This is a fairly timeless sentiment. Even more ubiquitous is looking back and laughing at just how primitive formerly-sufficient technology was. You'd think we'd know by now to just not have this thought anymore!
I'm not talking about stuff you can't get to run faster. If people had the same kind of care they had in the 80s and 90s, the software would be incredibly fast. But these days they can be more careless.
Sometimes it's a good thing, as things ship faster and frameworks are more expressive. Other times, though, it's a PITA. Like when half the time we upgrade just to run the ton of bloatware from different vendors on the web / multitasking OSes at a decent speed. Chrome is usually the biggest hog on my machine. Or like when Apple's iOS keeps making the iPhone slower with every version, until you're practically forced to upgrade. Is it really that hard to make an OS which hardly does any major multitasking from getting slower and slower with every release? One almost wonders if they do it on purpose ;-)
The changes which have actually made a difference to my computing experience over the last decade were the arrival of affordable SSDs and a jump upward in LCD pixel density.
Everything is just...adequate. People don't log on to an older machine and lament how slow it is anymore, they don't balk at how huge the new hard drives are, and consumer-grade disk speeds havent gone up since 7200rpms were introduced. Video hardware is pretty much the only technology in most computers that still has reason to march on, because game developers can always add higher textures and cooler shaders.
Furthermore, while people aren't really after bigger screens, getting to a point where a laptop or desktop has similar DPI to a smartphone would require vastly higher resolutions for which there isn't the technology yet.
Almost nothing I want to do on computers runs fast enough for me, and I'm too stupid to invent good algorithms. But I haven't given up on solving hard problems, so naturally I will look around and see machines incapable of even getting started.
Maybe, at some point, we will come up with a use-case that needs more compute that what we currently have, but for most people's current usecase, compute is not the limiting factor.
Very few hard requirements are valuable today for most users: battery lifetime, hd video for skype/facetime. And maybe SSD, which are now large and cheap enough for people to switch, and get back the bliss of instantaneous reaction again. Other than that no important task will require more than a core 2 duo. They won't saturate Excel (which was not long ago still single core only). In the case they really do, that spreadsheet is probably worth enough to buy a full fledge MBP.
I could literally eat a 1000X speed increase without blinking.
I think everyone else could too. The difference isn't about speed, but about quality of interaction.
Today's minimal UIs are a good impedance match for the limited cycles available. iOS was particularly good at leading the way on this, creating a UI that was relatively undemanding of raw cycles but still created an unexpectedly immersive experience.
A big part of that was the move from type-and-point to drag-and-tap. The tactile mode, the skeuomorphism, and the portability and instant access compensated for having to squeeze the UI into a tiny screen.
It felt like a more responsive experience even though it was quite a bit slower and more rationed than equivalent interaction on a full-sized desktop.
That's still true today. You can get stuff done on a tablet, but the most efficient and productive UIs pre-ration the affordances. A desktop is much more flexible/general and gives you much more visual bandwidth.
If you apply this idea to hypothetical 1000X hardware, you won't get more of the same but faster - you'll get completely new modes of interaction, which will likely combine the immediacy of touch with the high bandwidth of desktop interaction, and take them much further than anything available today.
It's possible to imagine all kinds of science fiction experiences. The reality will probably still be a surprise.
Even so - the point is really that today's users feel like they're doing everything they want, because UIs are designed without affordances that would be slow and frustrating given today's hardware.
That will change with faster hardware and new designs, and phone UIs will seem as quaint as a 720 x 350 monochrome desktop PC screen running DOS.
on Windows: http://superuser.com/questions/565347/how-does-windows-limit...
On Linux: https://wiki.archlinux.org/index.php/CPU_frequency_scaling
You can also play with the BIOS, according to https://en.wikipedia.org/wiki/Underclocking
Computing isn't a limiting factor in almost anything we do (I occasionally code and test on a raspberrypi) but we do notice and appreciate the improvements that performance bring. Even naive users notice.
Try to encode high-resolution 10bit RGB 444 video and stream it to the internet. While playing a AAA game on the same machine. Try to get the latency as low as possible and make sure the video is smooth, without artifacts on the stream.
This is the kind of problem a lot of folks are trying to solve right now and the current hardware is barely good enough to keep up...and that's with adding highly specialized add-on cards.
Try emulating Sega ST-V games on your machine -- and realize quickly that accurate emulation is a matter of clock speed over most other factors...and which has mostly slowed down since the Intel C2Q era.
When you provide compute power/bandwidth/whatever, someone will find a need for it.
As technology becomes more centralized, the computing power of your phone or laptop becomes much less important than the power of Google's and Amazon's data centers. Your devices essentially become clients for computation done in the cloud.
Technological advancements generally don't require increases in the client's computing power anymore.