Lorem ipsum dolor sit amet, consectetur adipiscing elit.
Last edited:
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.
M1 Ultra Doesn't Beat Out Nvidia's RTX 3090 GPU Despite Apple's Charts
- Thread starter MacRumors
- Start date
- Sort by reaction score
You missed the point. Apple could risk hurting their integrity if its core audience found out that the chart was off. When in fact they don't need to display such a chart in the first place given what they have already accomplished.
Just because you can’t read a graph?
The ultra is a joke it’s not needed it’s just for the fact they can say it’s faster and charge double the m1 max is enough for anything macs can do, the only reason 3090 is needed is for gaming and macs can’t game
Lorem ipsum dolor sit amet, consectetur adipiscing elit.
Last edited:
I've always maintained that Apple's "performance" charts were nothing but marketing fluff - they always shy away from dropping actual names of competing devices, and they use vague terms like "it's X times faster!"
If they said "this is 1.8x faster on calculating this floating point equation than a Threadripper," ok. That's something we can distinctly say "yeah, holy balls that IS faster!"
Otherwise, I could have said that my old Corolla was 200x faster - without specifying that I was comparing it to a scooter.
If they said "this is 1.8x faster on calculating this floating point equation than a Threadripper," ok. That's something we can distinctly say "yeah, holy balls that IS faster!"
Otherwise, I could have said that my old Corolla was 200x faster - without specifying that I was comparing it to a scooter.
HobeSoundDarryl
macrumors G5
Bring on the chorus of "PER WATT" defenses, as if desktop users are very, VERY concerned about using the same amount of extra power as an old incandescent bulb or two to get much more power.
I purchased Studio with Ultra knowing that those charts were much more likely Marketing team puffery than objective reality. Why? Because my needs are not maximum power but Apple functionality-centric... and this is THE way to maximize the latter.
However, I'm glad there is real competition and that Apple will feel some pressure to try to ACTUALLY live up to puffery claims. That will spur on further advancements in M2, M3 and so on. Competition is great here. If Apple actually won all contests in their first try, they could potentially coast for a while going forward (see Siri for example).
Go PC hardware advancements, go! Apple will feel pressure to innovate a way to actually win such contests measured objectively and PCs will feel some pressure to find a way to use less power while staying ahead. Somewhere a few years down the road, perhaps BOTH camps end up winning by being pushed by the other? Nothing at all wrong with win:win.
Bonus for the "working Mac" crowd: now that Bootcamp seems doomed on Silicon, those of us who must be able to run Windows too face the reality of probably needing to buy a PC when our Intel Macs conk. PC hardware makers trying to "beat Apple" means we can get a fantastic Windows machine to revive Bootcamp the old fashioned way.
I purchased Studio with Ultra knowing that those charts were much more likely Marketing team puffery than objective reality. Why? Because my needs are not maximum power but Apple functionality-centric... and this is THE way to maximize the latter.
However, I'm glad there is real competition and that Apple will feel some pressure to try to ACTUALLY live up to puffery claims. That will spur on further advancements in M2, M3 and so on. Competition is great here. If Apple actually won all contests in their first try, they could potentially coast for a while going forward (see Siri for example).
Go PC hardware advancements, go! Apple will feel pressure to innovate a way to actually win such contests measured objectively and PCs will feel some pressure to find a way to use less power while staying ahead. Somewhere a few years down the road, perhaps BOTH camps end up winning by being pushed by the other? Nothing at all wrong with win:win.
Bonus for the "working Mac" crowd: now that Bootcamp seems doomed on Silicon, those of us who must be able to run Windows too face the reality of probably needing to buy a PC when our Intel Macs conk. PC hardware makers trying to "beat Apple" means we can get a fantastic Windows machine to revive Bootcamp the old fashioned way.
Last edited:
It's not surprising that their charts are wrong, but a 50% difference is large enough that it indicates to me there might be some issue with either the shipped devices or the benchmark.
This is the kind of thing that breaks trust.
Not to mention that the RTX3090 price is literally 1/3 of a max out Studio Ultra. That if you are lucky enough to find one at $2600.
The Verge is not very good at benchmarks, and, they talk about that specific Geekbench 5 Compute one, @MacRumors I would wait for the Anandtech review with multiple tests and data points before assessing how the M1 Ultra and the RTX 3090 stand relatively to each other. As someone mentioned, in WoW the M1 Ultra beats it the RTX 3090 by a few fps, WoW being actually optimized for AS...
Of course, Apple should've specified which benchmark they used for that chart, as I'm sure they weren't lying per se.
Of course, Apple should've specified which benchmark they used for that chart, as I'm sure they weren't lying per se.
Compete with what and more important for whom?
The 25.6 non-miners that got a 3090?
Most PC users neither need or can afford that kind of power and would be quite happy with the power of the 2 base configs. Try to match (12th gen Intel, DDR5 and mid to upper tier GPU) and actual retail prices won't be that far off from what Apple asks.
I also don't think "noone care about power consumption" is true. Electricity rates can be pretty high in some places and when you're in an area where you need to run AC against that heat output it adds up fast even on low rates.
Also heat==noise and a quiet computer is a value of it's own.
So yeah, Apple can truly compete TODAY. Wether they will be able/willing to scale up to the highest tier is to be seen.
Let me see if I am getting this interpretation. The competitor, when burning through “unlimited” power, i.e., 500 watts, outperforms the Mac Studio at under 120 watts, and that is hailed as a victory?
How nVidia could be anything other than concerned / embarrassed by this trend is unclear to me, especially since this is Apple’s first generation of their desktop SOC.
How nVidia could be anything other than concerned / embarrassed by this trend is unclear to me, especially since this is Apple’s first generation of their desktop SOC.
Edit: It has come to light that geekbench is not an accurate benchmark for the Ultra, hence the low score. This was actually already known for the M1 Max. The Max 32 core and Ultra are designed to ramp up more slowly; too slowly for the fast Geekbench process. The truth is that games are showing within a few fps to the 3090. Not much else to say.
Last edited:
i think the CPU cooling system they had in the current Mac Pro is fine for any extreme chips.Won’t be long before they start using liquid cooled systems to compete.
In Geekbench. That seems to be as far as the comparison goes. We know Apple wasn't using that benchmark for their marketing.
This is incredibly disheartening. I was really looking forward to Apple finally toppling the powerhouse, and it looks like they fell short.
Yes, watt-efficient computing is cool and cost-effective. But when you need performance, you don't care that your system is drawing more power. You just get the task done.
Yes, watt-efficient computing is cool and cost-effective. But when you need performance, you don't care that your system is drawing more power. You just get the task done.
The surprising part to me is that the fastest graphics card in the world is only twice as fast as Apple's system-on-chip graphics.
That anyone could have taken that chart seriously, what with a y-axis specifying a nebulous variable of "relative performance", is beyond me. Kudos to Apple's marketing department; anyone who thought that graph showed anything relevant has probably purchased a bridge or two.
You need to buy a board, RAM, CPU, power supply, storage, windows license, case, cooling too.
Also need air con because you’re going to have 600W+ fan heater in your room.
Even my old Ryzen 3700X + 1660 GTX gaming PC made a British summer very uncomfortable.
To note it’s probably cheaper to rent a 3090 box from a cloud provider if you aren’t running one constantly for compute. If you want to play games buy something cheaper and cooler.
Apple does list the models they compete against in the fine print at the bottom of the charts. In the case of that chart, the competition was a PC with Intel Core i9-12900K using DDR5 memory and equipped with a 3090.
M1 Ultra Mac studio isn’t anywhere near the theoretical double performance on M1 Max mac studio. 2x M1 max shouldn’t result in a 25% increase in gpu performance. Something seems off IMO, but I’m far from an expert
Yeah sound’s like Apple.
I remember the fake charts of Apple PowerPC vs Intel.
This also explains why even simple Terminal Software like Midnight Commander (installed over brew.sh) takes so long to start on AppleCPU, compared to an old 2014 IntelMac or Linux.
Same for the Apples own Activity Monitor, on IntelMac its like POW there, on AppleCPU it takes ages to show all the processes.
I remember the fake charts of Apple PowerPC vs Intel.
This also explains why even simple Terminal Software like Midnight Commander (installed over brew.sh) takes so long to start on AppleCPU, compared to an old 2014 IntelMac or Linux.
Same for the Apples own Activity Monitor, on IntelMac its like POW there, on AppleCPU it takes ages to show all the processes.