graphics and hi-rez video. Built-in ray tracing and shading functions in the m3 max may even the score in some uses. Real world comparisons await.
 
Mac Studio Ultra users just lost half of its value.
Unlikely. I just bought a new Mac Studio M2 Ultra a couple of weeks ago. It's the lower of the two Ultra models, fewer GPU cores, but the same base Ultra. It will totally outperform any of these M3 models.
The Studio M2 Ultra has 800GB/s memory bandwidth, twice the speed of the Studio M2 Max, and far surpassing the M3 (150GB/s memory bandwidth). Also Studio M2 Ultra's SSD is extremely speedy, the fastest storage I have ever seen. I'm sure glad I bought a Build To Order model upgraded to 4Tb SSD (wow was that expensive). Anyway, these are not likely to be features you will find in any laptop. Until today, the M2 Studio is the most recently updated Mac (only like 3 months ago?) so I figured a Mac Studio M3 Ultra won't be out any sooner than late 2024.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.