Geekbench Blog
www.geekbench.com
Weird choice of baseline CPU and even weird is that the baseline score is 2500.
i7-12700 does hardly 2000 in GB5 with the fastest DDR5.
Not saying I don't believe you, but source? And are there numbers for M3 as well? Just curious.One M4 P core + $L2 is about 4.06mm2
Not saying I don't believe you, but source? And are there numbers for M3 as well? Just curious.
The m4 tested with no cooling at all still gets approx 3750 in geekbench 6.3. They were trying to get the highest result possible and one test managed to get just over 4k with ln2 being poured over it. When macs come out, it would be easy to get that score with fans. 12900k in comparison scores approx 2700.It has nothing to do with intel cpus and the fans are indeed at 0 rpm.
The m4 was tested under ln2 and still lost. So even though it's unfair that I didn't have Ln2 to use as well, it's fine, don't need it.
So as I imagined an M4 pcore is larger in transistor count vs an intel pcore, isn't it?
Of course transistor count is what matters when you are trying to compare CPUs with different architectures, instructions and nodes. What can ARM achieve with X amount of transistors vs what an x86 CPU can achieve with X amount of transistors. With the data we currently have a 3 year old x86 beats the latest and great arm by around 40% while having less than half the transistor count. At least in GB6. If you wanna compare something else, sure, just ask and ill have no problem testing it for you.The m4 tested with no cooling at all still gets approx 3750 in geekbench 6.3. They were trying to get the highest result possible and one test managed to get just over 4k with ln2 being poured over it. When macs come out, it would be easy to get that score with fans. 12900k in comparison scores approx 2700.
I have no ideas your obsession with transistor counts. It is a non-issue. If a CPU manufacturer can double speed with double transistor count for a core, they would be over the moon. It is just not a realistic scenario. If you really believe this, you have absolutely no clue about CPU design.
FYI - I really think you are trying to compete with Igor in this thread for comedy. Keep it coming.
Show us a screenshot of your diagnostic that the fans were off. In my experience, desktops fans are never turned off unless it is faulty. They might be running at its lowest speced RPM though.It has nothing to do with intel cpus and the fans are indeed at 0 rpm.
Seriously, you've never seen a 0rpm fan? Here you go. And check the fan curve im using on the CPU, it's 0% until 80c or roundabout.Show us a screenshot of your diagnostic that the fans were off. In my experience, desktops fans are never turned off unless it is faulty. They might be running at its lowest speced RPM though.
Ok, sure it has been a while. Interestingly, the system seems to have 4 other fans operating and with a large heat sink. I would hardly call this a passively cooled system.Seriously, you've never seen a 0rpm fan? Here you go. And check the fan curve im using on the CPU, it's 0% until 80c or roundabout.
View attachment 102471
Well - that's the minimum rpm I can run those other fansOk, sure it has been a while. Interestingly, the system seems to have 4 other fans operating and with a large heat sink. I would hardly call this a passively cooled system.
Limited to 45w. Still faster than the m4. Usually the chip wasn't even hitting the 45w power limit - it was sitting between 25-30. Only clang maxed it out.
Micro-Star International Co., Ltd. MS-7D28 - Geekbench
Benchmark results for a Micro-Star International Co., Ltd. MS-7D28 with an Intel Core i9-12900K processor.browser.geekbench.com
Not sure it was running at those frequencies since it was limited to 45 watts.With 8P+4E at 3.6GHz the M2 Max matches your 12900K.Mac Studio (2023) - Geekbench
Benchmark results for a Mac Studio (2023) with an Apple M2 Max processor.browser.geekbench.com
Most of the m2max is the GPU. It also has so much more than just the CPU cores. NPU, secure enclave, display engines, encode engines. Do you not even know this?Not sure it was running at those frequencies since it was limited to 45 watts.
Why do clockspeeds matter though? Performance per transistor count is what matters, clockspeeds and number of cores is kinda irrelevant. If a CPU can be faster while using 10.000 cores, great, got no issue.
EG1. The m2 max is what, 8 times bigger than the 12900k? And that's not disappointing?
So it is not passively cooled then.Well - that's the minimum rpm I can run those other fans
It's not LN2 cooled either, still beats the ln2 cooled m4So it is not passively cooled then.
I've already said it, when the ipad is not cooled at all, the speed difference is less than 8%. LN2 was only to get the highest possible speed in the ipad.It's not LN2 cooled either, still beats the ln2 cooled m4
Sure, sure. For context, a 4080 super + a 12900k is still 17b less transistors than the m2 max. Practically it should be faster than a 4080s + 4x12900ks. Is it?Most of the m2max is the GPU. It also has so much more than just the CPU cores. NPU, secure enclave, display engines, encode engines. Do you not even know this?
Still a lot of thing missing even with adding the 4080 and 12900k. And uses approx 10X more power under load.Sure, sure. For context, a 4080 super + a 12900k is still 17b less transistors than the m2 max. Practically it should be faster than a 4080s + 4x12900ks. Is it?
What workload is the 4080 and 12900k 10x more power? As youve just pointed out yourself at 45w the 900k is matching the m2 max in gb6, no?Still a lot of thing missing even with adding the 4080 and 12900k. And uses approx 10X more power under load.
But the 4080 GPU demolishes the m2Max GPU though.
It's now pretty obvious you are into a crusade to validate your confirmation bias.EG1. The m2 max is what, 8 times bigger than the 12900k? And that's not disappointing?
Ah, now I get it. For the last 23 pages that you thought M4 was winning, GB6 was fine. Now that you realize it's actually losing, it's not fine to use GB6. Kay, no worries.It's now pretty obvious you are into a crusade to validate your confirmation bias.
M2 floorplan: https://www.semianalysis.com/p/apple-m2-die-shot-and-architecture
12900k floorplan: https://www.techpowerup.com/review/intel-core-i9-12900k-alder-lake-12th-gen/2.html
Adding to the pain you compare GB6 MT scores which we know don't scale linearly. Piling up on invalid premises will only lead to incredibly stupid conclusions.
GPUs are a different story. Nvidia leads in perf/w and absolute performance. No one one is catching up to Nvidia this year or next year with the RTX 40/50 series.4080 GPU demolishes the m2Max GPU though.
Who said it was losing? M3max demolishes the 12900k while using much less power. The m4 in an iPad pro that in a fanless 5mm thick chassis has much higher per core performance than the 12900k (or any intel processor). You really have a one track mind and absolute no sense of reality.Ah, now I get it. For the last 23 pages that you thought M4 was winning, GB6 was fine. Now that you realize it's actually losing, it's not fine to use GB6. Kay, no worries.
Eg1. It was you specifically arguing the last 10-15 pages about how great GB6 is against everyone else that says otherwise. Oh gosh
Not sure about performance/watts, but definitely in absolute performance. The m3 gpu made huge advances compared to the m2 especially in compute.GPUs are a different story. Nvidia leads in perf/w and absolute performance. No one one is catching up to Nvidia this year or next year with the RTX 40/50 series.
The m2 max (posted by nothingness above) and the m4 loses by a lot to the 12900k. The m3 max barely scratches a win, which is embarrassing considering it is what, ~19 times bigger in transistor count? 19 freaking times. Let that sink in...Of course any modern high end chip from amd or intel walks all over it, but sure, let's compare it to the 12900k, no problem.Who said it was losing? M3max demolishes the 12900k while using much less power. The m4 in an iPad pro that in a fanless 5mm thick chassis has much higher per core performance than the 12900k (or any intel processor). You really have a one track mind and absolute no sense of reality.