*** Apple event 10th November @ 6pm - One More Thing: Apple Silicon Macs Reveal ***

[HACO]

Given the GPU is massively better than the integrated Intel ones, even under Rosetta anything that uses metal should be faster.

But GPU performance isn't related to ISA as you only need to make sure everything is compatible, i.e. no emulation so I'm sure that will be fine. That's why Apple showing Tomb Raider on Rosetta during WWDC wasn't exactly a tough task for their chips.

CPU performance is what's mostly affected by emulation.

It was more aimed at your last comment saying "Maybe once those results are out, the "Geekbench is bad" crew will turn into "Cinebench is bad" crew "

Which seems to insinuate certain things

It's a common occurrence that people tend to dismiss any benchmarks that show ARM closing the gap with x86, or in this case, completely bridge it, as "not relevant", "not real", "not good", or "not representative" and they're just running out of benchmarks where x86 has any meaningful lead. In most cases it stems from either not doing the actual research into how much ARM-based chips have improved, or in some cases x86 (or Intel/AMD/PCMR) fanboyism that wants to pretend like only Intel and AMD can make "real" CPUs.

 

[HACO]

Hopefully the emulation is going to be good, but that is something that people will be able to see soon, hopefully some review units are out and reviews come out fairly soon as this will be a big thing, whether the emulation works properly as advertised. I wasn't around for the PowerPC to Intel days with Rosetta so don't know what to expect.

Sadly Apple doesn't send pre-release review units to the people who actually review them property and put them to real tests. It will just be YouTubers at first who will always say it's great. But yeah, information will begin to trickle starting next week.

 

[HACO]

Interesting Video from Jonathan Morrison, not a huge fan of his content, and also know nothing about video editing but take a look and see what people think. If that's a comparable export on both then that's interesting in terms of potential performance.

https://twitter.com/tldtoday/status/1326610187529023488?s=20

That's not a good comparison of the two CPUs, A14 definitely has hardware encoding blocks for H265 and that Intel one is running on software or QuickSync. However, most end-users don't care about any of that. They just want their projects done faster.

 

[HACO]

The M1 will have all the same encoding blocks etc and as you say it doesn't matter how it's obtaining the performance the fact is the new Mac's are likely to be monsters when it comes to video encoding.

As long as Apple has implemented the encodings (source and target) and configurations are within what Apple has implemented, yes. If you move outside of these, it will be software encoding and that can be used as an actual benchmark of Apple's M1 performance cores.

 

[HACO]

Nothing like that at all - for most of us here though we don't believe "leaked" benchmarks from any party (whether that's AMD, Intel, NVIDIA or Apple), and the majority wouldn't accept a single cherry-picked benchmark to demonstrate e.g. the performance of the latest NVIDIA graphics card.

Why should we accept a single leaked benchmark (geekbench) as gospel that the new Macs are suddenly the most powerful chips available?


I have no doubt that the new Apple chips will be fast in certain circumstances (e.g. video encoding and other common tasks that are the target market of most Mac users), but how much of that will be down to specific hardware accelerated fixed function blocks, as opposed to actual raw CPU power?

And Geekbench/SPEC don't run on fixed function blocks, either. And Anandtech has reviewed A14 so it's not Apple leaks or cherry picked benchmarks, and to extrapolate that M1 would be similar is reasonable, unless you disagree?

But this discussion is pointless since in a week these devices will be out and we'll see more

If it was easy to take an off the shelf ARM design, sprinkle some magic fairy dust on the relevant parts and provide across the board gains, then Intel or AMD would have done it by now, and found a way to sell it to the growing ARM server market.

Never said it was easy or magical. There's nothing special about ARM ISA, the competition just stagnated themselves into this situation where their 50x lead in performance 10 years ago is now more or less completely gone (against Apple) and is down to maybe 1.5x (against other ARM competitors).

 

[HACO]

So you must really just love ARM then?

These arm chips are just going to perform well on apps optimised on the mac and their own 1st party apps, doesnt mean they'll be anywhere near AMD (and possibly intel) when it comes to other real world applications

Another personal shot like you did before, I'm just going to ignore these.

As for your second comment, I'll just leave it there until we see more benchmarks out of the retail units, even though Anandtech's SPEC results already showed you're very wrong about that (SPEC isn't a first party Apple app, lol).

 

[HACO]

I think Apple keynot was fine, 99% of people who watch the keynote wouldn't even know what an i5/i7 XXYXZZ is, so Apple cut to the chase, Its faster. In specific workloads no doubt but the workloads that a majority use I imagine.

Totally, the graphs and numbers in the keynote were just marketing stuff for generic non-tech news websites (and for YouTubers to rely on, lol). They didn't even say what they were comparing them against.

 

[HACO]

https://twitter.com/x/status/1326866126635143169

More benchmarks.

An integrated GPU not doing that badly against a 580X is amazing.

That's the lead developer of Affinity Photo as well, not just a random guy. I seriously hope he's not breaking an NDA

Unsure if that's native or Rosetta, either way, very good result.

 

[HACO]

Can someone more familiar with benchmarking put into context the GPU ~1200 in the raster test score? How would this be compared to the iGPU in the current lineups?

https://forum.affinity.serif.com/index.php?/topic/124022-benchmark-1900-results/

These are the user posted benchmarks. Ignore the non-Metal ones as they use different frameworks on Windows.

 

[HACO]

These new machines are a poor offering but show promise, its really the 2nd generation that will sign the changes are really maturing. There are plenty of unknowns that need to be sorted out over the next 12months plus seeing how this translates into the more powerful 'pro' offerings. Thunderbolt ports are lacking, ram offerings low, and the same design as intel. Maybe when intel is phased out some real design changes can be seen, for instance a lot of empty space in that Mac mini.

Staying just behind the curve due to this massive change and the huge potential - keeping in mind CPU is not everything - is probably safer. No need to be a guinea pig.

There's always an early adoption tax (whether in cash, or support, etc), and this one is no different really. Doesn't mean being an early adopter is always a bad idea. If you need a new mac right now, buying these is definitely a better idea than buying Intel ones. Buying them for the sake of buying them? There's that early adopter tax for you.

 

[HACO]

Or too much money to spend I wish I could buy one for the sake of it haha.

Me too. I love my toys, but no money left after buying Zen 3 and 3080