3 ms·

> PC cameras are comical in comparison. PCs usually do not have cameras. Laptops do, sometimes. But they are not meant to take photos - would you carry your la

by onli 6y ago

> PC cameras are comical in comparison. PCs usually do not have cameras. Laptops do, sometimes. But they are not meant to take photos - would you carry your laptop around for that? The article does not get the facts right. There is huge R&D investment on the side of the big PC hardware manufacturers, just look at the big ones, AMD, Intel and Nvidia. There are also really nice Windows/Linux laptops. And great peripherals, with big changes over the years, just compare modern cases like a Lancool II 215 (which isn't even expensive) to what was common a decade ago. And of course the PC market did not grow as much as the smartphone market in the last decade - smartphones started from zero (well, add a few years), while many people that need a laptop or a PC do have one already. Nonetheless the available market numbers for PC components and systems are not as bad as articles like this make it seem - and that was true even before the pandemic. Writers like this just always forget that when they want to paint a negative picture of PCs in favour of smartphones/tablets or when there is another round of "consoles will end PC gaming", which was never true in the slightest. There is no need for a closed off hardware platform like the Mac to save the PC.

EricE 6y ago
"There is huge R&D investment on the side of the big PC hardware manufacturers, just look at the big ones, AMD, Intel and Nvidia." Those are the only significant sources of R&D for PC hardware. None of the big vendors - Dell, HP, Acer are NOT doing basic R&D. Crap, HP is where good technology goes to die or whither away. All the stuff from DEC flushed down the toilet including Alpha (amazing processor - loved our NT Alpha systems back in the day), StrongARM (how ironic), Compaq (Still have my Personal Juke Box PJB-100 - first hard drive MP3 player with a solid lead on Creative and their awful thing, years before Apple and the iPod[and the PJB guys came from DEC now that I think about it]), Palm - don't get me started on Web OS. Itanium? Talk about making a bad partner choice, Intel. Didn't pay attention with Alpha? Then again Intel got StrongARM from HP and then pissed it away too. But I digress... The problem with the PC vendors is they are all pulling parts out of the same bin of least-common-denominator parts. Apple no longer is. We haven't seen the chips aimed at desktop use where battery life and thermal management will be far more flexible. I don't think Apple will focus on one single aspect like single core performance (which they are crushing all but AMD's newest with a 10 watt mobile part - 10 watts vs 50 watt desktop parts!) but overall system integration. Hardware accelerators. Neural Engine. ML cores. Stuff like that. Stuff we haven't even thought of yet. There is much speculation that Rosetta 2 got specialized hardware assistance which explains its ridiculous performance, esp. compared to the original Rosetta. Why not? Apple owns every transistor. M1 runs Intel code faster than the machines they replace. Apple can add what they want - and leave out what they don't need. Not just in the CPU but ANYWHERE in the system now. I think people are dramatically underestimating just how powerful having full vertical control really is. I suspect we are going to see it in a BIG way when they do release the desktops. Remember they really didn't need to do this processor transition; the only reason to totally disrupt their ecosystem is if the benefits DRAMATICALLY outweighed the disruption. They wouldn't commit to replacing all Mac's, including the Mac Pro, unless they thought they could carry the performance difference through all models. It just can't be marginally better. There will be no hackentosh with Apple Silicon. If Apple follows through and the desktop performance scales to what we have seen with what they introduced on their entry level hardware I guarantee you very few people are going to care AT ALL that these Mac's don't run Windows or really old software - especially in professional settings where computers aren't a hobby but tools to earn you money - and time is something you can't make more of. Needing less time is a competitive advantage. Religious debates will go out the freaking window in a heart beat if these new Macs are as dramatically different from everything else as I fully expect them to be. Otherwise why bother with Apple Silicon? If you can't have such a stark difference that detractors will just look silly, then why bother? But let's say they aren't embarrassingly better in performance and instead stay above M1 performance but not embarrassingly so (say multicore doesn't scale that well) they will still be the best Mac's made and still best the majority of PC's out there. Also just because the M1 Mac's don't support PCI Express or GPUs doesn't mean no chips will. Many people didn't think we would see Thunderbolt - yet these chips have it. And with AMDs latest graphics cards, I really don't care about Apple's rift with nVidia any more.
- onli 6y ago
  We have different definitions of what amounts to R&D expenses. Just look at the regular stuff like the tooling for a new case - we have lots of that, and that stuff is expensive. Mouses, keyboard, displays - nothing stood still. It's normal and a positive thing that the field became more stable after a while, like that all the alternative processor architectures vanished. Until now of course. > Apple can add what they want - and leave out what they don't need. Not just in the CPU but ANYWHERE in the system now. Yeah. The "courage" headphone jack company can now make even more decisions I will disagree with and create unupgradeable systems that are unrepairable. Can't even change the ram in those things, pure throwaway garbage. Apple is the very last company I want to be in a decision making position over my desktop system. This is not a positive development. And yes, they have thunderbolt, but you can't run GPUs with those ports. How useful. > Otherwise why bother with Apple Silicon? If you can't have such a stark difference that detractors will just look silly, then why bother? Money. And being able to release new systems when they are able to, not when Intel gets something out of the door for once. Which is a plus on its own, they won't and do not need to be radically different.

by onli 6y ago

EricE 6y ago

"There is huge R&D investment on the side of the big PC hardware manufacturers, just look at the big ones, AMD, Intel and Nvidia." Those are the only significant sources of R&D for PC hardware. None of the big vendors - Dell, HP, Acer are NOT doing basic R&D. Crap, HP is where good technology goes to die or whither away. All the stuff from DEC flushed down the toilet including Alpha (amazing processor - loved our NT Alpha systems back in the day), StrongARM (how ironic), Compaq (Still have my Personal Juke Box PJB-100 - first hard drive MP3 player with a solid lead on Creative and their awful thing, years before Apple and the iPod[and the PJB guys came from DEC now that I think about it]), Palm - don't get me started on Web OS. Itanium? Talk about making a bad partner choice, Intel. Didn't pay attention with Alpha? Then again Intel got StrongARM from HP and then pissed it away too. But I digress... The problem with the PC vendors is they are all pulling parts out of the same bin of least-common-denominator parts. Apple no longer is. We haven't seen the chips aimed at desktop use where battery life and thermal management will be far more flexible. I don't think Apple will focus on one single aspect like single core performance (which they are crushing all but AMD's newest with a 10 watt mobile part - 10 watts vs 50 watt desktop parts!) but overall system integration. Hardware accelerators. Neural Engine. ML cores. Stuff like that. Stuff we haven't even thought of yet. There is much speculation that Rosetta 2 got specialized hardware assistance which explains its ridiculous performance, esp. compared to the original Rosetta. Why not? Apple owns every transistor. M1 runs Intel code faster than the machines they replace. Apple can add what they want - and leave out what they don't need. Not just in the CPU but ANYWHERE in the system now. I think people are dramatically underestimating just how powerful having full vertical control really is. I suspect we are going to see it in a BIG way when they do release the desktops. Remember they really didn't need to do this processor transition; the only reason to totally disrupt their ecosystem is if the benefits DRAMATICALLY outweighed the disruption. They wouldn't commit to replacing all Mac's, including the Mac Pro, unless they thought they could carry the performance difference through all models. It just can't be marginally better. There will be no hackentosh with Apple Silicon. If Apple follows through and the desktop performance scales to what we have seen with what they introduced on their entry level hardware I guarantee you very few people are going to care AT ALL that these Mac's don't run Windows or really old software - especially in professional settings where computers aren't a hobby but tools to earn you money - and time is something you can't make more of. Needing less time is a competitive advantage. Religious debates will go out the freaking window in a heart beat if these new Macs are as dramatically different from everything else as I fully expect them to be. Otherwise why bother with Apple Silicon? If you can't have such a stark difference that detractors will just look silly, then why bother? But let's say they aren't embarrassingly better in performance and instead stay above M1 performance but not embarrassingly so (say multicore doesn't scale that well) they will still be the best Mac's made and still best the majority of PC's out there. Also just because the M1 Mac's don't support PCI Express or GPUs doesn't mean no chips will. Many people didn't think we would see Thunderbolt - yet these chips have it. And with AMDs latest graphics cards, I really don't care about Apple's rift with nVidia any more.

onli 6y ago
We have different definitions of what amounts to R&D expenses. Just look at the regular stuff like the tooling for a new case - we have lots of that, and that stuff is expensive. Mouses, keyboard, displays - nothing stood still. It's normal and a positive thing that the field became more stable after a while, like that all the alternative processor architectures vanished. Until now of course. > Apple can add what they want - and leave out what they don't need. Not just in the CPU but ANYWHERE in the system now. Yeah. The "courage" headphone jack company can now make even more decisions I will disagree with and create unupgradeable systems that are unrepairable. Can't even change the ram in those things, pure throwaway garbage. Apple is the very last company I want to be in a decision making position over my desktop system. This is not a positive development. And yes, they have thunderbolt, but you can't run GPUs with those ports. How useful. > Otherwise why bother with Apple Silicon? If you can't have such a stark difference that detractors will just look silly, then why bother? Money. And being able to release new systems when they are able to, not when Intel gets something out of the door for once. Which is a plus on its own, they won't and do not need to be radically different.