I would disagree that this is first order. First order is making the transition as smooth as possible, which obviously means having a very good translation layer. Only then you should even think about competing on comparability.
Does this not depend on how one sees the Arm transition matter playing out?
It at least conceivable and IMHO, plausible for Qualcomm to see Apple, phones on ARM and aging in demographics all speaking to a certain Arm transition?
I wouldn't be so sure. Windows on ARM has existed for more then a decade with almost zero adoption. Phones, both Apple and Android, have been ARM since forever. The only additional player is that Apple has moved their Macs to ARM. This to me means it would be pretty stupid for them to just throw up their hands and say "they will come". Because it didn't happen for a decade prior.
Maybe. Just trying to see it from other points of view.
A decade ago, Apple was on Intel and Microsoft had not advanced many plans in play today. Depending on the smoke they are blowing people's way, one could get an impression ARM is a sure thing.
Frankly, I have no desire to run Windows on ARM.
Linux? Yep.
And I am already on a Mac M1.
I sort of hope it fails personally. I want to see the Intel PC continue in some basic form.
That superhuman capability of "multiplying two large numbers instantly" has transformed human society like not even plumbing has. I really can't see how this you could not consider this useful.
You don't need to flex the material you are measuring to get tens of microns difference. Simply both sides of the caliper will not stay parallel since there is always some dust or fluid between the sliding pieces. The dust will simply flex.
Of course. To get a good reading you have to clean both surfaces. I mean if you want to wring blocks together it won't work without clean surfaces either.
Why don't you think it's a thing? A trivial example is audio. A ton of audio speakers can produce frequencies people cannot hear. If you have an unprocessed audio recording from a high end microphone one of the first compressions things you can do is clip of imperceptible frequencies. A form of compression.
Why would a consumer care about what node something is on? You should only care about a set of processors that is available in the market at the same time. The M4 is available now and Zen 6 is not. Once zen 6 is here we probably have an M5.
The question wasn't entirely serious, no. My point is: afaik the M processors aren't actually available on the market, they only come as one component of a much more expensive product.
I don't mean to take away from how impressive they are.
I don't get it. You can literally go buy it right now. You have been able to buy it for months and months by getting an Ipad. If you are saying you can't buy it because it's used inside a product then the same goes for basically all mobile processors. I can't "buy" an AMD Ryzen HX 370. I can't buy an Intel Core Ultra 258V. And neither can I "buy" a Qualcomm Snapdragon X1E-80-100. This has never even been a factor.
You are correct, the situation is similar for most mobile processors. They are unavailable on the market for consumers looking to build a system.
Apple goes one step further though: M processors aren't just unavailable to consumers, there's also no way for OEMs to build systems using these chips. In this point they differ significantly from the examples you mentioned. For people that do not want to buy into the Apple ecosystem, M chips are effectively not on the market, and benchmark comparisons to desktop or server CPUs are meaningless.
Why does it matter that they aren't available to OEMs? This is moving the goalpost from your original argument.
The second part of your argument is currently correct but that will change shortly. There is no reason to lock into the apple ecosystem. M4 support is underway in Linux though not yet available. You can easily use a Mac mini as a server running Linux.
>This is moving the goalpost from your original argument.
No, them only being available as part of Apple products and thus not on the CPU market was my original point. I should probably have been more explicit in my original comments. I don't believe I have been moving goalposts, but your interpretation of the point I was trying to make might have changed.
Please forgive the weird analogy, but if I'm on the market for a radio, I really don't want to buy a car just for its radio. Especially if there's no way to use the radio without that specific car.
My issue with that analogy is that the CPU is more like the engine of a car, and people certainly do buy cars just to have a vehicle with a given engine.
When it was only the iPad with the M4, it was easier to be sympathetic to your cause, since the iPad is totally locked down and isn't a general purpose computing device since it can't run arbitrary code. But now the Mac Mini is available. It is a general purpose computing device, and you can install Firefox and Linux or whatever you want.
It doesn't meet the level of hardware vendor purity you're asking for, sure, but that's a more ideological now that there's a general purpose computer with the M4 for sale. (Just wish it were cheaper.)
And since it just went on sale today, I was highlighting that, since other readers might want to know that they can now get a computer with an M4.
What isn't compatible with PyPy? I can run large frameworks using pypy no problem. There certainly will be package that aren't compatible. But far and away most of the ecosystem is fully comaptible.
That's not entirely true. Retraining is very expensive. If you can train on a very large dataset including proprietary knowledge and then postprocess the model cheaply to forget things saves you retraining for every variation.
reply