mjtomlin

About

Username
mjtomlin
Joined
Visits
192
Last Active
Roles
member
Points
4,861
Badges
2
Posts
2,699
  • Apple's M1 Max GPU is at least 3x faster than M1, Metal benchmark shows

    KITA said:
    The Apple M1 scores ~18,000 in OpenCL (1) and ~20,000 in Metal (2) on Geekbench.

    The Macbook Pro with M1 Max 32 core scored ~60,000 in OpenCL (3) on Geekbench. The ~68,000 Metal score above is likely the 32 core version as well.

    The RTX 3080 scores much higher than 90,000. A 130W version of the RTX 3080 in a laptop scores ~120,000 in OpenCL (4) on Geekbench.

    Apple used a 165W version of the RTX 3080 in their comparison:




    Apple stopped supporting and optimizing OpenCL (and OpenGL) on Macs a few years ago when they started designing their own GPU's optimized for their Metal APIs. I'm not sure those are good benchmark comparisons unless you're comparing "like" systems.
    tmaymagman1979OutdoorAppDeveloperwilliamlondonwatto_cobra
  • Despite potshots, Intel & Samsung want to make Apple Silicon

    lkrupp said:
    Analysts and big tech are taking notice of Apple Silicon. The race to copy Apple’s designs is probably already afoot.

    It's not that anyone can't design similar chips to Apple's, it's that Apple's designs are very expensive. You need scale and huge profit margins. Each generation of their SoC designs end up in hundreds of millions of "higher end" devices over the course of several years.

    Plus, Intel and AMD have the problem that their CPU's are based on the bloated x86 architecture that's never going reach the efficiency of ARM.
    killroywatto_cobra
  • Apple's M1 Max is 1.5x faster than M1 in supposed benchmark

    crowley said:
    mjtomlin said:
    crowley said:
    red oak said:
    Have to be honest, this benchmark seems low if it is true.   With everything they launched today (impressive),  I thought we would see single core at 2,000 and multi at 14,000+ 

    We'll know the real answer by end of this week 
    Why would the single core score go up?  It's the same M1 cores, just more of them.  You'll need to wait for M2 for the single core needle to move substantially.
    And for multi they went from 6 high performance cores to 8, a 33% increase in the number of cores; so a 50% improvement seems better than you might otherwise expect.

    The real improvements are in the memory capacity and bandwidth, the GPU, and possibly the other processing units for video and neural.

    The M1 has 4 performance cores and 4 efficiency cores. So performance core count doubled in Pro/Max SoC. I would say the GeekBench results are close. But we won’t really know until more people get their hands on these systems and start testing them.

    But yeah, Apple has moved a lot of the heavier tasks off the CPU and onto their own specialized IP, so these systems even with only a 10 core CPU are going to be wicked fast, especially for content creators.
    Of course you're right, I assumed that the regular M1 had 2 efficiency cores just like the Pro and the Max, but it actually has 4.  So they've gone from 4 high performance cores in the M1 to 8 in the M1 Max.  I guess that does make a 50% Geekbench multi-core improvement look a little underwhelming.

    Well Apple stated the performance increase was about 70%, which should put the multi-core numbers around 13,000. The results listed here are within the outlier range normally associated with Geekbench scores. If you look at M1 scores they range from 5,500 to almost 7,800. Lower scores are a result of the CPU being used by other processes while the benchmark was running. Furthermore, CPU performance does not scale linearly.

    That compares close to the 18-core iMac Pro. That CPU has a TDP of 140W. The new M1’s have a TDP of 30W.
    killroymuthuk_vanalingamwatto_cobra
  • The bell is tolling for Intel Macs with the arrival of the first Apple Silicon specific fe...

    safi_cz said:
    swineone said:
    many of the best new macOS Monterey features will only work on Apple Silicon. And so they should.

    Planned obsolescence much? Indeed that's the case as will be clear from the breakdown below.

    [pretend engineering snipped]

    Conclusion: yeah, it's all marketing, and Apple pressuring us into an upgrade we don't really need. So much for being so "green", it's all about the greenbacks (as well it should be, but let's not pretend Apple is a saint).
    Nah, you just made up a bunch of crap and claimed it as fact.
    Swineone is right and you are wrong. It is planned obsolesce and also logical step. Fact: Apple I9 Macbook pro is still the most powerfull Macbook Apple has in its lineup. Even today Apple continues to promote it as a powerful tool for graphic and video professionals. And such pro tool (claimed by Apple)  is not capable to handle real time blur? Or almost real time OCR?  Yes it is. 

    Second - "green environment".  You may not remember, but there were days, when you can change battery in your laptop without changing half of its case. Milled aluminium case. 

    However - Apple would be stupid not to make some new features available only for M1 otherwise Apple would act against its long time strategy. After all - it is company, which primary goal is to make money:-) 


    It is not planned obsolescence. "Planned obsolescence" is when a product has an inherent feature that is designed to fail at some point, making that product unusable and forcing an upgrade. There is no such thing in Intel Macs. Making a new products more desirable by giving them extra features, does not make older products obsolete. All this is, is incentivizing users to move forward and onto new hardware.

    The fact is, the M1 has specialized processors that can perform tasks and process data without affecting the rest of the system. This is what makes Apple's SoC's so powerful and efficient. Apple could very well add all these "features" to Intel Macs, but those systems would probably take a major performance hit and Apple didn't think it was worth wasting the resources for a sub-par experience.


    Fidonet127williamlondonJFC_PAsphericBeatswatto_cobra
  • Spatial Audio head tracking on Apple TV automatically resets when you get up from the couc...

    Beats said:
    What a mess. It calibrates according to the user looking in a direction for a certain amount of time?

    What if user is talking to someone? What if they’re looking down at their iPhone? What if they’re in the kitchen?.....

    If the user is sitting off to the side and not directly in front of the TV will the objects follow their assumed panning? (Admittedly this would be cool).
    For example you’re on the left side of the room. Batmobile speeds off to the left in the movie. It sounds like you’re gonna be ran over. 

    I’ve wanted a TrueDepth camera array for Apple TV for years pre-Spatial Audio and the lost benefits are compounding. The camera would have solved these problems.

    Seriously, man. This is a solution for the rare individual that sits in front of the TV and actually looks at the screen while watching a movie or TV show. Who then may at some point briefly get up or momentarily look away. This isn't meant for Cirque du Soleil performers who just want background noise.

    You sound like one of those people who are constantly asking questions, talking over dialog because you're not paying attention to the movie. The only mess here, is your wild, defeatist imagination. 
    jahbladeBeats