Hacker News Reader

Comment by Jonnax | Hacker News Reader

"the i3-1000NG4 Intel processor that is the cheapest option for the MacBook Air is not yet for public sale; it probably costs around $150, with far worse performance than the A13"

In so curious so see this. Unless I missed it we haven't seen consumer high end ARM chips that are readily available running a general purpose OS.

Like the Raspberry Pi is cool, but it's still trash in comparison to an average x86 laptop.

Let's see if those really high geekbench scores translate to an excellent computing experience.

View on HN

I mean, is the iPad Pro not evidence enough? The A13 is much, much better than the chips in RPis.

I'd really like to see an iPad Pro crunch through as many VSTs as my old 2600k did in its 2 millisecond budget

I do some light VST development with JUCE and have tested out my synth on my iPad Pro, worked a charm.

> I do some light VST development with JUCE and have tested out my synth on my iPad Pro, worked a charm.

I mean, of course a single VST works like a charm, single VSTs already worked fine on 500 mhz CPUs with win98-era computers and Cubase VST. The question is how many tracks with synth + 20 filters behind can you layer before having to freeze things.

As highlighted in the article, Apple's A13 is on par with Intel's i9-9900K in SPECint2006 and only 15% behind in SPECfp2006.

those are benchmarks, but to restate my comment I'd really like to see that hold up in actual real-world use cases

I think it'd do just fine. The A12X has 10 billion transistors. Sandy Bridge has 1 billion.

But my sandy CPU had not much trouble being OC'ed to 4.5ghz while apparently the top freq. of the A13 is 2.7 ghz? How many, say, convolution reverbs can you chain on the A12 or A13 would for instance be a super relevant metric to compare things.

How is this a valid comparison?

You have active cooling for the sandy CPU the iPad does not.

And I bet your sandy CPU had active cooling. The A13 tops out at 2.7GHz because that's what Apple decided was safe for a passively cooled phone. I don't doubt that it could run at 4.5GHz in a desktop package, with a proper fan and heatsink.

I don't understand, isn't this whole conversation about using these ARM chips for the next desktop macs ?

> I don't doubt that it could run at 4.5GHz in a desktop package, with a proper fan and heatsink.

well, I don't know at all. I'm pretty sure you can't take a, say, 2.something ghz U-series intel i7 CPU and make it go a lot beyond 3ghz

You'd almost certainly need to redesign the hardware for that; the A13 already pushes its efficiency curve during boosts.

And the A13 was built on ULP node with a maximum power budget of less than 6W. There is no reason why an A13 built on HP node with power budget of 20W and active cooling cant reach 4Ghz +.

The iPad Pro uses the A12Z (https://en.wikipedia.org/wiki/IPad_Pro). The A13 is used in the iPhone 11 variants + iPhone SE. Your point stands, these chips have insane geekbench scores.

Reply on news.ycombinator.com