Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The Great CPU Stagnation

databasearchitects.blogspot.com

Post reply on HN
Print view View original

161–170 of 222 posts

Re: The Great CPU Stagnation

#161

Post by bcrl »

borissk wrote: post #150

The slowdown of x86 processor progress was caused by the very slow adoption by Intel of new fab processes. 14nm CPUs came out in 2014, 10nm mobile only (Tiger Lake) came out in the end of 2020. Almost three years later still no commercial CPUs by Intel on 7nm. They used to be a lot faster, 22nm CPUs (Ivy Bridge) came out in 2012 - so around 2 years to go from 22 to 14.

Not really. Clock speeds for CPUs have stagnated because they hit the power wall back in the P4 days. Part of what enabled the relentless clock increased from the 1970s until the early 2000s is that each generation of CPU ran at a substantially lower voltage than the previous generation while being able to use more power. Going from 5 volts to 3.3 volts is a substantial power saving that gives you headroom to increase clock speed, but going from 1.1V to 1.05V is not nearly as much of an improvement. Once CPUs started dissipating 100-150 watts of power and voltages only drop by a few percent each generation, there's simply no power budget left to increase clocks. Current cores can run at upwards of 7 GHz (just look at the overclocking records) with liquid nitrogen cooling to dump the power increase, but virtually nobody will use such a system in the real world.

The reason we see increasing core counts is that 2 cores consume roughly 2x the power of 1 core (well, slightly more). Doubling clock speed from 5 GHz to 10 GHz would cost way more than 2x the power. Furthermore, pumping that much heat out of a sub-100mm^2 die gets harder and harder. You can have more transistors in a small space, but you simply can't have them change state much more than the previous generation did.

Re: The Great CPU Stagnation

#162

Post by NovaDudely »

jorvi wrote: post #142
lynx23 wrote: post #44

The perfect time to shed ourselves of the idea that "optimisation is a waste of dev-time". Mobile computing was the last godsent t actually rethink performance a little bit, but we still have a lot of relatively low-hanging fruit. I sometimes dream about a month-of-no-new-features, where everyone would just have a bit of time to clean up and improve on existing stuff.

This is why I loved every console pre-PS4, from a technical standpoint. The amount of performance (think God of War 2 or Black) devs were able to squeeze out of the PS2 was nothing short of staggering.

Because they are dirt cheap I recently bought a Xbox 360 and Ps3 and I agree, it is astounding even up til then just how much was squeezed out of those machines.

On the Ps3 side, I had completely forgotten just how hamstrung it was in terms of memory, the OS and the load times on games are SLOOOOOW! And yet, they got Uncharted 3 and Crysis 3 out of that thing.

One of the most impressive feats I have seen would have to be Daytona USA 2 in the arcades. That thing is running on a single 166Mhz PowerPC 603 and a Real3D GPU that still needed all the polygon setup done on the CPU. How they got that out of that hardware is just beyond me. It is optimised to the max!

https://www.youtube.com/watch?v=ttAz1mYTuv0

Re: The Great CPU Stagnation

#163

Post by flenserboy »

NovaDudely wrote: post #151
dietrichepp wrote: post #92

Earlier quoted context omitted.

A more extreme version of that is “permacomputing”— https://permacomputing.net/projects/ I still regularly use old systems for fun. I grew up with Macs, so that’s the point of reference. When I use these old systems, I pine for a few specific things I’m used to on newer systems—but it almost feels like nitpicking. The things that I really want from a computer are pretty basic. Like good, consistent copy/paste and dra…

I am always more than happy to see Permacomputing mentioned. I really think it is a wave of the future that will not be fully realized for another decade or so. Similar to Solar punk, it is still trying to find its feet and a lot of the most optimistic propositions will look silly in retrospect but the potential is there. It also kind of drives me up the wall when you see what was being done on computers in the 80's…

2+ decade-old word processors were fantastic at their jobs, and did 98+% of whatever needs to be done today. It would be great if old codebases could be liberated, cleaned-up, and turned into all-but-perfect tools for the job. Older spreadsheets, databases, etc., could also be renewed this way — but I doubt the corporations holding the rights to them would be interested in making this sort of thing happen. Too little in it for them, especially as (ideally) spy/tracking hooks would not be included in the package.

Re: The Great CPU Stagnation

#164

Post by NovaDudely »

Animats wrote: post #156
kayson wrote: post #2

We've known about this for a long time. Everyone expected it to happen. There are some key upcoming technologies that have the potential to cause a step in scaling (CFETs, backside power delivery) but it's still not going to be anywhere near Moore's law levels. I think this is part of why GPU power is skyrocketing and why Apple, Qualcomm, and the like are trying to shift towards services.

> GPU power is skyrocketing Not quite. NVidia's entry level price/performance has not improved much since 2016. What's skyrocketing is the price of the top of the line models.

When the Voodoo 2 card launched it was $299. Adjusted for inflation that would be about $550. The latest 4090RTX is RRP of $1599.

High end is now becoming a case of throwing us much money at the problem as possible.

Re: The Great CPU Stagnation

#165

Post by moosedev »

WaxProlix wrote: post #155
ilyt wrote: post #141

Earlier quoted context omitted.

my i7-4790 served me nicely for 8 years. Only upgraded because now most games were CPU-limited, even on "only" GTX 1070. But I went "all in" "I don't want to upgrade for quite a long time" with 7800X3D. Maybe GPU upgrade in 2-3 years...

Did something similar a bit ago with the 5900X and a 6900XT. Wonderful setup, low wattage, quiet, performant. Just lost it to a fire and not sure what to replace it with, alas. Too busy in the aftermath to justify building and parting stuff independently, so that probably leaves me with an off the shelf or a workstation + GPU combo.

Yikes. What started the fire?

Re: The Great CPU Stagnation

#166

Post by MindSpunk »

bhouston wrote: post #83

This "stagnation" is nothing like the stagnation during AMD's poorly performing Bulldozer era (the post Athlon era) where they were consistently beat by Intel's offerings and there was a general lack of innovation in the prosumer space. During that era for the most part Intel's i7 prosumer CPUs started with 4 cores with the Bloomfield Nehalem chips in 2008 (which at the time were awesome and a game changer) and ended…

Not saying you're wrong, because intel stagnation was very real, but please don't use userbenchmark as a source of numbers for anything. They're a terrible source with biased benchmarks and reporting. Just look at the dribble they write for basically any AMD product on their site (https://cpu.userbenchmark.com/SpeedTest/1817839/AMD-Ryzen-7-..., or this https://cpu.userbenchmark.com/SpeedTest/2081998/AMD-Ryzen-7-... which just find-and-replaces half the 5800X3D comments).

Re: The Great CPU Stagnation

#167

Post by crote »

mhh__ wrote: post #15

IPC data seems a bit sus

Even worse: IPC data is pretty much useless when it comes to comparing CPU performance!

How is it going to account for something like the introduction of AVX512, doubling the data throughput per instruction? What about increased cache greatly reducing the number of instructions waiting for memory access, or a faster memory controller? How would you even begin to quantify a new PCIe generation doubling SSD bandwidth, or a better core interconnect reducing NUMA penalties?

IPC is basically just a marketing number with today's CPUs. If you want to get a proper comparison, use a real-world benchmark.

Re: The Great CPU Stagnation

#168

Post by MichaelZuo »

infinite8s wrote: post #159
danielvaughn wrote: post #139

Earlier quoted context omitted.

We might need to block all software features for the next decade so we can figure out what's going on.

The C memory model underlying all modern processors does not scale. Most of the time your cpu is sitting idle.

My favourite comparison is that the A16 chip can do 1 million times more operations per second compared to the top of the line System 360 mainframe.

That is to say, a theoretically ideal OS running on the A16 could run every mainframe program in existence until the late 1970s, simultaneously.

Re: The Great CPU Stagnation

#169

Post by crote »

nologic01 wrote: post #47

In fact the CPU stagnation is worse. In contrast to the shrinking chip size and higher frequencies of past scaling laws, CPU core count does not easily translate to performance gains. It needs specialized and careful designed software to tap the capability. In fact if CPU core count did translate more easily to performance gains I think already with the existing CPU's we'd have a fairly signficant one-time boost. May…

I'm not convinced that this is the case.

A lot of computation-heavy workloads do in fact scale rather well with an increased core count. That's why GPUs are essentially using thousands of cores, after all: nontrivial computation virtually always implies a sizeable amount of data, and a lot of data usually implies that you can split it up into multiple sections to compute in parallel.

A lot of desktop code is still single-core, that is true. However desktop CPUs are idle >90% of the time, only waking up to do a small burst of computation every once in a while. You're not going to notice some event loop finishing in 0.8ms instead of 1ms. Even then it'll probably be running multiple tasks from multiple processes in parallel, making good use of the available cores.

CPUs have simply gotten fast enough that they aren't a bottleneck for desktop use anymore!

Re: The Great CPU Stagnation

#170

Post by crote »

lynx23 wrote: post #44

The perfect time to shed ourselves of the idea that "optimisation is a waste of dev-time". Mobile computing was the last godsent t actually rethink performance a little bit, but we still have a lot of relatively low-hanging fruit. I sometimes dream about a month-of-no-new-features, where everyone would just have a bit of time to clean up and improve on existing stuff.

That's still the case, though.

Like it or not, computation is cheap and developers are expensive. Code isn't worth optimizing until it either becomes a bottleneck, or you are running it on thousands of machines.

Post reply on HN