m1 ridiculously fast

[537635]

This is ridiculous. I did read the reviews, but was like "ok, Safari is always snappier, sure". But this feels like iOS felt for the first time when coming from a slow Nokia interface.

Got a Mini (8/256), coming from a 2020 13'' Pro 16/1TB. Connected to an Ultrafine 5k. Macbook was always struggling with the 5k display, even moving windows around too fast resulted in audible vents. Half screen Google Maps and the laptop was on the verge of death.

With M1, nothing. Dead quiet. Cold to the touch. Scrolling the timeline in Google Photos full screen 5k is smooth as butter with tens of thousands of photos. This was always like 2fps on the 2020 i5 Intel.

Compatibility wise I was afraid important stuff wouldn't work, read some reviews about Dropbox not working. Well, everything works, except some kinks with the Ultrafine monitor not showing resolution correctly, but it appears to be working in 5k actually.

For me this is much more about the smooth interface experience and comfortable operations than some sheer benchmarks.

Apple really did something truly great here.

 

[One2Grift]

Great write up.

I’m 12 days in with my base M1 MBA. The form factor, lack of heat, lack of ambient noise, very quick response, very good image, great keyboard inter, surprisingly good speakers for its form factor, excellent battery life. Definitely appreciate the iOS app commonality. I take some of my cellphone calls on the MBA, the caller feedback has been very positive on the quality of the mic array performance.
So far this is a great machine for everyday use, collab, stream. (I don’t game much and I’ve done no major 4K editing. I can’t comment on their perfs). 849$ at the Apple refurb store ?

No real negatives (at least not yet) except for what I expected with a crossover windows task. Maybe the onboard camera, it is definitely adequate but is pedestrian when compared to its other stand out qualities.

 

[Ev0d3vil]

Great write up.

I’m 12 days in with my base M1 MBA. The form factor, lack of heat, lack of ambient noise, very quick response, very good image, great keyboard inter, surprisingly good speakers for its form factor, excellent battery life. Definitely appreciate the iOS app commonality. I take some of my cellphone calls on the MBA, the caller feedback has been very positive on the quality of the mic array performance.
So far this is a great machine for everyday use, collab, stream. (I don’t game much and I’ve done no major 4K editing. I can’t comment on their perfs). 849$ at the Apple refurb store ?

No real negatives (at least not yet) except for what I expected with a crossover windows task. Maybe the onboard camera, it is definitely adequate but is pedestrian when compared to its other stand out qualities.

I found one on the second hand market for 867ish USD. I thought of getting the 16GB model but then thought again, this is only the first iteration of the M series. There will be faster chipsets coming soon. 8GB for the Air is plenty good, but 16gb should be reserved for the pro series.

 

[George Dawes]

And this is just the first base chip ! !

Imagine the new mac pro with apple silicon inside !

do we have a mind blown smiley ?

 

[One2Grift]

I found one on the second hand market for 867ish USD. I thought of getting the 16GB model but then thought again, this is only the first iteration of the M series. There will be faster chipsets coming soon. 8GB for the Air is plenty good, but 16gb should be reserved for the pro series.

my gut instinct says you’re right. Yet so far my MBA largely outperforms my 3 windows laptops that all have 16GB (to be fair my beat up HP is 4 years long in the tooth, I have to use a usb keyboard, heat and fans - you can fry bacon on it, getting video lines, one USB port doesn’t work. Lol For the use and beating it took it was a damn! fine laptop). My 2020 Lenovo is a very solid laptop(except the fan kicks off with very little happening and it likes to get quite warm). I bought the 8gb model but in the first week I ordered an 8gb strip and added to it, it clearly was needed. But it’s a good laptop and was my go to before my MBA. The MBA is just a better to use. So I’m cautiously optimistic/wait and see how the Apple silicon SOC performs with less “RAM” But ultimately you may definitely be right. We’ll see….
One thing I’m not worried about is too much swap writes. A competitor cajoled ploy imho.

 

[George Dawes]

Don't forget the outrageous battery life on the macbook !

M1 is so much more than raw power , its a game changer

 

[Apple_Robert]

I think the M1 is grand. It has been able to do anything I have needed done and all in the same manner as previously mentioned. I look forward to seeing what Apple does with the new M series MBA.

 

[UBS28]

And this is just the first base chip ! !

Imagine the new mac pro with apple silicon inside !

do we have a mind blown smiley ?

It is not the base chip, the base chip is the A14 that is in the iPhone.

 

[Khedron]

And this is just the first base chip ! !

Imagine the new mac pro with apple silicon inside !

do we have a mind blown smiley ?

?

 

[cmaier]

It is not the base chip, the base chip is the A14 that is in the iPhone.

That’s sort of an arbitrary position to take, given that even every transistor and wire in the a14 core is different than in the M1 core.

 

[Mcckoe]

The M1 is an amazing SOC, and what your talking about is exactly why it was made. The Chip’s ability to limit bottlenecks, and utilize insanely fast transfer speeds has essentially given Apple: a scalable, low cost, Hi-performance powerhouse; that offers display requirements for 99% of the population.

The only thing left is for Apple to convince people this chip is all they need, and with each generation; that is going to get easier and easier. In four generations 8k@120hz should be a comfortable native resolution to hit(not that anyone will need that before 2030).

 

[UBS28]

That’s sort of an arbitrary position to take, given that even every transistor and wire in the a14 core is different than in the M1 core.

It is not, the M1 is basically the iPad version of the A14, aka the A14X.

That is how Apple was able to put it in the MBA without a fan, as it is an iPad chip.

Even the Dev kit of the M1 Mac Mini had the A12Z from the previous iPad.

 

[Astralis56]

It is not the base chip, the base chip is the A14 that is in the iPhone.

He was talking about the Mac Silicon chips

 

[cmaier]

It is not, the M1 is basically the iPad version of the A14, aka the A14X.

That is how Apple was able to put it in the MBA without a fan, as it is an iPad chip.

The next version, the M2 and M2X will be based on the A15 from the iPhone.

No it’s not. The a12x cores and the a12 cores are physically identical. A14 and M1 have completely different core physical designs. They are the same microarchitecture, but completely different from a circuit standpoint.

And the M2 is not based on the A15. The A15 and the M2 share the same core microarchitecture - neither is “based on” the other. In fact, the M2 was designed prior to the a15.

 

[NARadyk]

Cannot imagine how the M1X will be later this year on the new Mac Pro’s. It will almost feel unnecessary?

 

[UBS28]

No it’s not. The a12x cores and the a12 cores are physically identical. A14 and M1 have completely different core physical designs. They are the same microarchitecture, but completely different from a circuit standpoint.

And the M2 is not based on the A15. The A15 and the M2 share the same core microarchitecture - neither is “based on” the other. In fact, the M2 was designed prior to the a15.

The A12X/Z was a 8-core chip with 4 high-performance and 4-efficiency cores. Does that sound familar? So the M1 is basically the A14X that was coming the iPad Pro in the first place.

That is how it is able to also run iPhone and iPad apps.

That is also why Apple put the M1 in the iPad Pro, as it is basically the A14X that the iPad Pro would have received in the first place.

 

[Mcckoe]

It is not, the M1 is basically the iPad version of the A14, aka the A14X.

That is how Apple was able to put it in the MBA without a fan, as it is an iPad chip.

Even the Dev kit of the M1 Mac Mini had the A12Z from the previous iPad.

Realistically, the A10X was Apple’s first steps toward the M1, Bionic chips followed; but, the closest iteration before the M1 was released was the A12X; A12Z was basically a optimized reprint of the A12X.

 

[cmaier]

The A12X/Z was a 8-core chip with 4 high-performance and 4-efficiency cores. Does that sound familar? So the M1 is basically the A14X that was coming the iPad Pro in the first place.

What does that prove. Once again: the physical design of each firestorm/icestorm core in M1 is completely different than the physical design of the firestorm/icestorm cores in a14. Just look at them - you can easily see that the standard cell placements and the routes are completely different. They have completely different physical designs.

This is different than A12X vs. A12, where the physical designs were essentially the same.

In other words, M1 is not to A14 as A12X is to A12. The only similarity between the two situations is that in both cases the “newer” chip had the same number of cores as each other, and the same microarchitecture as the older corresponding chip.

People want to sound like they know something so they keep saying “m1 is just a14x and m2 will be based off of a15” but that just shows that these people have never designed chips.

 

[UBS28]

Realistically, the A10X was Apple’s first steps toward the M1, Bionic chips followed; but, the closest iteration before the M1 was released was the A12X; A12Z was basically a optimized reprint of the A12X.

True. The iPad Pro’s have always been faster and energy efficient than Intel MacBook Pro’s in the past. So it made sense for Apple to put those iPad chips in their Mac’s at some point.

 

[UBS28]

What does that prove. Once again: the physical design of each firestorm/icestorm core in M1 is completely different than the physical design of the firestorm/icestorm cores in a14. Just look at them - you can easily see that the standard cell placements and the routes are completely different. They have completely different physical designs.

This is different than A12X vs. A12, where the physical designs were essentially the same.

In other words, M1 is not to A14 as A12X is to A12. The only similarity between the two situations is that in both cases the “newer” chip had the same number of cores as each other, and the same microarchitecture as the older corresponding chip.

People want to sound like they know something so they keep saying “m1 is just a14x and m2 will be based off of a15” but that just shows that these people have never designed chips.

The A14 is 50% faster in single core performance than the A12 and so is the M1/A14X over the A12X/A12Z. If the M1 was not the A14X, you would not see the same performance bump also.

The iPad Pro was expecting a 50% performance jump with the A14X, which it exactly got from the M1.

 

[Mcckoe]

True. The iPad Pro’s have always been faster and energy efficient than Intel MacBook Pro’s in the past. So it made sense for Apple to put those iPad chips in their Mac’s at some point.

It was always Apple’s plan to develop the chip on the iPad Pro line, and utilize it elsewhere. However, the results after the 12X surprised even Apple engineers; which moved up the roadmap considerably, and lead to product unification 3 generations before initially planned.

Keep in mind the whole market was/is moving in this direction: Apple just had a slight lead, and that multiplied exponentially as they found success.

 

[cmaier]

The A14 is 50% faster in single core performance than the A12 and so is the M1/A14X over the A12X/A12Z. If the M1 was not the A14X, you would not see the same performance bump also.

The iPad Pro was expecting a 50% performance jump with the A14X, which it exactly got from the M1.

What? That doesn’t even make logical sense. So now, if a chip has a particular performance gain you can declare that it has the same relationship as another pair of chips that had the same performance gain? That makes no sense whatsoever.

Am I supposed to believe that strange theory, or my own eyes, that can see, plain as day, that the cores in the M1 are different than the cores of the A14?

And am I supposed to believe you, or people who actually designed the chips, about the relationship between A15 and M2?

 

[UBS28]

What? That doesn’t even make logical sense. So now, if a chip has a particular performance gain you can declare that it has the same relationship as another pair of chips that had the same performance gain? That makes no sense whatsoever.

Am I supposed to believe that strange theory, or my own eyes, that can see, plain as day, that the cores in the M1 are different than the cores of the A14?

And am I supposed to believe you, or people who actually designed the chips, about the relationship between A15 and M2?

How are the cores different when the M1 has the same performance that the A14X would have had? The M1 is now also in the iPad Pro, meaning it has been designed for a tablet, aka the A14X.

It is not a Mac / laptop chip, because else it would not perform like the A14X and would not fit in the iPad Pro.

 

[cmaier]

How are the cores different when the M1 has the same performance that the A14X would have had? The M1 is now also in the iPad Pro, meaning it has been designed for a tablet, aka the A14X.

It is not a Mac / laptop chip, because else it would not perform like the A14X and would not fit in the iPad Pro.

I just told you four times how the cores are different. They have different netlists, different layouts, different standard cell placements, etc. You have no idea what performance the “A14x would have had” because no such chip exists.

Your argument: 1984 chevy impala is 10% faster than 1983 Chevy impala. 1990 toyota Tercel is 10% faster than 1989 Chevy impala, so the toyota is just the “x” version of the impala.

It’s a logical fallacy.

 

[Khedron]

I just told you four times how the cores are different. They have different netlists, different layouts, different standard cell placements, etc. You have no idea what performance the “A14x would have had” because no such chip exists.

Your argument: 1984 chevy impala is 10% faster than 1983 Chevy impala. 1990 toyota Tercel is 10% faster than 1989 Chevy impala, so the toyota is just the “x” version of the impala.

It’s a logical fallacy.

The Car is just the HorseX.