Porting to Mac with Apple Silicon means instant access to all iOS devices too....over 1Bn instant potential customers. That’s a whole different prospect to only being able to target a small group of Mac gamers.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Apple Silicon M1 Chip in MacBook Air Outperforms High-End 16-Inch MacBook Pro
- Thread starter MacRumors
- Start date
- Sort by reaction score
Porting to Mac with Apple Silicon means instant access to all iOS devices too....over 1Bn instant potential customers. That’s a whole different prospect to only being able to target a small group of Mac gamers.
Bootcamp users were less than 1% of all users and I’d hazard a guess that Windows VM users were pretty low too. Apple care about the majority of its user base, hence this move. For the minority of users who want to run Windows, it’s pretty clear you’ll need to buy a legacy pc moving forwards.
Won't it be impossible for Apple to ever get a great GPU into their silicon?
Now I know Apple have VERY talented chip designers, but look at something like a Nvidia RTX 3080 with 28 Billion transistors.
Apple's M1 chip has only 16 billion on whole chip.
It's never going to get anywhere close to a high end graphics card is it?
Now I know Apple have VERY talented chip designers, but look at something like a Nvidia RTX 3080 with 28 Billion transistors.
Apple's M1 chip has only 16 billion on whole chip.
It's never going to get anywhere close to a high end graphics card is it?
Welcome to the world of Apple criticism, where a feature suddenly matters just because Apple doesn’t have it, then mysteriously stops becoming such a huge deal when Apple finally implements it.
Eg: 5g, folding phones, smart speakers, 120hz displays, netbooks, cheap smartphones.
I don’t see how this is any different compared to the existing macbook lineup, where users are already unable to manually upgrade the ram or internal storage after purchase.
And like you suggest, I am willing to bet that the number of people who do care about such functionality are in the minority.
The M1 is not going to go in the higher end computers. Have to wait and see what Apple does for the 16” MacBook Pro, iMac and Mac Pro.
the arm in the micro was not the primary cpu - i may have worded my comment poorly - it was a coprocessor option (there was other compatible coprocessors as well, including x86 and Motorola products). The coprocessor was rare, at least we had one out of a suite of bbcs.
BBC Micro expansion unit - Wikipedia
I believe that there are enough channels (though I’m not completely familiar with LPDDR4x - I never had to deal with it when I designed chips. But I think it’s 6 address bits per channel.)
Could be that they wanted a bigger L2 cache, or they need changes to the TLBs, or something else, though, to support more RAM. Or perhaps it just wouldn’t fit in the package, and they are reserving a bigger package for the high end machines that also have a bigger SoC (with more graphics cores and CPU cores).
Sorry, don't have time to read 32 pages, but did nobody stop to ask this question:
If this chip is truly faster and more efficient in day-to-day usage, why is Apple not selling it at a premium, above and beyond what the Intel equivalents cost? Surely a more efficient and powerful Mac is more valuable?
It is very unlike Apple to sell a product for less than what they could. In fact, their investors would/should scream bloody murder.
If this chip is truly faster and more efficient in day-to-day usage, why is Apple not selling it at a premium, above and beyond what the Intel equivalents cost? Surely a more efficient and powerful Mac is more valuable?
It is very unlike Apple to sell a product for less than what they could. In fact, their investors would/should scream bloody murder.
Could be. It’s a very interesting question that requires simulation and testing to determine. Would be a fun chapter in a dissertation. Nice 3-D surface graphs of task latency and performance/watt as a function of the number of each type of core. Would also vary wildly by workload, of course. But most workloads don’t scale very well across multiple CPU (as opposed to GPU or neural engine) cores, meaning you would generally be relying on multiple high performance cores because you are running multiple independent tasks. I think.
You are probably right but the underlying architecture will be similar, whatever they call it. This is purely a guess from someone who doesn’t design chips but they will probably have more cores and more memory and better graphics because they will have multiple SOC’s. I don’t know how they would tie all of the SOC’s together so that they can access memory or graphics but honestly it’s not my problem.
Well, a 16GB, 1 TB, M1 MBP config is $1800. Only $100 cheaper than the equivalent spec’d intel i5. Apples margins stay similar - I’ve heard it to be in the 40% range. I expect these new machines to drop in price in 2-3 years as they recoup R&D costs.
I do suspect the new ARM MacBooks might have a slightly more gross margin than the previous Intel ones. This is baseless speculation, I have no idea how cost accounting works at Apple.
Also price elasticity won't allow them to sell entry level thin and light MacBooks that much over what Intel ones were priced at, even if their performance is great, especially if Apple has more powerful versions of the M1 lined up for the more powerful products.
Apple's company-wide gross margin has consistently been around 38% for a while, long enough that I'm getting a bit suspicious of the number. Either way that number includes all the products including iPhones, services, accessories, etc, which all have higher gross margin than Macs.
The Macs traditionally have had a lower margin, around 30%. You will see that in the pre-iPhone Apple financial statements where the gross margin was around 20-30%, and that's including all the expensive models. That makes me think the gross margin on entry models is probably lower than people expect from Apple overall.
Nothing is impossible, time and money is all they need.
For me everything panned out as expected [except no re-design which I still think was a mistake].
The M1 chips seem awesome and I imagine will be amazing for Apple developed apps.
I am looking forward to seeing the Rosetta 2 performance on the MBP 13 vs native on a MBP 16. Suffice to say I will have both soon and will be definitely putting it through the ringer. If there are any issues it will go back and I will just wait it out for 6 months [or just get a base air for mucking about].
The M1 chips seem awesome and I imagine will be amazing for Apple developed apps.
I am looking forward to seeing the Rosetta 2 performance on the MBP 13 vs native on a MBP 16. Suffice to say I will have both soon and will be definitely putting it through the ringer. If there are any issues it will go back and I will just wait it out for 6 months [or just get a base air for mucking about].
It has the same memory options and number of ports as the 13" base model it replaced. You can still get the 13" Intel with better options. The only change is with the number of monitors supported, it appears.
This year is ahead of next year.
The GPU in the M1 is pretty impressive according to the post below. And this being an integrated GPU.
M1 8-core gpu is equal to gtx 1060 6gb!!
https://browser.geekbench.com/v5/compute/1802234 https://browser.geekbench.com/metal-benchmarks gtx 1060 6gb metal: 18441 m1 gpu: 18656 This is pretty good for a entry level chip!
forums.macrumors.com
As an IT guy, I have lost count of the number of times I have helped out friends and family when their computers became horribly slow. 99% of the time it's because they upgraded the OS and applications over the years until the RAM can't deal with it, so their machine was page faulting and had slowed to a crawl. Simple and cheap fix - I would go and buy some cheap RAM from the local IT shop, install it in 5 minutes, and boom, their machine is running like new. But for 5 years now Apple has killed this off, entirely and purely for greed and profit, thus rendering all such machines doomed to an early, unnecessary, and environmentally wasteful death. And to cap it off, Apple charge 4x the market retail rate to pre-upgrade RAM when you buy the machine. Which is even worse when you consider that 3 or 4 years down the track when you actually need to upgrade the RAM, the prices have halved, so in effect, the Apple tax on RAM is 8x. Pure disgusting greed, and environmental sabotage.
This topic has been beaten to death, over and over and over and over and over and over. Facts have no effect on people's views about it. But this is probably not the best thread to beat it again in, in any case.
Game devs don't think about which processor architecture they're targeting. Unless it's Quake III, haha. There are other factors for incompatibility, mostly DirectX vs OpenGL vs Vulkan vs Metal.
Btw, the Nintendo Switch is ARM. The Xbox 360 and Wii U were both PPC.
Last edited:
This too, Mac<->iOS cross compatibility is going to be huge. Though as I understand it, that has less to do with them using the same instruction set and more to do with Apple unifying the libs and such. Idk, I gave up Mac dev years ago.
Last edited:
This one?
When was this, though? RAM space/$ used to double every few years, now it doesn't anymore. It even became more expensive at some point due to supply problems. So it's less of an issue than it was in, say, 2012. Case in point, I'm still happily using my 2015 MBP with the baseline 8GiB RAM it came with.
Really? 28 vs 16 is a smaller difference than I expected given how much larger the 3080 is. But that's besides the point...
I doubt Apple can compete with dedicated GPUs. Intuition (I'm no chip designer) tells me that GPUs don't have to be as small as CPUs because their computations are more parallelized, therefore integrated graphics will always have a big disadvantage.
Last edited: