4 ms·

Apple's graphs without labels, and comparing to "top PC laptops" without naming them actually made me assume the M1 was going to be a flop, and in practice wors

Apple's graphs without labels, and comparing to "top PC laptops" without naming them actually made me assume the M1 was going to be a flop, and in practice worse than a current high-end Dell laptop. It now looks like that isn't true, and the M1 really is revolutionary, but that certainly isn't what I took away from Apple's advertising.

BossingAround 6y ago
> actually made me assume the M1 was going to be a flop, and in practice worse than a current high-end Dell laptop Precisely my thinking. However, I don't think we're the target audience for this Apple generation. If my mother saw it, she'd probably think "great, finally something faster than my computer." Despite the article's praise of the graphs, I think it's a highly risky move that looks highly suspicious, and can pay off only when you know your target audience, and you know you're 100% right. If I saw a cloud provider, let's say Oracle Cloud, having a similar graph against the leader, like AWS, they'd lose credibility in my mind. Unless I'd read later they were right, in which case I'd think someone in their marketing just brain farted when making the graphs.
- philistine 6y ago
  I absolutely agree with you that a graph like this is highly suspect. However, here the less nefarious thing happened: Apple simplified their graph so much we assumed they were hiding something.
- mirthflat83 6y ago
  Not suspicious once you realize their target audience. Explaining something in a way that common people can understand without using technical jargon is a skill that many HN readers should value and learn.
coldtea 6y ago
>Apple's graphs without labels, and comparing to "top PC laptops" without naming them actually made me assume the M1 was going to be a flop That's because you aren't the target audience for the graphs. The targed audience (the average user) would be confused and bored for with labels, and axis, and statistical minituae.
- GordonS 6y ago
  I don't know that I'd call an labels and axis "statistical minutiae" - they are about the minimum information required to gain any meaningful value from a bar chart.
  - mirthflat83 6y ago
    All information you’re supposed to take away from their charts is just this: “Wow, that looks faster than their competition.” None of the general Apple consumers care if their SIMD or whatever benchmark is precisely 1.321353x faster. That’s the definition of minutiae.
Zanni 6y ago
What's missing from this article is that this has to be a single prong of a multi-prong approach. You're right that this sort of language is going to flop with a technical audience, and, in fact, we've all seen these sort of vague claims from other companies before, to the point that we instinctively distrust them. What makes this work for Apple is not simply the accessible language, but the fact that they can back it up, and that an army of tech journalists did just that with benchmarks.
jdxcode 6y ago
I think they focused more on real-world performance that wouldn’t have been represented so well. Things like lots of floating point ability for JavaScript (which has no integer types) or optimizing ref counting (obj-c and swift do this a lot). I imagine those things wouldn’t impact traditional benchmarks.
- rsynnott 6y ago
  That's the interesting thing; they don't seem to have. If they'd gone for ref-counting benchmarks, that would have shown something more like 5-10x; it's really good at that. The graphs they showed seem to be reflective of performance in general-purpose benchmarks (SpecINT et al).
  - jdxcode 6y ago
    I'm saying they focused on real-world performance in the chip and therefore couldn't show benchmarks that would've represented well. Not that they should've had more specific benchmarks. In no way would this audience be able to understand something like reference counting.
astrange 6y ago
That is not their problem, it’s your problem. A graph with labels would not be any more true but it wouldn’t make Intel happy.

BossingAround 6y ago

> actually made me assume the M1 was going to be a flop, and in practice worse than a current high-end Dell laptop Precisely my thinking. However, I don't think we're the target audience for this Apple generation. If my mother saw it, she'd probably think "great, finally something faster than my computer." Despite the article's praise of the graphs, I think it's a highly risky move that looks highly suspicious, and can pay off only when you know your target audience, and you know you're 100% right. If I saw a cloud provider, let's say Oracle Cloud, having a similar graph against the leader, like AWS, they'd lose credibility in my mind. Unless I'd read later they were right, in which case I'd think someone in their marketing just brain farted when making the graphs.

philistine 6y ago
I absolutely agree with you that a graph like this is highly suspect. However, here the less nefarious thing happened: Apple simplified their graph so much we assumed they were hiding something.
mirthflat83 6y ago
Not suspicious once you realize their target audience. Explaining something in a way that common people can understand without using technical jargon is a skill that many HN readers should value and learn.

coldtea 6y ago

>Apple's graphs without labels, and comparing to "top PC laptops" without naming them actually made me assume the M1 was going to be a flop That's because you aren't the target audience for the graphs. The targed audience (the average user) would be confused and bored for with labels, and axis, and statistical minituae.

GordonS 6y ago
I don't know that I'd call an labels and axis "statistical minutiae" - they are about the minimum information required to gain any meaningful value from a bar chart.
- mirthflat83 6y ago
  All information you’re supposed to take away from their charts is just this: “Wow, that looks faster than their competition.” None of the general Apple consumers care if their SIMD or whatever benchmark is precisely 1.321353x faster. That’s the definition of minutiae.

Zanni 6y ago

What's missing from this article is that this has to be a single prong of a multi-prong approach. You're right that this sort of language is going to flop with a technical audience, and, in fact, we've all seen these sort of vague claims from other companies before, to the point that we instinctively distrust them. What makes this work for Apple is not simply the accessible language, but the fact that they can back it up, and that an army of tech journalists did just that with benchmarks.

jdxcode 6y ago

I think they focused more on real-world performance that wouldn’t have been represented so well. Things like lots of floating point ability for JavaScript (which has no integer types) or optimizing ref counting (obj-c and swift do this a lot). I imagine those things wouldn’t impact traditional benchmarks.

rsynnott 6y ago
That's the interesting thing; they don't seem to have. If they'd gone for ref-counting benchmarks, that would have shown something more like 5-10x; it's really good at that. The graphs they showed seem to be reflective of performance in general-purpose benchmarks (SpecINT et al).
- jdxcode 6y ago
  I'm saying they focused on real-world performance in the chip and therefore couldn't show benchmarks that would've represented well. Not that they should've had more specific benchmarks. In no way would this audience be able to understand something like reference counting.

astrange 6y ago

That is not their problem, it’s your problem. A graph with labels would not be any more true but it wouldn’t make Intel happy.