Hacker News Reader

Comment by mromanuk | Hacker News Reader

You are right. But is not Apple's fault, the long line of chip and computer makers long gone (DEC, Sun, also IBM in the PC Market, etc.) All crushed by the Wintel duopoly. Everybody trusted on x86 and specially Intel, to bring us forward, but they stopped innovating at least a decade ago.

I'm happy that Apple is showing that the king is naked, I hope that a new wave of ARM based devices (not just from Apple) will influx new life on the PC Market.

View on HN

I think one problem for CPUs specifically is that people plainly just don't care as much about speed anymore. Modern processors are extremely fast.

So other metrics like power draw get more prominence. Apple managed to get both with a chip design, but I don't see it as a trend yet. I wouldn't even call it innovative in itself, it is just a very good iteration of CPU tech. I think that is awesome because it challenges other companies again.

For performance intensive applications like gaming or CAD, GPUs have become more relevant.

There is a lot of special software for physical calculations (lenses, materials, etc.) that require raw CPU power for n-th degree numerical solutions, but these are niches that cannot sustain an industry.

My main device I use in private for a lot of development has a standard intel CPU and I only looked at power usage because I knew it would be fast enough. I wanted it to be as silent as possible. Turns out it was indeed fast enough for anything I wanted to do. Maybe I buy a new one in a few years, but it didn't yet fail me in any way.

>I think one problem for CPUs specifically is that people plainly just don't care as much about speed anymore. Modern processors are extremely fast.

This is just not true. I do NOT have an M1 yet. I have a group of co-workers that do. This group is normally what I would call the critics that are never happy with anything. With out fail each one of them has raved about how fast and responsive their M1s are. They say it nothing like anything else and is a game changer. They also rave about the fact that they can go a day without needing to find a wall socket. This is a group of devs and hardcore networking people.

From my view, no matter how powerful of an Intel system I have running Windows, macOS or Linux, the desktop UI are not up to the speed of an iPad. Getting that in a laptop...that would be great and I plan to try it out in the form of an M1 Air.

> From my view, no matter how powerful of an Intel system I have running Windows, macOS or Linux, the desktop UI are not up to the speed of an iPad.

What I read: Desktop environment should have more UI effects like iPad.

> I think one problem for CPUs specifically is that people plainly just don't care as much about speed anymore. Modern processors are extremely fast.

The rave reviews that the M1 Macs are getting for performance and power efficiency shows that large step improvements in CPU performance are noticed and greatly appreciated. We just didn't have anything as good as the M1 before to really understand how much better computing could be.

> I think one problem for CPUs specifically is that people plainly just don't care as much about speed anymore. Modern processors are extremely fast.

I disagree. At least in the datacenter there are still lots of workloads that are CPU bound. And many of these workloads are not easy to parallelize. Right now there are about a bajillion startups in a Python codebase that are hitting scaling limits, where the single-core GIL limits are starting to hurt responsiveness.

Yes, they could re-write into Go and get higher performance and better concurrency at the same time. But most orgs don't have the bandwidth for re-building their software stack from scratch. They'd much rather throw money at hardware.

My guess is if you had a datacenter CPU with 2-3X the single-core threaded performance of top-level Xeons, it would sell like hotcakes.

That particular issue is much better solved by a JIT compiling GIL-free Python implementation like GraalPython.

Better solved depends under the organizational and operational calculus you’re working under.

Many companies would rather throw money at beefier hardware than play around with experimental JITs that may or may not break standard behavior in unpredictable ways.

That's taking slow software and complaining that it's slow due to hardware rather than making the software itself faster, either by rewriting it or optimizing it another way.

Well you’re not totally wrong but the one place where consumers care about raw single threaded performance is the web browser and these new rigs are making huge strides. The M1 browser is remarkably fast and it unlocks new territory in what you can do without bringing the browser to its knees. The new 11th gen Intel laptop part has also opened a big gap between itself and its predecessors and competitors, but it’s not as fast as the M1 and it draws way more power when it’s running. So I think this is an area they still have room to improve and consumers will really notice the benefit of it.

Intel 10th gen to 11th Gen is not really that different

In many single threaded benchmarks the 11th gen is 50% faster than the 10th. I think it’s a remarkable bump. The Zen3 has also brought a significant leap in single threaded performance, but not on mobile yet.

Most consumers don't understand computer "speed," and so any of these speed claims are either 1) marketing, or 2) aimed at enthusiasts who do understand. Most people seem to think that you own a computer for a few years, and then the computer "gets slow," and then it's time to buy a new computer.

This kind of condescension isn't helpful. People understand when an app or a site is slow, or doesn't work, compared to when it works well and quickly.

I legitimately didn't intend to be condescending. In my experience people don't understand that for a computer to be "fast," you have multiple possible bottle-necks: hard drive read/write, CPU, RAM, network bandwidth, etc. My very intelligent father-in-law (I mean this seriously, he's very intelligent, and was a programmer for years.) wanted to boost his internet speed, and upgrade from DSL to cable. (something like 15mbs to 100mbs) Well, he didn't really notice much speed difference. And this was because he was seldom, if ever, saturating his 15mbs link. In fact, on a modern webpage, there are so many new TCP connections that each one can be a bit slow despite bandwidth. (and this is to say nothing of the delay incurred by actually trying to render the js and other elements on the page.)

My point was that most consumers don't understand these things, and just assume that "computer speed" is one-dimensional: it's either fast, or not fast.

Yet they use Facebook and Instagram nonetheless.

Free dancing bears and network effects will always win.

Reply on news.ycombinator.com