How much RGP light tax is required for that to occur? 🤣
Kidding aside there are other metrics than just that.
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.
M2 Ultra Chip Benchmark Results Reveal Impressive Performance Gains
- Thread starter MacRumors
- Start date
- Sort by reaction score
Kidding aside there are other metrics than just that.
Remember how everyone doubted M1 before it came out? Nothing seems to have changed.
Passmark 😂😂😂😂 use a benchmark that actually runs properly on Apple silicon. (Hint, its the one every professional reviewer uses to give balanced comparisons. Its the one in this article)So we are comparing an M2 Ultra with Intel 28 Core unit from 2016, which seems unfair.
These days Intel and AMD have CPUs with 56 and 64 cores with access to 4TB of RAM.
The most significant benefit of the M processors is power consumption, but workstation has fewer restrictions and needs to deliver raw power.
I would love to see a comparison with the newer CPUs from AMD and Intel.
Apple is not even in the first 100 when it comes to raw power
PassMark Intel vs AMD CPU Benchmarks - High End
PassMark Software - CPU Benchmarks - Over 1 million CPUs and 1,000 models benchmarked and compared in graph form, updated daily!www.cpubenchmark.net
Not to mention will any x86 chip in that performance range be acceptable in a silent room or silent PC use case?
Mac Studio's active HSF is designed to make it as silent as possible.
The Ultra's die shrink makes it possible to run cool enough for that overengineered active HSF and yet have the highest performance per watt in its chip class.
i9 & RTX 4090 may make sense in a room where having multiple airblowers are acceptable like say gaming but the Ultra was never positioned for that. Apple took 3 years to get Game Porting Toolkit (GPT) out... that's how serious they are with gaming on the Mac.
Everyone thought it would be absurd that a iGPU would by equal to a 4yo dGPU until Apple did so for a $499 3.7L desktop that sipped less than 30W of power.
Might want to edit your post. Whatever you linked to has some random Xiaomi Arm chip, eating the 13900k’s lunch 😂. You also must have typoed on your stated score of “255,000”, a score that high might not come in our lifetime… we are at 30k peak on all manufacturers right now.
There are clearly some fakes in there. Also all the higher than expected results for the 13900k which aren’t fakes are highly overclocked using extreme liquid cooling. Those use well over 500w just for the cpu
I'd like primate labs to explain what happened between Geekbench 5 and 6. The Ultra used to score almost double of the M2 Max, then with Geekbench 6 it was much lower. Is this internal Geekbench programming not being as good, are there some new limitations in GB 6 that weren't there in GB5?
I read their blurb, but nothing stood out to me to explain this
I read their blurb, but nothing stood out to me to explain this
And does M2 Ultra supports water cooling? It's all about the choice.
No doubt. You wouldn't believe the arguments I've been getting in with people who can't accept that a device that is likely 12 months from shipping is completely locked hardware wise. It's like the Apple Silicon DTK, with an A12z, they're showing this off with hardware that exists so developers can start working on software, but when they're ready to sell it, it will be the best device they can make. Shipping this device in 2024 with GPU cores from 2020 and on essentially the same node as the M1 when there's a 3nm chip with 3+ years of development that could give it potentially 20%+ performance increase or 20+ battery life increase just to keep specs in line with an announcement video would be insane
I will wait for real benchmarks to come out. Geekbench doesn't sell me anymore.
The i9 wouldn’t have been in a Mac Pro, so it wouldn’t have been an appropriate answer to the question.
No Apple does not allow overlocking so there is absolutely no need for liquid cooling.
Those overlocked i9-13900k results are impressive though.
Except those Intel systems (especially with an nVidia graphics card) don't run macOS. If someone is building a computer to just crunch numbers, then an Intel system can be faster and cheaper potentially. But it doesn't run macOS, so right there it's going to be a hard stop if macOS is what you want to run...
I don’t see how this is expected to compete with only 192GB RAM when the old one had 1,536GB RAM. I mean it depends on your workload but idk how you can keep the fixed amount so low and expect everyone to be fine upgrading to that from the old machine. It’s not modular if you can’t upgrade the RAM. They need to figure out some faster interconnects for that.
Or are they just going to completely ignore that market segment moving forward? It’s just such a strange device that seems like a half-assed attempt being pseudo upgradeable. I mean I don’t even think you can upgrade GPUs in it anymore. And yet it’s so big.
I honestly don’t see the point of it vs. the Mac Studio, and this seems to be by design. Apple can point to poor Mac Pro sales and spew out some PR crap: “Users love the power and portability of Mac Studio, and we look forward to bringing exciting new updates to this platform moving forward.” And maybe they’ll bring back the iMac Pro and call it the iMac Studio with a 32” 6K display.
I just don’t think the Mac Pro is long for this world.
Or are they just going to completely ignore that market segment moving forward? It’s just such a strange device that seems like a half-assed attempt being pseudo upgradeable. I mean I don’t even think you can upgrade GPUs in it anymore. And yet it’s so big.
I honestly don’t see the point of it vs. the Mac Studio, and this seems to be by design. Apple can point to poor Mac Pro sales and spew out some PR crap: “Users love the power and portability of Mac Studio, and we look forward to bringing exciting new updates to this platform moving forward.” And maybe they’ll bring back the iMac Pro and call it the iMac Studio with a 32” 6K display.
I just don’t think the Mac Pro is long for this world.
How about you post a video of you or someone running Geekbench 6 on a stock i9-13900K system? No overclocking, no overvolting/undervolting, no water-cooling, no fudging results. Then run the benchmark on a stock M2 Ultra. Post that video as well. Total CPU power use is also a factor to consider (Apple's engineers are not giving Apple Silicon the same TDP as Intel's high end chips; they could though) so also include performance per watt as an additional metric.
Also running other benchmarks than just Geekbench will be necessary.
Then, do real-world workloads of equally optimized software.
Then have at least 30 other people independently do the same thing so we have a little more data to be comfortable with statistical comparisons.
Because what it looks like is the typical result for an i9-13900K is 2940 single core: https://browser.geekbench.com/processors/intel-core-i9-13900k
That's great performance on the benchmark but it's not matching your cherry-picked other scores. The cheery-picked results might be interesting but if they are not readily reproducible, they aren't particularly meaningful and are not representative of real world workflow performance.
I'm not discounting benchmarks, they are important but they are not the only important thing.
I will wait patiently for your videos and results, or at least a measured reply providing more evidence to support your claims.
Last edited:
Yeah but it sounds like you're saying like they have been working on it for 3 years, irl they may have made it in few months porting DXVK guy's code to be compatible with metal instead of vulkan
What I was pointing out is Apple's priority for Mac gaming.
If it took them 3 years to get GPT via DXVK guy's code then it isn't that high up in their priorities list.
Apple Makes More on Games than PlayStation, Xbox, Nintendo + Activision + Windows COMBINED.
Their incentive to get into Mac gaming is lazy or smart scheduling considering PC gamers for the past 3 years keep pointing out that the M1, M2 and hopefully not the M3 have an iGPU too weak compared to dGPU to play triple-A titles.
Last edited:
Why would I when PC has way more potential than Apple Silicon which Apple restricted so heavily? You cant do that? Well, that's Apple Silicon's problem, not PC. You only care about power by watt which is totally meaningless and pointless especially toward desktop users who need high performances. Beside Intel 13th gen are Intel 7 which is more like TSMC 7nm, not 5nm which you didn't even care. It is Apple who decided to make power by watt focused chips with a lot of restrictions so I dont see your point. I dont care about power by watt when the performance is too low or poor while working in real life. That's how Apple GPU failed as they compared M1 Ultra to RTX 3090 and yet, it never ever close to it. Typical Apple's lie that only Apple fanboys praise for it.
Last edited: