Live data from Hacker News

2020 Mac Mini – Putting Apple Silicon M1 To The Test

anandtech.com

231–240 of 767 posts

Re: 2020 Mac Mini – Putting Apple Silicon M1 To The Test

#231
post #190

Earlier quoted context omitted.

Not to mention 5950X alone without cooling ($799) costs almost as much as an entry level MacBook Air The M1-based Mac mini starts at $699.

Yeah, forgot about that. Everything else being equal (ostensibly), the M1 Mac Mini is $200 cheaper than the crappy Intel i5 Mac Mini, more if you upgrade the Intel CPU. As an owner of a decked out 2019 Mac Mini, in hindsight I made a shitty purchase decision.

I sold my 2018 Mac Mini with high specs 1 week before the keynote.

The guy must be feeling bad right now.

Re: 2020 Mac Mini – Putting Apple Silicon M1 To The Test

#232
post #13

Dave2D found the air to be on par with the Pro, at least for tasks that took under ~8.5 minutes. It only really throttled after that point, according to him.

So.. if I place it on a slab of ice, it would work the same as Pro? Tbh, the only reason I didn't even think of buying a pro is because I don't want the touch bar. I might still buy an air if there's no touch bar on the pro, but the decision will be a lot harder.

Both currently released M1 Pros have the touch bar, so it looks like your decision will be quite simple!

Re: 2020 Mac Mini – Putting Apple Silicon M1 To The Test

#233

The Cinebench R23 results seem kinda weird to me. The 5950X would have almost a 40 % clock speed advantage over the M1 (~5 GHz 1C vs 3.2 GHz 1C), yet the M1 is only about 8 % slower in an entirely ALU-limited SIMD benchmark? This suggests the M1 core has like 50 % more EUs and achieves much higher throughput than Zen 3. The SPEC results are... decisive to say the least. Without Zen 3, x86 CPUs would look, well... lik…

You overestimate how ALU limited Cinebench is.

I managed to find an AVX vs AVX off benchmark run for Cinebench r20 [1]. Going from 128bit SSE to 256bit AVX and doubling the ALUs only results in a 10-12% increase in performance.

I assume this has to do with how each SIMD lane of calculation might need to branch independently, limiting the performance speedup from just throwing wider SIMD ALUs at it.

[1] https://www.techpowerup.com/forums/threads/post-your-cineben...

Re: 2020 Mac Mini – Putting Apple Silicon M1 To The Test

#234
post #23

Earlier quoted context omitted.

That's one data point that's particularly interesting to me. As someone who (normally) travels a great deal, I'd probably go with the Air unless there were real throttling compromises, especially given that I use a different computer for multimedia editing at home.

Wish they re-introduce the discontinued Macbook 12 inch with the same specs as air. It weighed only 970grams vs 1.29 kg for air. In fact air feels bulky compared to other light weight laptops like LG Gram, not to mention the design is outdated. Always wondered why Apple killed the smaller model. Perhaps they want to push the iPad pro so killed off the netbook line. The wannabe traveller inside me keep drooling at 12…

There are several rumours about a return of the 12-inch in 2021H1.

Re: 2020 Mac Mini – Putting Apple Silicon M1 To The Test

#235
post #66

Earlier quoted context omitted.

That conclusion is quite misleading, in my opinion. They write "outperforms the core performance" and the keyword here is "core". What they mean is that if one had a single-core Zen3 and a single-core M1, then the M1 would win some and lose some. But in the real world, most Zen3 CPUs will have 2x or more cores, thus they'll be 2x to 4x faster. So what they mean to say is that they praise Apple for having amazing per-…

It bodes really well for future chips with higher power budgets. The Pro seems a bit underwhelming for what it could be though.

They only need to add 2 ports to the Pro to differentiate it from the Air.

Re: 2020 Mac Mini – Putting Apple Silicon M1 To The Test

#236
post #128

I'm not buying into Apple's eventually-closed desktop computer systems - I don't care what the performance is. They've been slowly marching towards iOS's closed ecosystem model on the Mac and with an in-house CPU, they can effectively lock users out of alternate OS choices on their hardware. Buyer beware.

I feel exactly the same way, but I do wonder how this gap is going to close. If Microsoft team up with another ARM vendor and make similarly closed-off, proprietary glue sandwiches as a response, where does that leave Linux? I doubt PC-compatible, x86 laptops are going to disappear off the face of the Earth anytime soon, but... if there's basically two types of machines, and ours have worse performance and half the b…

You just have to look at the phone community (XDA Developers, etc) to see how this will (eventually) go.

A good example were the old Asus Transformer tablets. They were a super niche device, but it still lended itself to Linux and so a small team of people managed to load Ubuntu on it.

Another are Samsung phones. They try to lock people out, but they have popular enough devices that people find a way to put LineageOS on them.

Finally, even iPhones aren't immune. Small teams of people have managed to load Android on them and get it (partially) working. More people would give them even greater functionality.

If laptop manufacturers lock things down with ARM, there will be people who work around those mitigations and install their own OS on that hardware. Tooling will be developed to make that process easier and easier for the next round of people with that device (or future devices). It'll suck up front until the community grows large enough to work around issues faster and faster.

And that's even supposing worst case scenario. I'm not fully buying the idea that you _won't_ be able to change the OS on these laptops. Microsoft has tried (and failed) to lock other OS's out of their laptops. Chromebooks are (currently) the largest market of ARM laptops and you're able to change the OS on them. Apple might be the only company even remotely able to hinder freedom on their devices.

Either way. In the war on general computing, I'm generally optimistic for the users.

Re: 2020 Mac Mini – Putting Apple Silicon M1 To The Test

#237
post #60

Earlier quoted context omitted.

So a 60W TDP dGPU (gtx 960M) is ~35% faster than this iGPU? I think this is what they called a Rival.

Agree. But don't you think this would come off as a lot less impressive if Anandtech had included all of the old rivals from 1-4 years ago that still rank above the M1? "If you currently own a 2016 15" Acer, buying the new 2020 MacBook will be a downgrade." sounds pretty lame to me. That's why I said they had a very Apple-friendly comparison set.

> But don't you think this would come off as a lot less impressive if Anandtech had included all of the old rivals from 1-4 years ago that still rank above the M1?

Not really. Why would you compare against old rivals instead of the current market? They had 1660 Ti's on the charts, too, which both obliterated the M1 & are not at all the high-end of discreet mobile GPUs.

The "discreet rival" was because the M1 was competing favorably against the discreet 1650 & 560(X). As in, entry-level discreet GPUs make increasingly less sense (they already weren't making much sense with Intel's new Xe and AMD's Vega 8 & 11 integrated, but more nails in that coffin with the M1)

Re: 2020 Mac Mini – Putting Apple Silicon M1 To The Test

#238
post #48

Earlier quoted context omitted.

Interesting that your takeaway from all this is "oh, it can't beat some of the top x86 chips in existence—it can only meet them on even footing. Guess it'll be falling behind next year." This is Apple's first non-mobile chip ever. You think this is the best they can do, ever?

I'd expect NVIDIA to join the ARM CPU race, too. And they have experience with the tooling for lots and lots of cores from CUDA. So I'd expect to have 5x to 10x the M1's performance available for desktops in 1-2 years. In fact, AMD's MI100 accelerator already has roughly 10x the FLOPS on 64bit. That said, it's an amazing notebook CPU.

>I'd expect NVIDIA to join the ARM CPU race, too.

Nvidia has been making ARM SoCs since 2008. They have been used in cars, tablets, phones, and entertainment systems.

What do you think powers the Nintendo Switch?

Re: 2020 Mac Mini – Putting Apple Silicon M1 To The Test

#239

The last sentence on the first page has a typo, I think. I am guessing it would be `tough competition`, not `though competition`.

I never understand downvotes in this situation. Like are you mad someone is calling out typos? I mean I'm not surprised, I got downvotes for the literally the same thing [0] a few days ago when anandtech published another article full of typos... I like their reporting but the typos and the multi-page nonsense is a huge turnoff.

[0] https://news.ycombinator.com/item?id=25052892

Re: 2020 Mac Mini – Putting Apple Silicon M1 To The Test

#240
post #201
post #159

Earlier quoted context omitted.

I watched and read multiple reviews and Dave2D seems to be the only one who tried to quantify the throttling to some extent, all the others only had useless statements like "The Pro will probably be able to sustain unthrottled workloads for longer thanks to it's active cooling" - No shit, sherlock. For me the fact that it only throttles after 8-9 minutes (!) of heavy use is going to be the deciding factor that will a…

Couldn't the Pro just turn on its fans after 8-9 minutes (to avoid throttling), thus giving the best of both worlds?

Best of both of worlds is:

- active cooling

- lack of a touchbar

Post reply on HN