New iPad Pro rumored to debut with M4 chip

2»

Comments

  • Reply 21 of 40
    blastdoorblastdoor Posts: 3,425member
    narwhal said:
    Qualcomm and Microsoft will debut Snapdragon X Elite Windows laptops in May that they claim match the specs of a base model M3. I suspect Apple wants to release a faster iPad before then to take a bit of wind from their sails.
    We’ll see… the M1 Pro beats the m3 on multithreaded benchmarks. And if Apple had been willing to overclock a single core on the m1, it might have been able to match the m3 on bursty single core benchmarks too. But an M1 Pro uses a lot more power and requires more cooling. 

    So if all they’ve done is create something equivalent to an M1 Pro, then they are still behind Apple (and really no better than Intel).
  • Reply 22 of 40
    9secondkox29secondkox2 Posts: 2,902member
    The only way this makes sense is if the Mac Studio or a returning large iMac gets an m4 ultra this year. The Mac Pro with m4 “extreme, Super, hyper” etc. to arrive either this year or next. 

    Otherwise id expect an m3. 

    But with a “rumor” this bold, there just might be merit. We know the original 3nm process had yield issues and isn’t as efficient as could be. The updated process is better all the way around. Makes sense to move to it asap. Also, with the designs being different, makes sense to go ahead with m4 a bit “early.” 

    Ex opted for the possibilities. 
    jellybelly
  • Reply 23 of 40
    MarvinMarvin Posts: 15,402moderator
    blastdoor said:
    narwhal said:
    Qualcomm and Microsoft will debut Snapdragon X Elite Windows laptops in May that they claim match the specs of a base model M3. I suspect Apple wants to release a faster iPad before then to take a bit of wind from their sails.
    We’ll see… the M1 Pro beats the m3 on multithreaded benchmarks. And if Apple had been willing to overclock a single core on the m1, it might have been able to match the m3 on bursty single core benchmarks too. But an M1 Pro uses a lot more power and requires more cooling. 

    So if all they’ve done is create something equivalent to an M1 Pro, then they are still behind Apple (and really no better than Intel).
    According to recent reports, they were lying:

    https://www.semiaccurate.com/2024/04/24/qualcomm-is-cheating-on-their-snapdragon-x-elite-pro-benchmarks/
    https://www.tomshardware.com/tech-industry/qualcomm-faces-benchmark-cheating-allegations-snapdragon-x-eliteplus-benchmarks-claimed-to-be-fraudulent

    OEMs got less than half the performance they claimed. They are also being sued by ARM over their license.
    nubus said:
    It doesn't make sense to ship any AI product before WWDC. There is no new iPadOS/AI to make use of that new AI power.
    Even worse for Air as it will be obsolete on arrival.

    Launching M4 now will make MBA M3 "the last student computer without AI". It is, but Apple shouldn't put back-to-school sales at risk.
    Same for iPhone. We're 4-5 months from next gen. No way that Apple will go full Osborne and kill all shipping products now without new products.

    WWDC should be XCode AI + frameworks using very limited memory + the message that everything is possible with 18 TOPS and 6-8 GB as that is A16/M3.
    A18/A18 Pro/M4 can deliver a new level of performance but iPhone has to be first.
    Re WWDC and need for a new iOS with new AI— wouldn’t the new AI need new hardware to have that ML/AI capability?  And developers need that hardware in hand to try the new code at WWDC.  (I’m finding myself starting to see a logic in this rumor being true)
    New AI software would work on existing hardware and OS, M3 + current OS runs everything just fine. It uses too much memory for mobile though, easily 10GB RAM for basic language models and 20GB+ for image generators and there's no swap on mobile. The higher iPad Pros go up to 16GB RAM but the 8GB models won't handle them without a new memory setup e.g running them from SSD.

    The Neural Engine in M3 is fast enough for local AI and I don't see M4 being able to provide a significant boost for mobile hardware. In laptops, they can ship a 2x Neural Engine because it would only use 20W but there's no cooling in the mobile devices.

    I expect the upcoming iPad Pro to use M3 and the focus being the Pencil. If Apple has AI announcements later, such as local AI model shipped with next iOS, the iPad Pro will run them, just as the existing M3 lineup will.

    I'd expect a small hardware upgrade with M4 like go from 35TOPs to 45TOPs Neural Engine and GPUs get a small boost. The responses running local AI LLMs on M1 is instant. It slows down as the conversation goes on as it has to use the content of the whole conversation but they can just cap it or have some more efficient way to handle it.
    tmaydanox
  • Reply 24 of 40
    mattinozmattinoz Posts: 2,388member
    The only way this makes sense is if the Mac Studio or a returning large iMac gets an m4 ultra this year. The Mac Pro with m4 “extreme, Super, hyper” etc. to arrive either this year or next. 

    Otherwise id expect an m3. 

    But with a “rumor” this bold, there just might be merit. We know the original 3nm process had yield issues and isn’t as efficient as could be. The updated process is better all the way around. Makes sense to move to it asap. Also, with the designs being different, makes sense to go ahead with m4 a bit “early.” 

    Ex opted for the possibilities. 
    If rumours are true and the Ultra(Hidra) is a big single Beast chip, in which Apple feels like the E cores in those desktop machines are going to waste so has cut them along with a single unified larger ML block. Then surely having the Mac Pro and Studio update together makes sense, having both of them shipping binned versions of the Ultra(Hidra). It gives 2 machine options for developers to train new ML models for deployment on the same hardware. With an M4 Ultra they could offer the M3 Max in the lower slot until the M4 Pro/Max die is ready.
    danox
  • Reply 25 of 40
    The Ai-Pad cometh... I'm not clear on who (outside of Apple's Marketing Team) might have asked for this though.
    edited April 28 22july2013williamlondon
  • Reply 26 of 40
    We cannot expect a sleepyhead waking up in the morning to delivery something other techs have worked overnight. Without huge memory plug-ins, Apple AJET LLM could be just a juke.
    williamlondon
  • Reply 27 of 40
    brianusbrianus Posts: 167member

    If the M4 iPad Pro debut is accurate, it would be the shortest cycle of Apple Silicon updates yet, with the gap between releases being only seven months, versus about 10 from M2 to M3.

    Eh? M2 MacBook Air debuted at WWDC '22, and the "scary fast" M3 unveiling was this past Halloween. So that's just shy of 17 months, not 10. The previous cycle was 19 months.  
  • Reply 28 of 40
    entropysentropys Posts: 4,221member
    Seems a bit far fetched, although if it was to truly take advantage of ML/AI features that require the new chip can understand it.
    williamlondon
  • Reply 29 of 40
    I suppose any chance of the iPad mini getting the M4 is nonexistent..? 
    Mini is my favorite iPad.
  • Reply 30 of 40
    danvmdanvm Posts: 1,449member
    Marvin said:
    blastdoor said:
    narwhal said:
    Qualcomm and Microsoft will debut Snapdragon X Elite Windows laptops in May that they claim match the specs of a base model M3. I suspect Apple wants to release a faster iPad before then to take a bit of wind from their sails.
    We’ll see… the M1 Pro beats the m3 on multithreaded benchmarks. And if Apple had been willing to overclock a single core on the m1, it might have been able to match the m3 on bursty single core benchmarks too. But an M1 Pro uses a lot more power and requires more cooling. 

    So if all they’ve done is create something equivalent to an M1 Pro, then they are still behind Apple (and really no better than Intel).
    According to recent reports, they were lying:

    https://www.semiaccurate.com/2024/04/24/qualcomm-is-cheating-on-their-snapdragon-x-elite-pro-benchmarks/
    https://www.tomshardware.com/tech-industry/qualcomm-faces-benchmark-cheating-allegations-snapdragon-x-eliteplus-benchmarks-claimed-to-be-fraudulent

    OEMs got less than half the performance they claimed. They are also being sued by ARM over their license.
    From what I know, it's just one report that claim that Qualcomm is lying.  The second link you posted makes reference to the www.semiaccurate.com article.  I'm looking forward seeing what the real results are.  Based in recent benchmarks, most results have been positive.  We'll see what happens.  
  • Reply 31 of 40
    blastdoorblastdoor Posts: 3,425member
    danvm said:
    Marvin said:
    blastdoor said:
    narwhal said:
    Qualcomm and Microsoft will debut Snapdragon X Elite Windows laptops in May that they claim match the specs of a base model M3. I suspect Apple wants to release a faster iPad before then to take a bit of wind from their sails.
    We’ll see… the M1 Pro beats the m3 on multithreaded benchmarks. And if Apple had been willing to overclock a single core on the m1, it might have been able to match the m3 on bursty single core benchmarks too. But an M1 Pro uses a lot more power and requires more cooling. 

    So if all they’ve done is create something equivalent to an M1 Pro, then they are still behind Apple (and really no better than Intel).
    According to recent reports, they were lying:

    https://www.semiaccurate.com/2024/04/24/qualcomm-is-cheating-on-their-snapdragon-x-elite-pro-benchmarks/
    https://www.tomshardware.com/tech-industry/qualcomm-faces-benchmark-cheating-allegations-snapdragon-x-eliteplus-benchmarks-claimed-to-be-fraudulent

    OEMs got less than half the performance they claimed. They are also being sued by ARM over their license.
    From what I know, it's just one report that claim that Qualcomm is lying.  The second link you posted makes reference to the www.semiaccurate.com article.  I'm looking forward seeing what the real results are.  Based in recent benchmarks, most results have been positive.  We'll see what happens.  
    I bet they aren’t “lying” to the extent semi accurate claims. 

    But I suspect they are misleading in the sense I described, in which they compare a chip with transistor counts and thermals more like an m1pro (or m2pro) to a vanilla m3. 

    Another thought — nuvia was working on a server chip. I wonder if they added SMT? If so, it would help with multithreaded benchmarks but it would increase power consumption. Also, if qcom’s benchmarks were run with SMT but the OEMs didn’t have SMT then that could be a source of discrepancy 
  • Reply 32 of 40
    charlesncharlesn Posts: 939member
    I suppose any chance of the iPad mini getting the M4 is nonexistent..? 
    What's "nonexistent" is the need for an M4 in an iPad Mini -- like, seriously, what are you doing on a Mini that would require that kind of power? The small screen size dictates its use primarily for media consumption, web browsing, emailing, etc. That Apple doesn't even bother with an accessory keyboard for the Mini tells you a lot. Not that you literally can't do anything else on it--the current A15 chip is actually quite powerful--but the small screen makes other types of work a real chore, especially in iPadOS where multitasking is STILL a kludge. I supposed Apple might drop an M-chip into the Mini's next iteration if it standardizes on M-chips across the whole iPad line, especially if AI capabilities require it--or we may continue to see the entry level iPad and Mini running on an A-chip. 
    edited April 29 williamlondon
  • Reply 33 of 40
    JFC_PAJFC_PA Posts: 937member
    Could be fun, my iPad is a goto device for me. 
  • Reply 34 of 40
    MarvinMarvin Posts: 15,402moderator
    blastdoor said:
    danvm said:
    Marvin said:
    blastdoor said:
    narwhal said:
    Qualcomm and Microsoft will debut Snapdragon X Elite Windows laptops in May that they claim match the specs of a base model M3. I suspect Apple wants to release a faster iPad before then to take a bit of wind from their sails.
    We’ll see… the M1 Pro beats the m3 on multithreaded benchmarks. And if Apple had been willing to overclock a single core on the m1, it might have been able to match the m3 on bursty single core benchmarks too. But an M1 Pro uses a lot more power and requires more cooling. 

    So if all they’ve done is create something equivalent to an M1 Pro, then they are still behind Apple (and really no better than Intel).
    According to recent reports, they were lying:

    https://www.semiaccurate.com/2024/04/24/qualcomm-is-cheating-on-their-snapdragon-x-elite-pro-benchmarks/
    https://www.tomshardware.com/tech-industry/qualcomm-faces-benchmark-cheating-allegations-snapdragon-x-eliteplus-benchmarks-claimed-to-be-fraudulent

    OEMs got less than half the performance they claimed. They are also being sued by ARM over their license.
    From what I know, it's just one report that claim that Qualcomm is lying.  The second link you posted makes reference to the www.semiaccurate.com article.  I'm looking forward seeing what the real results are.  Based in recent benchmarks, most results have been positive.  We'll see what happens.  
    I bet they aren’t “lying” to the extent semi accurate claims. 

    But I suspect they are misleading in the sense I described, in which they compare a chip with transistor counts and thermals more like an m1pro (or m2pro) to a vanilla m3.
    When they say they have a market-leading ARM chip by comparing a much higher TDP chip to Apple's lowest one, that's lying.

    Their TDP for the Elite chip is 80W, which is Max-level power usage.

    Benchmarks for GFXBench Aztec Normal offscreen:
    Snapdragon X Plus = 136FPS (~25W)
    Snapdragon X Elite = 312FPS (up to 80W)

    M3 = 323FPS (20W)
    M3 Pro = 579FPS (30W)
    M3 Max = 867FPS (60W)

    Baldur's Gate 3 test, they show the Elite running at 30-40FPS, M3 Max can run this at 90FPS.

    Their Plus chip is half M3 and their Max power chip is performing like M3. Their own performance-per-watt tests showed 50-60% better than Intel. Apple is 4-5x better than Intel on performance-per-watt.

    This is the kind of thing Intel does where they put more CPU cores in, ramp up the power usage then compare it to Apple's weakest fanless chips.

    Snapdragon chips will be competitive with entry-level Intel notebooks but so is a potato. It's well behind even last-gen Pro Macs and Apple will soon launch their next-gen version.
    williamlondon
  • Reply 35 of 40
    KITAKITA Posts: 409member
    Marvin said:
    blastdoor said:
    danvm said:
    Marvin said:
    blastdoor said:
    narwhal said:
    Qualcomm and Microsoft will debut Snapdragon X Elite Windows laptops in May that they claim match the specs of a base model M3. I suspect Apple wants to release a faster iPad before then to take a bit of wind from their sails.
    We’ll see… the M1 Pro beats the m3 on multithreaded benchmarks. And if Apple had been willing to overclock a single core on the m1, it might have been able to match the m3 on bursty single core benchmarks too. But an M1 Pro uses a lot more power and requires more cooling. 

    So if all they’ve done is create something equivalent to an M1 Pro, then they are still behind Apple (and really no better than Intel).
    According to recent reports, they were lying:

    https://www.semiaccurate.com/2024/04/24/qualcomm-is-cheating-on-their-snapdragon-x-elite-pro-benchmarks/
    https://www.tomshardware.com/tech-industry/qualcomm-faces-benchmark-cheating-allegations-snapdragon-x-eliteplus-benchmarks-claimed-to-be-fraudulent

    OEMs got less than half the performance they claimed. They are also being sued by ARM over their license.
    From what I know, it's just one report that claim that Qualcomm is lying.  The second link you posted makes reference to the www.semiaccurate.com article.  I'm looking forward seeing what the real results are.  Based in recent benchmarks, most results have been positive.  We'll see what happens.  
    I bet they aren’t “lying” to the extent semi accurate claims. 

    But I suspect they are misleading in the sense I described, in which they compare a chip with transistor counts and thermals more like an m1pro (or m2pro) to a vanilla m3.
    When they say they have a market-leading ARM chip by comparing a much higher TDP chip to Apple's lowest one, that's lying.

    Their TDP for the Elite chip is 80W, which is Max-level power usage.

    Benchmarks for GFXBench Aztec Normal offscreen:
    Snapdragon X Plus = 136FPS (~25W)
    Snapdragon X Elite = 312FPS (up to 80W)

    M3 = 323FPS (20W)
    M3 Pro = 579FPS (30W)
    M3 Max = 867FPS (60W)

    Baldur's Gate 3 test, they show the Elite running at 30-40FPS, M3 Max can run this at 90FPS.

    Their Plus chip is half M3 and their Max power chip is performing like M3. Their own performance-per-watt tests showed 50-60% better than Intel. Apple is 4-5x better than Intel on performance-per-watt.

    This is the kind of thing Intel does where they put more CPU cores in, ramp up the power usage then compare it to Apple's weakest fanless chips.

    Snapdragon chips will be competitive with entry-level Intel notebooks but so is a potato. It's well behind even last-gen Pro Macs and Apple will soon launch their next-gen version.

    From Anandtech (Oct. 2023)

    "Thin and Light" version has a 23W TDP:

    QSXE

    Also, I find it odd that they're showing off old smartphone benchmarks anyways.

    NUVIA cores aside, Qualcomm tends to have pretty good graphics for their SoCs:

    image

    Solar Bay tests modern graphics features such as ray tracing - the peak score of the Snapdragon 8 Gen 3 (smartphone SoC) is close behind the Apple M2.
    williamlondon
  • Reply 36 of 40
    danvmdanvm Posts: 1,449member
    blastdoor said:
    danvm said:
    Marvin said:
    blastdoor said:
    narwhal said:
    Qualcomm and Microsoft will debut Snapdragon X Elite Windows laptops in May that they claim match the specs of a base model M3. I suspect Apple wants to release a faster iPad before then to take a bit of wind from their sails.
    We’ll see… the M1 Pro beats the m3 on multithreaded benchmarks. And if Apple had been willing to overclock a single core on the m1, it might have been able to match the m3 on bursty single core benchmarks too. But an M1 Pro uses a lot more power and requires more cooling. 

    So if all they’ve done is create something equivalent to an M1 Pro, then they are still behind Apple (and really no better than Intel).
    According to recent reports, they were lying:

    https://www.semiaccurate.com/2024/04/24/qualcomm-is-cheating-on-their-snapdragon-x-elite-pro-benchmarks/
    https://www.tomshardware.com/tech-industry/qualcomm-faces-benchmark-cheating-allegations-snapdragon-x-eliteplus-benchmarks-claimed-to-be-fraudulent

    OEMs got less than half the performance they claimed. They are also being sued by ARM over their license.
    From what I know, it's just one report that claim that Qualcomm is lying.  The second link you posted makes reference to the www.semiaccurate.com article.  I'm looking forward seeing what the real results are.  Based in recent benchmarks, most results have been positive.  We'll see what happens.  
    I bet they aren’t “lying” to the extent semi accurate claims. 

    But I suspect they are misleading in the sense I described, in which they compare a chip with transistor counts and thermals more like an m1pro (or m2pro) to a vanilla m3. 

    Another thought — nuvia was working on a server chip. I wonder if they added SMT? If so, it would help with multithreaded benchmarks but it would increase power consumption. Also, if qcom’s benchmarks were run with SMT but the OEMs didn’t have SMT then that could be a source of discrepancy 
    I wouldn't say they are lying, but they can manipulate data and benchmarks to their benefit. Even Apple did this when they compared the M1 Ultra GPU with the Nvidia RTX 3090, and with the M2 processors.
    M1 Ultra GPU comparison with Nvidia was misleading - 9to5Mac
    Don't Believe the Hype: Apple's M2 GPU is No Game Changer | Tom's Hardware (tomshardware.com)

    At the end, we'll have to wait to see how Qualcomm processors perform in real life scenarios.  I think devices will be available this summer, so we'll see how it goes.  
    edited April 29
  • Reply 37 of 40
    brianusbrianus Posts: 167member
    charlesn said:
    but the small screen makes other types of work a real chore, especially in iPadOS where multitasking is STILL a kludge.
    Sorry, you’re just wrong. People with this opinion have simply not given iPad multitasking a chance. It is vastly superior to macOS. It should not work like a traditional desktop, which would be a step down. 

    The issues with “real work” on iPads have more to do with “dumbed down” apps offering less than desktop functionality, and some issues with the way the OS handles (or randomly drops) background tasks. Some aspects of input in the touch interface can be trickier, slower and less precise than on a Mac, even with a keyboard and mouse. But that’s about it.
    tht
  • Reply 38 of 40
    danoxdanox Posts: 3,096member
    danvm said:
    blastdoor said:
    danvm said:
    Marvin said:
    blastdoor said:
    narwhal said:
    Qualcomm and Microsoft will debut Snapdragon X Elite Windows laptops in May that they claim match the specs of a base model M3. I suspect Apple wants to release a faster iPad before then to take a bit of wind from their sails.
    We’ll see… the M1 Pro beats the m3 on multithreaded benchmarks. And if Apple had been willing to overclock a single core on the m1, it might have been able to match the m3 on bursty single core benchmarks too. But an M1 Pro uses a lot more power and requires more cooling. 

    So if all they’ve done is create something equivalent to an M1 Pro, then they are still behind Apple (and really no better than Intel).
    According to recent reports, they were lying:

    https://www.semiaccurate.com/2024/04/24/qualcomm-is-cheating-on-their-snapdragon-x-elite-pro-benchmarks/
    https://www.tomshardware.com/tech-industry/qualcomm-faces-benchmark-cheating-allegations-snapdragon-x-eliteplus-benchmarks-claimed-to-be-fraudulent

    OEMs got less than half the performance they claimed. They are also being sued by ARM over their license.
    From what I know, it's just one report that claim that Qualcomm is lying.  The second link you posted makes reference to the www.semiaccurate.com article.  I'm looking forward seeing what the real results are.  Based in recent benchmarks, most results have been positive.  We'll see what happens.  
    I bet they aren’t “lying” to the extent semi accurate claims. 

    But I suspect they are misleading in the sense I described, in which they compare a chip with transistor counts and thermals more like an m1pro (or m2pro) to a vanilla m3. 

    Another thought — nuvia was working on a server chip. I wonder if they added SMT? If so, it would help with multithreaded benchmarks but it would increase power consumption. Also, if qcom’s benchmarks were run with SMT but the OEMs didn’t have SMT then that could be a source of discrepancy 
    I wouldn't say they are lying, but they can manipulate data and benchmarks to their benefit. Even Apple did this when they compared the M1 Ultra GPU with the Nvidia RTX 3090, and with the M2 processors.
    M1 Ultra GPU comparison with Nvidia was misleading - 9to5Mac
    Don't Believe the Hype: Apple's M2 GPU is No Game Changer | Tom's Hardware (tomshardware.com)

    At the end, we'll have to wait to see how Qualcomm processors perform in real life scenarios.  I think devices will be available this summer, so we'll see how it goes.  

    A new post to GeekBench: Qualcomm ultimately is at the mercy of Microsoft these numbers if true puts its performance just ahead of the last Intel based 27 inch iMac from 2019-2020 and just behind a Apple Silicon MacBook Pro (13-inch Late 2020).

    Note: There are 40 Apple Silicon devices ahead of the new Snapdragon X Elite - X1E80100 - Qualcomm Oryon CPU

    40th place single core 2302 Single-Core Score 8105 Multi-Core Score MacBook Pro Late 2020 13 inch laptop 1 Processor, 8 Cores

    41st place single core 1765 Single-Core Score 4367 Multi-Core Score Qualcomm new chip inside a Dell Inspiron 14 Plus 7441

    https://browser.geekbench.com/mac-benchmarks Apple devices 40 devices ahead the one just ahead was made 4-5 years ago (note: the Apple M1 SOC is 8 cores and the new Qualcomm is 12 cores)…..

    https://browser.geekbench.com/v6/cpu/5889224  Qualcomm Oryon CPU 1 Processor, 12 Cores Uploaded April 28 2024 09:07 AM

    https://www.techpowerup.com/319630/qualcomm-snapdragon-x-elite-x1e80100-cpu-gets-geekbenched  What machine is it? Another test Intel, AMD, Qualcomm going at it in the comments.

    edited April 30
  • Reply 39 of 40
    KITAKITA Posts: 409member
    danox said:
    danvm said:
    blastdoor said:
    danvm said:
    Marvin said:
    blastdoor said:
    narwhal said:
    Qualcomm and Microsoft will debut Snapdragon X Elite Windows laptops in May that they claim match the specs of a base model M3. I suspect Apple wants to release a faster iPad before then to take a bit of wind from their sails.
    We’ll see… the M1 Pro beats the m3 on multithreaded benchmarks. And if Apple had been willing to overclock a single core on the m1, it might have been able to match the m3 on bursty single core benchmarks too. But an M1 Pro uses a lot more power and requires more cooling. 

    So if all they’ve done is create something equivalent to an M1 Pro, then they are still behind Apple (and really no better than Intel).
    According to recent reports, they were lying:

    https://www.semiaccurate.com/2024/04/24/qualcomm-is-cheating-on-their-snapdragon-x-elite-pro-benchmarks/
    https://www.tomshardware.com/tech-industry/qualcomm-faces-benchmark-cheating-allegations-snapdragon-x-eliteplus-benchmarks-claimed-to-be-fraudulent

    OEMs got less than half the performance they claimed. They are also being sued by ARM over their license.
    From what I know, it's just one report that claim that Qualcomm is lying.  The second link you posted makes reference to the www.semiaccurate.com article.  I'm looking forward seeing what the real results are.  Based in recent benchmarks, most results have been positive.  We'll see what happens.  
    I bet they aren’t “lying” to the extent semi accurate claims. 

    But I suspect they are misleading in the sense I described, in which they compare a chip with transistor counts and thermals more like an m1pro (or m2pro) to a vanilla m3. 

    Another thought — nuvia was working on a server chip. I wonder if they added SMT? If so, it would help with multithreaded benchmarks but it would increase power consumption. Also, if qcom’s benchmarks were run with SMT but the OEMs didn’t have SMT then that could be a source of discrepancy 
    I wouldn't say they are lying, but they can manipulate data and benchmarks to their benefit. Even Apple did this when they compared the M1 Ultra GPU with the Nvidia RTX 3090, and with the M2 processors.
    M1 Ultra GPU comparison with Nvidia was misleading - 9to5Mac
    Don't Believe the Hype: Apple's M2 GPU is No Game Changer | Tom's Hardware (tomshardware.com)

    At the end, we'll have to wait to see how Qualcomm processors perform in real life scenarios.  I think devices will be available this summer, so we'll see how it goes.  

    A new post to GeekBench: Qualcomm ultimately is at the mercy of Microsoft these numbers if true puts its performance just ahead of the last Intel based 27 inch iMac from 2019-2020 and just behind a Apple Silicon MacBook Pro (13-inch Late 2020).

    Note: There are 40 Apple Silicon devices ahead of the new Snapdragon X Elite - X1E80100 - Qualcomm Oryon CPU

    40th place single core 2302 Single-Core Score 8105 Multi-Core Score MacBook Pro Late 2020 13 inch laptop 1 Processor, 8 Cores

    41st place single core 1765 Single-Core Score 4367 Multi-Core Score Qualcomm new chip inside a Dell Inspiron 14 Plus 7441

    https://browser.geekbench.com/mac-benchmarks Apple devices 40 devices ahead the one just ahead was made 4-5 years ago (note: the Apple M1 SOC is 8 cores and the new Qualcomm is 12 cores)…..

    https://browser.geekbench.com/v6/cpu/5889224  Qualcomm Oryon CPU 1 Processor, 12 Cores Uploaded April 28 2024 09:07 AM

    https://www.techpowerup.com/319630/qualcomm-snapdragon-x-elite-x1e80100-cpu-gets-geekbenched  What machine is it? Another test Intel, AMD, Qualcomm going at it in the comments.

    "these numbers if true"

    Considering the Snapdragon 8cx Gen 3 (early 2022) running Windows 11 scores ~1600 single core and ~6400 multi core, I'm just going to say, that's probably not true.

    Qualcomm's last smartphone SoC, the Snapdragon 8 Gen 3 (early 2024), scores 2169 single core and 6773 multi core.
  • Reply 40 of 40
    danvmdanvm Posts: 1,449member
    danox said:
    danvm said:
    blastdoor said:
    danvm said:
    Marvin said:
    blastdoor said:
    narwhal said:
    Qualcomm and Microsoft will debut Snapdragon X Elite Windows laptops in May that they claim match the specs of a base model M3. I suspect Apple wants to release a faster iPad before then to take a bit of wind from their sails.
    We’ll see… the M1 Pro beats the m3 on multithreaded benchmarks. And if Apple had been willing to overclock a single core on the m1, it might have been able to match the m3 on bursty single core benchmarks too. But an M1 Pro uses a lot more power and requires more cooling. 

    So if all they’ve done is create something equivalent to an M1 Pro, then they are still behind Apple (and really no better than Intel).
    According to recent reports, they were lying:

    https://www.semiaccurate.com/2024/04/24/qualcomm-is-cheating-on-their-snapdragon-x-elite-pro-benchmarks/
    https://www.tomshardware.com/tech-industry/qualcomm-faces-benchmark-cheating-allegations-snapdragon-x-eliteplus-benchmarks-claimed-to-be-fraudulent

    OEMs got less than half the performance they claimed. They are also being sued by ARM over their license.
    From what I know, it's just one report that claim that Qualcomm is lying.  The second link you posted makes reference to the www.semiaccurate.com article.  I'm looking forward seeing what the real results are.  Based in recent benchmarks, most results have been positive.  We'll see what happens.  
    I bet they aren’t “lying” to the extent semi accurate claims. 

    But I suspect they are misleading in the sense I described, in which they compare a chip with transistor counts and thermals more like an m1pro (or m2pro) to a vanilla m3. 

    Another thought — nuvia was working on a server chip. I wonder if they added SMT? If so, it would help with multithreaded benchmarks but it would increase power consumption. Also, if qcom’s benchmarks were run with SMT but the OEMs didn’t have SMT then that could be a source of discrepancy 
    I wouldn't say they are lying, but they can manipulate data and benchmarks to their benefit. Even Apple did this when they compared the M1 Ultra GPU with the Nvidia RTX 3090, and with the M2 processors.
    M1 Ultra GPU comparison with Nvidia was misleading - 9to5Mac
    Don't Believe the Hype: Apple's M2 GPU is No Game Changer | Tom's Hardware (tomshardware.com)

    At the end, we'll have to wait to see how Qualcomm processors perform in real life scenarios.  I think devices will be available this summer, so we'll see how it goes.  

    A new post to GeekBench: Qualcomm ultimately is at the mercy of Microsoft these numbers if true puts its performance just ahead of the last Intel based 27 inch iMac from 2019-2020 and just behind a Apple Silicon MacBook Pro (13-inch Late 2020).

    Note: There are 40 Apple Silicon devices ahead of the new Snapdragon X Elite - X1E80100 - Qualcomm Oryon CPU

    40th place single core 2302 Single-Core Score 8105 Multi-Core Score MacBook Pro Late 2020 13 inch laptop 1 Processor, 8 Cores

    41st place single core 1765 Single-Core Score 4367 Multi-Core Score Qualcomm new chip inside a Dell Inspiron 14 Plus 7441

    https://browser.geekbench.com/mac-benchmarks Apple devices 40 devices ahead the one just ahead was made 4-5 years ago (note: the Apple M1 SOC is 8 cores and the new Qualcomm is 12 cores)…..

    https://browser.geekbench.com/v6/cpu/5889224  Qualcomm Oryon CPU 1 Processor, 12 Cores Uploaded April 28 2024 09:07 AM

    https://www.techpowerup.com/319630/qualcomm-snapdragon-x-elite-x1e80100-cpu-gets-geekbenched  What machine is it? Another test Intel, AMD, Qualcomm going at it in the comments.

    There are other Geekbench results with better numbers than the one you posted,
    Geekbench Search - Geekbench

    Looks like these numbers are from the Surface Laptop 6 that will be released soon. The results are very different from the Dell results you posted.  We'll have to wait for the release of the devices to see how they really perform.  
    edited May 1
Sign In or Register to comment.