I was ticked off that my 2007 Mac Mini couldn't be upgraded to Mountain Lion, until I realized Snow Leopard ran all the software I needed on that box. I think I'm happy with the hardware and form factor of my phone, too, so I've got all the electronics I need for years to come. Good thing, too, because my rent just went up, and I need a new couch.
Basically this, computers hit good enough a while ago, now you just have to replace parts when they die.
Yes, on paper, the latest processor is faster than the one released two years ago but you have to be doing specific types of workloads with it to really make a big difference.
> When your processor is too slow, buy a new CPU, or you get a new heat sink and over clock it
The motherboards for PCs built 5 years ago are completely different from those built today, and the CPU sockets have changed every other year. New processors from Intel will be soldered on.
The performance of a PC from five years ago is probably adequate for web browsing and office tasks. For anything more demanding, the advances in power consumption, execution efficiency and process node are huge leaps from five years ago.
> The performance of a PC from five years ago is probably adequate for web browsing and office tasks. For anything more demanding, the advances in power consumption, execution efficiency and process node are huge leaps from five years ago.
Anecdotally, my machine is approaching four years old and I can still run just about every new game on the highest settings. At just under $1,000 at the time, this isn't a luxury yacht type rig, either.
The performance of a PC from five years ago is probably adequate for web browsing and office tasks. For anything more demanding, the advances in power consumption, execution efficiency and process node are huge leaps from five years ago.
In 2000, a three-year-old PC couldn't run modern games. Today, a three-year-old PC simply forces you to switch to lower graphical settings. The race for hogging all the hardware resources did slow down in the past years, which is a good thing for consumers.
Well, the CPU/RAM/HDD systems do last a very long time. It's the GPU that needs periodic upgrading. Robert Space Industries for instance will be leveraging the Cryengine 3 with nearly 10 times as many polygons as with the average 3D FPS. Also, Microsoft keeps adding rendering features to the latest OS's which require hardware updates on the GPU level. I guess what I'm saying is: Nvidia will continue to be a sound stock to add to your portfolio.
This reminds me of a piece I wrote a couple years ago: http://jseliger.wordpress.com/2011/10/09/desktop-pcs-arent-g... , which makes a similar point. Both articles are less screechy and less likely to get readers than screaming headlines about OMG DEATH!!!
I built a dev/gaming machine back in early 2010. It's stout, but not a ridiculously expensive (~$1,000) behemoth. The only thing I've done since then is toss some more RAM in so I could have two sets of triple channel DDR3 instead of one. I can still run just about any modern AAA game at the highest settings.
The only time I felt like I've needed an upgrade is while playing Planetside 2, which is/was very CPU bound for my setup. However, when it was initially released, Planetside 2 ran like a three-legged dog even on some higher end rigs. It's much better after a few rounds of optimizations by the developers, with more scheduled for the next month or two.
I dual boot Linux boot on the same machine for my day job, 5 days a week all year. For this purpose it has actually been getting faster with time as the environment I run matures and gets optimized.
As good as it is now, I remember struggling to keep up with a two year old machine in 2003.
~$1,000 sounds quite cheap for a gaming machine which can still run any modern game at highest settings. I remember I built mine at about the same time (2010) for $2500, top notch video card, fastest cpu, lots of ram, but its 2013 now and I can not say that it runs any modern game at highest settings.
It was surprisingly easy. I'm not saying you over-paid, but for $2,500 I could have built something pretty ridiculous. Most of my money went into processor and GPU, which are typically your two big ticket items.
I trolled around Newegg looking for upper-middle tier components with a higher quantity of good reviews. A lot of the times you won't see a lot of the recently released stuff with useful reviews, so some of the parts were actually circa 2009'ish instead of being latest and greatest (2010).
I didn't splurge for a super expensive case, and my power supply wasn't modular (making it pretty cheap). i7 with a decent mobo. Went AMD for the GPU since (at the time) they were the best bang for buck. Got some cheaper G.SKILL 1600mhz DDR3 RAM (which has worked awesomely for me) for next to nothing, and I was ready to roll.
Looking at the Tom's Hardware system builder challenges from a year ago or two years ago, and the comparing them to the $500 and $1000 machines that they have now in their System Builder marathons should showcase the tradeoffs in performance vs price more starkly.
More like impossible. The Internet has many PC building communities and none of them could do it. Hell an i7 and good GPU at the time easily brought you over 600$, which means very little money for a motherboard than can keep up.
Not the parent poster, but I also bought a computer in 2010 for about $1000 and am very happy with it today. I still play games at max or near max settings with no problem. This does not include a monitor though; also I only bought a 1650x1050 monitor, so I save a bit on graphics power by not having the full HD pixel count (or more).
EDIT: don't know my full specs offhand; i5, gtx 460, either 4 or 8 gigs RAM (4 I think), a 120gb SSD and 2G HD
My i7-875K (overclocked, but not exhaustively; my Gigabyte motherboard has an auto-overclock function that raised the multiplier without killing stability) with 16GB of RAM, entirely too many SSDs for any one person, and a Radeon HD6870 is capable of playing every new game I've bought this year (Dishonored, Saints Row IV--I don't buy games from EA so the new title list is pretty short right now) at max settings at no worse than 35FPS--I personally can't detect a difference between 30FPS and 60FPS, so I don't care.
Neither Dishonered or Saints Row IV are particularly intensive games. I get in excess of 100FPS in both and my cards (GTX 670s SLI) barely hut 30-35% usage…
I don't doubt the OPs rig is powerful enough to play modern games, but I seriously question the "any modern game at highest settings" statement.
FWIW, I built a rig a great rig in 2010 which I still use from time to time. i7 920, 12GB DDR3, SSD, SLI GTX 460 2GB.
I bought a ~1000 setup in 2009 and it still runs most games acceptably (not necessarily highest settings).
it was a Dell boxing day deal $500 computer + $350 video card (GTX 260) + changed the case and PSU.
Maybe the OP doesnt play some of the most demanding games?
> Maybe the OP doesnt play some of the most demanding games?
I play some pretty demanding games, but I was able to get a lot of performance per dollar building custom. I knew where it was OK to spend more/less and did a lot of research/shopping around.
Well yeah, something like the Titan has no price / performance at all. Usually the best valuation is around $200, ie, an r7 270x or a 760. And if you want a premium card, 280x / 770 are both devouring modern titles. Anything past that is "I want the best there is right now, screw the cost".
That's not the point. He said he can run any modern game on high settings on the computer he bought 3 years ago, I doubt you can play any game in 2016 on high settings if you buy a computer for $800 today.
Why does no-one ever factor in the costs of transferring systems in this upgrade process? It's not trivial to set up a new system and bring your stuff over.
IMO, you don't need the fastest CPU for gaming. I saved a ton of money by getting an AMD FX-8350 (I already had an AMD motherboard on hand, so that helped too), and although it does bottleneck in really CPU-intensive games, I can run almost all games on the highest settings at 2560x1440. The graphics card is really the only component that you need to splurge on for a gaming rig these days, because it's the bottleneck for the vast majority of games.
It's not even GPU today. RAM is pretty much the bottleneck for everything. CPUs have been so much faster than RAM for so long now that they have to keep doing more and more tricks just to keep utilization up at all. With gaming, it's always texture memory and transferring from RAM to the video card that's expensive.
In terms of gaming performance, when you go from mid range machine to high end machine, you are often spending 2-4 times as much without 2-4 times the performance. You hit diminishing returns big time.
I go with generally mid range components with my gaming machine. Even then, I upgrade every 3 generations for CPU and every other generation for video card. CPU performance doesn't impact gaming as much as it used to.
This gives me reasonable performance in most games around high to ultra on a 1920x1080 monitor.
Bingo. And there's now some pretty fierce competition between Nvidia and AMD again, so you can often catch some pretty good deals on a GPU or a combo if you pay attention.
I still use an AMD processor on my gaming/development rig, because upgrading to a better Intel CPU would have cost me nearly twice as much (I already had an AMD motherboard, so that helped). I don't notice a bottleneck in most games.
I think the best strategy is to get mid-range components and just upgrade more often. You get way more bang for your buck, and you can always sell the old components on eBay or whatever.
2010? If you got an i5-2500K and a 580, you'd probably get a PC that came in around $1200-$1300. If you could reuse components like cases, hard drives, PSUs etc. you could easily get that in under $1000. That should still run 90% of modern games maxed out at 1080p, with the exceptions being pretty much just Metro 2033 and Planetside 2.
If you're doing multi-monitor or >1080p resolutions, then you might need to get something better than the 580 however.
Im running a 2500k at 4.3ish with a CM 912+ in push/pull and happily devour pretty much anything. I don't foresee a CPU upgrade anytime soon. I had a pair of 6950 (the older 1 gigs) and just put in a 7970 and another 4GB of RAM for less than 400 bucks and only did that because BF4 is a resource monster. Everything else was fine.
> I can still run just about any modern AAA game at the highest settings.
AAA games mostly target the console. Look at GTA5, which isn't even out on the PC. Most AAA games will run on a PS3, which came out in 2006, and has 512MB of RAM (combined system / graphics).
That said, there's a point of diminishing returns - making games look much more realistic will take obscene amounts of resources.
I expect there to be a system requirements explosion for PC games now that the new consoles are right around the corner and AAA game developers can finally target much higher specs.
An interesting opposite is the previously mentioned Planetside 2. It was initially released for PC only, and was a resource hog. Now they're working on bringing it to the PS4, so they're having to do an extremely aggressive round of optimizations to make it run decently. The optimizations will get a lot of testing on the PC version (think it's hitting the test server next week). PC players will benefit as a result of the console port.
Planetside 2 is a weird example, though. I don't think there will be many games that have a 1-year+ delay from landing on PC before they hit consoles.
I built mine in early 2007 for around $2000 (+-$300, can't remember exactly) and it's just now really starting to show its old age. It could use more RAM and an SSD (maybe), but for 99% of what I do, including gaming, it's plenty fast enough. I can't run the very absolute latest AAA games on highest, but if I turn the resolution down a hair or turn off antialiasing they run fine.
In fact the only thing I really want a faster machine for is some of the latest emulation techniques (Higan) and a vague desire to play around with some virtualization odds and ends.
I guess this will change for a little while as updated consoles come out and games can improve their graphics as a result, and also when 4k monitors start coming out. But yeah, until those two things come into play, older computers still play games just fine.
Like another poster said, with "game" terms replacing "data": " But, if you can run a [game] and it's [good framerate], it just means you should really be doing better [gaming] over more [pixels]."
Mainly we don't need new ones because the 3 year old one is still doing the job. That wasn't the case a decade ago - your 3 year old PC was seriously out of date and couldn't run most games released that year and probably not install the latest OS release. This rapid progress has flattened out considerably. Now people upgrade to get nice features such as retina displays or SSD drives, but that's optional (so you don't do it if you don't have spare money laying around) and the benefit is much smaller than going from a 90 MHz Pentium to a 450 MHz Pentium III.
I still use my 5 year old desktop (upgraded twice) for development. I like to open box and upgrade it myself , if I want to do similar on laptop I think twice . Freedom to upgrade it yourself is a bliss.
Wish I could downvote this. Snark like this is toxic and is neither useful nor interesting.
A lot of folks reading sensationalist articles about the PC market decline are making the conclusion that nobody likes or uses PC's anymore since sales are declining. The author is pointing out that it's a poor conclusion to make since there are other factors contributing to the decline, including the fact that the usable lifespan of today's PC's is longer than it used to be.
Personally, I upgrade incrementally, and I still use my PC on a regular basis. The machine I have now is a hodge-podge of parts from different ERAs. I have an Intel Q6600 but DDR3 RAM, and a modern, quite beefy graphics card that I bought when it was in the upper-echelons in early 2013. It runs most modern games pretty well. I have an SSD for most software but also three big HDDs, one of which I've had since my first build in 2004.
It used to be that things got faster at a much faster rate. And until this new E5-2690 v2 was released, the fastest CPU was only 14000 or so, which is less than 2x as fast.
It was pretty typical to get +4x performance in less than 2 years before after an upgrade cycle. The problem is that encryption standards aren't increasing at a logarithmic rate like CPUs were, so DoD had to step in and adjust the marketplace. By 2010, it was well known in certain cirles that the i7 would be the last serious CPU in the desktop space, that future CPUs would be targeted to cloudian budgets and applications, where they can be managed and hidden away from people who might use real modern cpus you've may have heard about like Tile-x that are used by NSA to do the very thing they were afraid normal people might do, like break https, farm personal data and use it to affect the market, and basically have control over their lives outside of the financial system. Now that surveillance state is complete, they don't really have to worry about drug dealers breaking away from the financial system anymore.
CPUs are actually getting slower after overclocking
I have water cooled 2600k at 5GHz running 24/7. Bigger GPU and more cores brings additional complexity and more heat. No other CPU can handle this without blowing up.
Depending on what you're doing with CPUs there have been improvements, even at the high end between Nehalem and Haswell (on the Intel side). Anandtech's data shows a nearly 2x improvement between those generations for certain tasks.
But the reality is that for other tasks (e.g. the winrar compression benchmark) the speedup is minimal compared to previous years (see Haswell vs Prescott). In many cases where there is a speedup, performance is already so good that you wouldn't notice.
It's pretty crazy to see that in the consumer world, not even games are demanding so much performance from our CPUs.
the article is falling under fallacy of assuming the wrong sample. Of course the author wouldn't buy new PC because he can upgrade his old one. Heck, almost any tech-savvy people can in fact upgrade or build one from the scratch. If not, chances are that you know at least one person who can help you and after the first time, it just gets easier.
the PC market isn't dead, it is slowly receding and it won't stop. It's because of the new alternatives, and assuming finite budge, when you get one of the alternatives, which cost roughly around a consumer-level laptop, you don't have enough for another PC that you don't need.
The article to me seems extremely narrow in both its oversight and scope. People don't care about processing power not because it's a marketing gimmick, but because they don't care. People who do care are the ones who know enough to care, and they will always be minority.
For many of the most common business and consumer applications (web browsing, using MS Office), adding a more powerful computer just doesn't do that much for you. Power users will upgrade their old machines, but most users will just keep using them.
I'm typing this on a PC where I did the same thing as the author. Over the past 10 years, I've swapped out a part every two years or so to keep it running the latest and greatest. But the CPU is five years old and still running fine. I'm planning to donate it to a non-profit to replace a computer that's almost 10 years old and also still running fine.
There was a time when you felt like a new PC was obsolete the second you took it out of the box. But that was because we were just scratching the surface of what we could do with new hardware. We're now at a point where it's hard to find consumer and business applications for all the spare hardware that you can afford.
Mobile adoption has been so quick because everyone is buying devices for the first time (tablets), or there is an incentivized two-year replacement cycle (phones). But I'm still using an original iPad that works just fine, and a 3 year old cell phone with no reason to upgrade. Eventually, I think we'll start to see the same leveling off in mobile as well.
Tablets need to be very small, lightweight and thin. Making them modular and configurable would go against that.
Look at some ipad teardown for instance and watch how everything is packed together. Not to mention that these days everything is in SoCs: you can't upgrade the GPU and CPU separately. Even the RAM can be stacked on the SoC package and if it's not it's soldered right next to it.
Contrast that with the innards of a desktop PC which is mostly empty space.
We are going in the oposite direction of #2 for at least a decade. Nowadays a motherboard has almost all the periferals you'll need in a typical PC, and at the ARM world (that is more flexible) we are getting into the periferal-less single chip CPU that has only a power supply and I/O pins for display, USB and network.
That's a natural step because connectors are expensive, power hungry and bulk. As computers get inexpensive, upgrading makes less sense, so it makes less sense to pay more for an upgreadable system.
"For what" is the obvious question. Web development with a remote testing environment, office applications, email, web browsing - sure, a Core 2 Duo is more than good enough if your software environment is kept in order. Audio / video / photoshop, gaming, developing software that does math, data analysis - you can never get fast enough.
The limiting factor is if your computer's feedback loop is tighter than your brain's perception loop. If you can type a letter and the letter appears, your computer is fast enough for word processing. But, if you can run a data analysis job and it's done before you release the "enter" key, it just means you should really be doing better analyses over more data. Certain use cases grow like goldfish to the limits of their environment.
Very true, but the vast majority of computer users aren't pushing the limits of their systems, and I think that's what the author is getting at. If you look at the market as a whole, the need for more powerful computers isn't nearly as big as it used to be.
And those users are overwhelmingly the ones that tablets suit just fine.
That's the problem in this type of "PCs are fine, people just don't have to upgrade" argument. (And I've made it myself, before I really worked it through and took stock of just how far tablets have already come.)
The use cases satisfied by an old PC could all be satisfied just as well by today's tablets. Yes, the software support for keyboards, external displays and server-type services isn't quite there. But that's solely a software limitation. Not a limitation of the hardware platform. It will be addressed and more PC sales will disappear.
And the actual remaining PC-justifying use cases are precisely those where an old core2duo isn't "good enough" and simply buying some more RAM or even an SSD isn't going to obviate the need for new hardware for 5+ years.
So inasmuch as people still need PCs, they have to show up in the sales charts. And inasmuch as we don't, mobile devices are going to eat that market as the software evolves.
There's really no way around it.
Frankly, I think android and/or iOS are one good upgrade-cycle away from decimating laptop sales. If either or both really focus on the 'docked' use case -- better support for external displays and keyboard support to facilitate top-flight bulk text entry/editing -- laptops are dead to the general public.
If a student can so much as use a keyboard case and tablet to write a term paper or code up some homework as easily as on a laptop, it's over. And there's no hardware preventing that. It's software. If Microsoft wasn't so inept, they'd have been way out ahead on this.
And, frankly, between the potential of the docked use case and "BYOD" policies, mobile devices could seriously eat into the enterprise desktop market as well.
You build a device cradle, that connects to an external display, power and a keyboard/trackpad, and enables the OS and apps to automatically sense that connection and shift their interfaces accordingly and you'll see how few people truly need a PC anymore.
I completely agree. I wasn't trying to say that PCs aren't dying, I was just trying to contend the notion that power users make up a significant portion of the market we're talking about. In fact, I think tablets directly show that the vast majority of consumers don't need powerful hardware, because tablets aren't very powerful machines at all.
I love the idea of device docks and I would very much like to see things go that way.
I think docks have some hurdles to jump in the consumer market, though. They're unsexy - I think most people associate them with boring enterprise computers and boring enterprise work, and they tend to be big and ugly. They're expensive too, considering that most people view them as nothing but plastic with a few connectors and wires inside.
If someone can figure out how to jump those hurdles, though, and make docks sexy to the consumer market, beige-box sales will plummet. Make them smaller, easier, cheaper. Make them less proprietary - at the very least, a Company XYZ consumer dock should be compatible with all/most of of Company XYZ's portable offerings. Make them wireless and automatic (OK, I realize this conflicts with "cheaper")! Let me drop my Win8 tablet on my desk and immediately have it display to my dual monitors and let me use it with my mouse, keyboard, printer, big external drive and wired network connection. Let me press a button and have it display to my TV or play to my big speakers. Have your cake and eat it too.
I mention Win8 specifically because the whole idea of it is that it's a tablet and a desktop OS in one. Why on earth is there no Microsoft-produced drop-in docking solution for the Surface so it can actually be both?! The consumer potential is crazy - got a crusty old desktop PC laying around? Toss it, keep the peripherals, buy a Surface+dock and you've got a new tablet/laptop and desktop. You can "dock" a Surface as-is with two cables (a micro HDMI and a USB hub with all your desktop peripherals wired in), but that leaves out wired network (maybe important), certain peripherals like speakers (maybe important), and power (critical), and even two cables is two too many.
Most people who say they need a PC don't need a beige box, they need a keyboard, mouse and monitors on a desk where they can work. The form factor of the box that the CPU, disk and memory come in doesn't matter when you're at the desk, so it might as well take the form of something you can take with you and use when you're not at the desk: a laptop, tablet or phone.
Docks will take off when you can make them wireless.
My tablet can connect to my wifi when I come home and sit at my desk. There's no reason at all why it can't connect to my mouse+keyboard at the same moment; so someone needs to solve the technical problems of it being able to connect to my large monitor wirelessly, and we're set.
Emphatic agreement. I wind up helping folks a lot with writing high performance software, and it's very easy to get to the point where the time to run a model is totally determined by how fast the CPU and IO are. I'm talking problems where naively written C would take an hour, but if I'm careful and use some blend of very clever c or Haskell, the computation is done in 2-5 minutes
It's not neccesarily which language is faster, but the which algorithm is faster.
He said naively written C, which mean the algorithm may be entirely different and run in O(n2) and much slower than a Haskell version which use a different algorithm and run in O(nlogn).
actually in the specific example I'm thinking about, i'm talking about memory locality being the performance difference (and in this case, array layout for matrix multiplication).
The naive obvious "dot product" matrix mult of two Row Major matrices is 100-1000x slower than somewhat fancier layouts, or even simply transposing the right hand matrix can make a significant difference, let alone more fancy things.
Often the biggest throughput bottleneck for CPU bound algorithms in a numerical setting is the quality of the memory locality (because the CPU can chew through data faster than you can feed it). Its actually really really hard to get C / C++ to help you write code with suitably fancy layouts that are easy to use.
Amusingly, I also think most auto vectorization approaches to SIMD actually miss the best way to use SIMD registers! I've actually some cute small matrix kernels where by using the AVX SIMD registers as a "L0" cache, I get a 1.5x perf boost!
oh, thats just me rambling about why i don't trust compiler autovectorization :)
well:
1) i've been slowly working on a numerical computing / data analysis substrate for haskell for over a year now.
2) the haskell c ffi is probably the nicest c ffi you'll ever see. Also pretty fast, basically just a known jump/funcall! And no marshaling overhead too!
3) theres a lot of current and pending over the next year work to make it easy to write HPC grade code using haskell. Some approaches involve interesting libs for runtime code gen (the llvm-general lib for haskell is AMAZING).
Theres also a number of ways where ghc haskell will likely get some great performance improvements for numerical code over the next 1-2 years! (i've a few ideas for improving the native code gen / SIMD support over the next year that I hope to experiment with as time permits)
in machine learning, as soon as you stop using linear models which are normally quite fast to compute, models will be as slow as you can tolerate.
for example: random forest, gradient boosting, gam, etc -- you will typically do parameter searches and the models you get are as good as your willingness to wait. Good software will run at a significant fraction of memory bus speed and the faster that bus goes the better your models will be.
exactly! This winds up being a memory locality / layout problem often times!
eg: most CPU memory traffic is in "cachelines" size chunks so you're best off trying to organize information so you can use all bandwidth! I've a few ideas on this i'm trying to bake into an array/matrix library i hope to release soon. :)
In what planet? I'm not even going to use myself as an example because I do other heavy stuff with my PC, I'm going to use my non-tech friends: one of them got a new laptop with 8GB of RAM, why? because she was complaining about webapps using too much memory and slowing down her previous system.
Regular users don't know or care about memory management, they don't even close old windows or tabs, its about convenience. That's not a problem in mobile where the need is the mother of invention so mem management is automatic and chrome reopens the tabs you had by itself, but in a desktop environment (specially windows) one wrong click and the session restore in chrome wipes your previous session.
But it was cheap, cheaper than an unlocked iphone and it gets the job done so its ok for her.
> one of them got a new laptop with 8GB of RAM, why? because she was complaining about webapps using too much memory and slowing down her previous system.
I suspect a few people on HN will be reluctant to see their role in this arrangement. ducks
We built really powerful computers and decided the best way to use them was to run web apps consisting of a poor performance scripting language with a poor performance visual-rendering language and half the people who build with it seem to think that anything done in a web browser is free.
"Modern" CSS for even a simple site is pretty silly. So many horribly inefficient rules, sometimes dozens of levels of overwriting properties, etc. And gobs upon gobs of Javascript that does very little but is constantly running, checking input boxes, animating every tiny little detail, doing things that can be done without Javascript, etc.
I have better system performance when running a Windows 7 VM, a semi-bulky image editor, or compiling the kernel than I do with a few bulky webapps in Firefox. And this is on a desktop built just 4 years ago.
I have a light-weight Linux setup that uses 80MB of RAM after logging in. It has 1GB of RAM. I can't run GMail and Facebook Firefox and browse the Internet at the same time with minimal open tabs. It's sad.
It's not like it's constrained to the web either. I had a bit of an epiphany about the state of desktop Windows a few years ago when for a hobby project I wrote a UI using only Win32 and C and no external dependencies. Maybe it took longer to write than it should have but I was really shocked at how the thing ran so much faster and smoother than just about any UI I was using on that machine. It dawned on me that all those layers of MFC, WinForms, WPF, Silverlight, WinRT and whatever else they'll come up with in the name of developer productivity are no match for a set of APIs designed to run at 25MHz with 8MB of RAM.
This is believable, but I'd like to dig a little deeper. Do you remember which Windows apps you were comparing yours against? And on what version of Windows?
This was Windows 7, I am pretty sure with dwm disabled, and it was comparing (albeit informally) against every Windows app I used at the time.
I know that some newer frameworks take advantage of GPU features (actually I used to work as a dev in the Windows org) but guess what, GDI is still faster.
I can pop-up Activity Monitor any time in the day, with multiple apps open, and it's guaranteed the app using the most memory is a browser (any browser).
As a data point: Safari is using 600+MB now with just HN and GitHub open. It's using more memory than all the other apps (editor, terminal, mail, Rdio, Skype) combined. 600MB is not much by today's standards, but comparatively, is ridiculously wasteful. It's a damn CD.
Hope the Mavericks update improves this a bit since I'm short on RAM on this machine (4GB).
Mavericks does amazing things to memory usage. My machine has 4GB as well, and safari went from 1GB+ memory usage to 200MB memory usage for the same amount of tabs. Under ML, I was swapping constantly; now, no swap at all.
I have 16MB, and usually have dozens of windows/tabs open in Safari, Xcode, Textmate2, and a few miscellaneous apps. Currently, Activity Monitor shows 15.9x MB used, 10.87 App, 1.26 File cache, 1.91 Wired, and 1.94 Compressed. Swap used 0 bytes, virtual memory 20.12 GB. No noticeable hesitations attributable to compressing/uncompressing when switching windows/apps.
4GB is plenty of RAM for a web browser with many tabs, especially with an SSD for backing virtual memory. If it isn't there is a memory leak, and if there is a leak, 8GB is as bad as 12GB.
My MBA has 4GB and I constantly forget, because it is enough, and I am constantly surprised that it doesn't have 8GB.
All modern web browsers save and restore sessions across restarts.
> 4GB is plenty of RAM for a web browser with many tabs,
I always up-to-date Firefox, so hopefully the memory leaks are minimal. I reboot my Linux machine every month-ish and don't exit Firefox unless necessary. I always keep a few "app" tabs pinned, only a couple of them are heavy. I have 4GB of RAM and turned swap off for an experiment. Sure enough, after a couple weeks of browsing and keeping about 25 tabs open, Firefox's memory consumption would creep up and up and eventually it would crash.
I can only guess/hope that there are memory leaks involved.
A bit over a year ago RAM was cheap enough you could just stock up. I have 16GB of RAM and paid 80€ or something for it. I keep stuff in tabs or windows now I used to stick in bookmarks. Right now I have Visual Studio open after work as I want to do a deployment in ~3h. No need to close and re-open the rather big project later.
Nothing is ever good enough for a development box when you use an IDE and work with larger and larger projects.
Faster CPUs, more memory, and faster storage are always welcome. I look forward to the day when Eclipse and other IDEs really start taking advantage of GPU stream processors for indexing and validation.
Indeed. I work with RAW photographs fairly often, and simply exporting an album with a few hundred RAW files to JPEG takes a surprising amount of time with fast, modern hardware.
None of those use cases are going to grow the PC market. The things you're describing have always been a niche (remember Workstation Class PCs?) that may add a few $$ to the bottom line, but they are not going to drive growth.
The PC market has relied on end users - consumers and business users - for it's growth engine for decades, and that appears to be drying up. One of the reasons for that is outlined in the article, for most use cases we don't need faster.
Even with gaming there isn't as much of a push as there used to be to constantly be on the cutting edge. This is mostly do to the fact that the industry as a whole focuses primarily on consoles first now and thus consoles tend to be the gating "LCD" target. If your PC is at least as good or a little bit better than a console released in 2005 or 2007 you're set. Of course, there will soon be a bump forward here with the next gen Xbox and Sony systems coming out in a month.
I fit into a lot of the special cases here: Developer, gamer, amateur photographer with many gigabytes of RAW files and even I don't feel the need to upgrade systems like I used to. Now it is about an every 3-4 year thing whereas previously it was yearly or more.
Below is what I feel is a relevant excerpt from Text of SXSW2013, Closing Remarks by Bruce Sterling [1]:
---
Why does nobody talk about them? Because nobody wants them, that’s why. Imagine somebody brings you a personal desktop computer here at South By, they’re like bringing it in on a trolley.
“Look, this device is personal. It computes and it’s totally personal, just for you, and you alone. It doesn’t talk to the internet. No sociality. You can’t share any of the content with anybody. Because it’s just for you, it’s private. It’s yours. You can compute with it. Nobody will know! You can process text, and draw stuff, and do your accounts. It’s got a spreadsheet. No modem, no broadband, no Cloud, no Facebook, Google, Amazon, no wireless. This is a dream machine. Because it’s personal and it computes. And it sits on the desk. You personally compute with it. You can even write your own software for it. It faithfully executes all your commands.”
So — if somebody tried to give you this device, this one I just made the pitch for, a genuinely Personal Computer, it’s just for you — Would you take it?
Even for free?
Would you even bend over and pick it up?
Isn’t it basically the cliff house in Walnut Canyon? Isn’t it the stone box?
“Look, I have my own little stone box here in this canyon! I can grow my own beans and corn. I harvest some prickly pear. I’m super advanced here.”
I really think I’m going to outlive the personal computer. And why not? I outlived the fax machine. I did. I was alive when people thought it was amazing to have a fax machine. Now I’m alive, and people think it’s amazing to still have a fax machine.
Why not the personal computer? Why shouldn’t it vanish like the cliff people vanished? Why shouldn’t it vanish like Steve Jobs vanished?
It’s not that we return to the status quo ante: don’t get me wrong. It’s not that once we had a nomad life, then we live in high-tech stone dwellings, and we return to chase the bison like we did before.
No: we return into a different kind of nomad life. A kind of Alan Kay world, where computation has vanished into the walls and ceiling, as he said many, many years ago.
Then we look back in nostalgia at the Personal Computer world. It’s not that we were forced out of our stone boxes in the canyon. We weren’t driven away by force. We just mysteriously left. It was like the waning of the moon.
They were too limiting, somehow. They computed, but they just didn’t do enough for us. They seemed like a fantastic way forward, but somehow they were actually getting in the way of our experience.
All these machines that tore us away from lived experience, and made us stare into the square screens or hunch over the keyboards, covered with their arcane, petroglyph symbols. Control Dingbat That, backslash R M this. We never really understood that. Not really.
---
This isn't universally wrong, but it is dead-wrong with present technology. And covering our eyes and pretending that we are currently that advanced doesn't make it so.
Because while basic computation is a universal commodity, what is implemented on top of it certainly isn't - a piece of software always functions as someone's agent. When the systems you end up relying on are entirely defined by someone else, the only thing that represents your will is your mind, and it is effectively executing a complex and ill-defined protocol against always-diligent computers.
You've done the computational equivalent of declining a lawyer.
But it doesn't seem like a big deal, since you're only compromising a little at any given time. But the software is always changing in ways that benefit its controllers while your expectations are mostly based on the capabilities that they've presented. So the progress you perceive is entirely in their desired paradigm. Features that would benefit you but at the expense of Google/Apple/etc are never explored, because you aren't the user of their software - you're its working set!
I can forgive the old-timers who were conditioned by broadcast media to see the world in hierarchal take-it-or-leave terms and don't understand what they're losing by sharecropping in walled gardens. And I can mostly forgive the unclued herd that just buys whatever is advertised.
But for everybody who knows the power of personal computers yet pretends webapps and locked appliances are actual progress, either out of personal laziness, cognitive dissonance, or longstanding need for social acceptance: shame on you for abandoning that self-determination you tasted the first time you truly experienced computing.
Good thoughts here. I am optimistic due to things like raspberry pi and arduino becoming popular, but at the same time these things just feel like this generations version of building radio kits and model airplanes. Ultimately it's the data rather than the computation that's important.
Yeah exactly, it's not that general purpose CPUs will be outlawed like we worried about in the 90s, it's that the generally popular ways of using technology won't be using their capabilities.
It's not just the data itself, but really about protocols used to access the data. Protocols mediate between parties, and by choosing to download a binary blob simply to check your email, you've given up any true bargaining power in that exchange. You still have some autonomy by hacking the blob (userscript injection, etc), but you're only building on unstable ground.
It's not that people don't need a new PC because their old PC does just as good a job as it did 5 years ago. It's also not because your average mom and pop are upgrading their own rigs themselves that new PC sales are slow.
It's that when tablets hit the scene, people realized they don't need their PC for 90% of what they do on a "computer". Email, social networking, shopping, music, video etc.
Us old geeks who swap hardware, play PC games, tweak OS settings and generally use yesterday's general purpose PC will be the ones remaining who keep buying new hardware and complete machines.
The general public meanwhile will only buy a PC if their tablet/smartphone/phablet needs expand beyond those platforms.
The market will shrink but it will turn more "pro". The quicker MS evolves into a modern IBM the better.
If 10 % of their needs aren't met by a tablet, they need something more that do, which is my problem with recommending a tablet over a PC. Sure, buy a tablet and use it 90 % of the time, but own a PC as well for when the tablet isn't enough.
> It's that when tablets hit the scene, people realized they don't need their PC for 90% of what they do on a "computer".
That's exactly what I see everywhere. People still need a PC for the 10% of the tasks they can't do on a tablet. So, they'll keep a good enough PC around all the times.
What is a different situation from a few years ago, when people were buying a PC for each family member. Now, it's a tablet for each person, a PC for each home, and spare space at the desks.
Replace 90% by 100%. The use case of the PC at home is disappearing fast. What can you do on a PC that you cannot do on a tablet? If you really think about it, not much at all. On the iPad Apple's creativity apps [1] go a long way: the basics for the amateur who wants to create content are covered. At home the only thing I actually need a "PC" for is to import CDs, I cannot think of anything else.
The rest is a matter of taste, I personally prefer PCs (well, Macs) but I can definitely understand people preferring tablets, if only because they are so much cheaper, and generally so much easier to use (to my astonishment my daughter figured out how to use my iPad to watch videos on YouTube while she was still 0 years old, before she could even speak...).
When it comes to work I need a computer. And event that might change in the future [2].
> What can you do on a PC that you cannot do on a tablet?
Easy copy pasting, selecting specific parts of a document... basically anything requiring precision and speed. You can add a keyboard and a mouse to it I suppose, but at that point you may as well have got a laptop.
That's just convenience. Do you really think many people will keep forking $X00 for PCs just for a little bit of convenience needed only once in a while? I respectfully disagree. A couple of anecdotes, for lack of evidence:
My grandfather (89 years old today) switched to the iPad. He was one of the pioneer of computer usage in France (in the steal industry)... at a time when programs were punched on cards! He retired before the mouse/keyboard thingy became popular, and never quite managed to fully grasp it when he got a computer ten years ago. He is not switching back, the iPad is way easier for him.
My parents went full iPad without knowing it: after getting one a year ago they realized that they just aren't using their laptops anymore. The iPad is just much nicer to browse the internet. In short the iPad is better for them 80% of the time, and worse 20% of the time. It can only get better with time, as we get better at making touch interfaces.
Don't get me wrong, I'm not advocating anything, I just think that's where the market is heading, and it's not coming back to the PC world I (and probably you) grew up in. I for one will keep using my laptop... but I know I'm in the minority.
Well, that "bit of convenience" varies in size a lot over the market. If you intend to do any typing over a long period of time, using a tablet just isn't a good idea ergonomically. There are may other reasons to stick to a PC, but this one seems inherent to the tablet format, and once you add a keyboard and a cradle to position the tablet ergonomically, you've lost every advantage that using it might have had in the first place.
So yes, I'd say that there's a large market of people who would rather shell out $X00 for a little bit of convenience rather than $Y00 for something that will be totally unfit for their application. "it's not coming back to the PC world I grew up in" is a truism, but no indicator of the future market shares of PCs and tablets. Also, is there any basis to the claim that you are in the minority?
It's really nice when some build process takes less time because of better hardware. Also, try running some upcoming games on an old PC. Obviously the need for some hardware depends on what you are planning to do.
The PC is dead, it's just not dead for computer professionals, and never will be. But for the rest of the world - think mom, dad, gramps,grammy - why on earth do the need the headaches of a full PC (mac or windows)? A good tablet is basically enough for almost everyone else.
I'll never get this point. Laptop - fine, yes, but a tablet? How do you chat, post FB status, or write an email from the virtual keyboard? It's painful. Especially if you are 50+. I don't see my mom or dad using these devices.
My grandmother is 90 and she does all of those things from her iPad. She's also almost blind and has arthritis, and yet she finds the tablet meets her needs - which mostly involve keeping track of her grandchildren and great-grandchildren via email and FB.
Anecdotally, I've found that "non-technical users" have trouble adjusting to any new UI, including Apple laptops, phones and tablets.
For them the choice is between keeping an old computer, on which they already know how to send emails and compose documents, and buying a new tablet on which they will need to learn a whole new set of UI idioms. In addition, physical keyboards and larger screens (and no "ipad neck") make laptops and desktop computers more attractive.
Add Windows 8 to that, and oh boy, there's a lot of tears. My aunt's laptop died so I bought her a new one. Every time I talk to her she's complaining about the Windows 8 UI. She still hasn't figured out how to do some tasks she would do before with Windows Vista (which she had no problem with).
I think even for the casual user there are a still a lot of use-cases in favor of a PC. Basically anything to do with managing large amount of content like photos, music or even fairly light content creation like writing long emails is much easier to do on a computer with a larger screen and mouse/keyboard.
I think the issue is that the rate of improvement has fallen pretty hard. I remember when nvidia moved from the 5 series to the 6 series, their new flagship card doubled the performance of any current card on the market. The same thing happened with the 8 series. Processors before multicore would show direct improvements in the speed of the machine, especially if (like the average consumer) your machine filled up with useless, constantly running crap over time.
These days I just don't see that. Graphics cards seem to improve by 30-50% each generation, and because so many games are tied to consoles now, they often aren't even taking advantage of what's available. With multicore processors and the collapse of the GHZ race, there's no easy selling point as far as speed, and much less visible improvement (now all that useless crap can be offloaded to the second core!) and most consumers will never need more than two cores. Crysis felt like the last gasp of the old, engine-focused type of game that made you think "man, I really should upgrade to play this"... and that was released in 07. Without significant and obvious performance improvements, and software to take advantage, why bother upgrading?
People snack on smartphones, dine on tablets, and cook on PCs.
A lot of people don't want to cook, so are happy with smartphones and tablets.
Why buy a desktop or laptop when an iPad will do everything you need for a fraction of the price? That's what people mean when they sound the death knell for the PC.
Why cook when you can eat chips and order pizza? Probably because it's better for you and because cooking has cultural significance that goes beyond simply replenishing calories.
People who cheerfully proclaim that PCs are dead forger that PCs aren't just devices, they also attained a certain level of cultural significance. IF the death of PCs also means the death of PC culture (which involves things like game modding, hobby website making and so on), then the death of PCs is a really, really bad thing.
It’s awesome to do cool, fulfilling things and for some of those things you need a computer. For others you don’t.
I really don’t see why everyone should use a computer, given the wealth of possibilities out there.
Also, I’m pretty sure the death of the PC doesn’t mean any of what you are insinuating. It will be more like the death of horse riding after the advent of the car. (If I want to go horse riding there is a club that offers that not five minutes from where I’m living. Horse riding is dead – but that doesn’t mean it’s impossible or even hard to go horse riding today.) Only that PCs will probably be an order of magnitude more relevant than horses are today and, while not always as relevant as in the past in certain contexts (at home), they will still be relevant in others (some work, academia, …).
I’m still pretty confident in the prediction that the PC at home will die. (Which will not mean that no one will have one at home. Just far less people than today.) I’m also pretty sure that the PC at work and in academia will not die.
Division of labor. Not everyone has to be a producer in every sphere; it's okay to be a producer in some, and a consumer in others.
Plenty of people are too busy with other aspects of their lives — doing things which may, for all we know, be of great cultural significance — to spend time being a producer in the digital sphere as well.
Some people devote their lives to cooking for others; others devote their lives to other pursuits, and only ever consume food produced by others.
That's where the analogy falls apart, people who only eat out are a small percentage of the overall popular whereas people who only need a tablet are purportedly the norm.
The numbers don't have to be the same for the analogy to work.
In any case, I think you're underestimating the number of people who never cook, or cook very infrequently. In your average family household, one person typically cooks the vast majority of the meals.
To clarify, by cooking I mean real cooking — beginning with raw ingredients, going through numerous stages of preparation requiring some degree of skill, etc.
That stuff is never going away. "Death of PCs" doesn't mean the complete disappearance of them, just their death as a dominant consumer item. Professionals and prosumers will never stop needing PCs, and they're the ones who constitute the groups you mentioned.
As for me, I'd much rather have my personal chauffeur carry around my full kitchen and always fresh ingredients so I can eat in luxury any time I wish.
Thank goodness for tablets with full XWindows support to my desktop and the university supercomputer. I like broken metaphors.
Does a tablet really constitute a full kitchen, with no compromise?
To me a tablet is a cramped working space (small screen, limited memory), difficult to use (requiring additional utensils like a keyboard, mouse to make certain tasks bearable), with limited storage space (no cupboards), limited processing power (more like a camp stove than an oven), and only usable in short bursts.
Edit: In any case, the analogy is as much about the time, effort and skill required to cook as it is about having the relevant equipment at your disposal.
A 64gb model of the iPad costs $700 (because 48gb of storage should cost $200 to pad those juicy margins).
I bought an amazing desktop from HP last year on a black friday sale for $779. For what's in it, you couldn't have assembled it from Newegg at that price.
In another generation or two the typical Chromebook will be superior to the iPad on performance, while being half the price.
You should buy a desktop or laptop because you get drastically more computing power at the same price.
Five years ago, I bought a MacBook Pro to replace my PowerBook G4, which was itself five years old. The list of obsolescences was enormous: It had only USB 1.1 in a market teeming with new USB 2.0 hardware that couldn't have existed with the slower speeds; it had a single-touch trackpad just as OS X was introducing all sorts of useful multi-touch gestures; it relied on clumsy external solutions for wi-fi and Bluetooth; it had a slow-to-warm CFL LCD that had been supplanted by bright new LED backlit screens; it was even built on a dead-end CPU architecture that Apple had traded for vastly more powerful, energy-efficient, multi-core x86 processors.
Today, the calendar says it's time for me to upgrade again. Yet the pain of obsolescence of a five-year-old laptop in 2013 just isn't the same as in 2008: USB 3.0? What new applications is it enabling? Anything I need Thunderbolt for? Not yet. New Intel architectures and SSDs at least promise less waiting in everyday use... but I'm hardly unproductive with my old machine.
I would just upgrade the RAM and put an SSD in there and you should be good for a while. Drooping an SSD in mine was the single biggest upgrade I've done for productivity and battery life.
I have an old Dell D830 with a 1.8 Ghz Core2. I put an SSD in there and it made it feel like a whole new machine. My parents were complaining out their slow computers. I had them buy SSDs.
SSDs help in most cases, however, I once happened onto a netbook (Atom 1.6GHz, 2GB RAM) where SSDs did not really help. Slow as mollasses in Web browsing, Flash video stuttering at 480p. I removed SSD and sold this netbook immediately.
The PC market isn't dead, but then again, the Mainframe market isn't dead either.
The Post-PC devices[1] (tablets / smartphones) are it for the majority of folks from here on out. They are easier to own since the upgrade path is heading to buy new device and type in my password to have all my stuff load on it. If I want to watch something on the big screen, I just put a device on my TV. Need to type, add a keyboard.
The scary part of all this is that some of the culture of the post-PC devices are infecting the PCs. We see the restrictions on Windows 8.x with the RT framework (both x86/ARM), all ARM machine requirements, and secure boot. We see the OS X 10.8+ with gatekeeper, sandboxing, and app store requirements with iCloud.
The PC culture was defined by hobbyists before the consumers came. The post-PC world is defined by security over flexibility. Honestly, 99% of the folks are happier this way. They want their stuff to work and not be a worry, and if getting rid of the hobbyist does that then fine. PC security is still a joke and viruses are still a daily part of life even if switching the OS would mitigate some of the problems.
I truly wish someone was set to keep building something for the hobbyist[2], but I am a bit scared at the prospects.
1) Yes, I'm one of those that mark the post-PC devices as starting with the iPhone in 2007. It brought the parts we see together: tactile UI, communications, PC-like web browsing, and ecosystem (having inherited the iPods).
2) I sometimes wonder what the world would be like if the HP-16c had kept evolving.
> I truly wish someone was set to keep building something for the hobbyist
I really don't understand your concern.
Hobbists have a wider selection of computing tools than ever before (altough, that statement was true at any time since the 50's). We have the entire arduino ecosystem for hardware hobbists, throwaway PCs like the Raspberry Pi for embebbing real computers everywhere, several different standards of desktop-capable parts for more powerfull systems, and the server ecosystem for the real beefy ones.
Most of those computer types aren't even able to run Windows or OSX. iCloud and Secureboot won't make them go away.
> Hobbists have a wider selection of computing tools than ever before
I don't think that's quite true. We had Heath kits and a lot more variety of computers from the late 70's to the early 90's. There is no under $200 computer sold at major retailers like there was in the 80's.
I don't consider inflation a valid excuse when we continually hear computers get cheaper every year and the Mac mini is selling at the Apple II's old price point. The PC industry abandoned the sub $200 because of Windows licensing, Intel, and Apple taking the old Apple II price as a floor.
$570 is a lot father out of reach today for many than $200 in the 80's.
Computers DO get cheaper every year. That Core 2 Duo machine you bought in 2006 might have cost $2000, but those components (if you can even find a place to buy them) would likely cost about $300 today.
It's unfair to say that they don't get cheaper considering that a Mac Mini is an entirely different class of machine from an Apple II, in so many ways that it's ridiculous to even try to list them here.
As for your statement about $570 vs $200... what do you base that on?
No, computers in the same price range are more powerful. The price range for computers under $200 has disappeared. The Mac mini and Apple II inhabit the same price range. You can get more power in the same price range next year but that same power never trickles down to the price ranges that disappeared in the 90's.
It's a big problem for the starting hobbyist. A kid will probably receive an iPad rather than something to start them on their way to being a programmer or EE.
400 comments
[ 4.7 ms ] story [ 131 ms ] threadYes, on paper, the latest processor is faster than the one released two years ago but you have to be doing specific types of workloads with it to really make a big difference.
The motherboards for PCs built 5 years ago are completely different from those built today, and the CPU sockets have changed every other year. New processors from Intel will be soldered on.
The performance of a PC from five years ago is probably adequate for web browsing and office tasks. For anything more demanding, the advances in power consumption, execution efficiency and process node are huge leaps from five years ago.
Anecdotally, my machine is approaching four years old and I can still run just about every new game on the highest settings. At just under $1,000 at the time, this isn't a luxury yacht type rig, either.
In 2000, a three-year-old PC couldn't run modern games. Today, a three-year-old PC simply forces you to switch to lower graphical settings. The race for hogging all the hardware resources did slow down in the past years, which is a good thing for consumers.
The only time I felt like I've needed an upgrade is while playing Planetside 2, which is/was very CPU bound for my setup. However, when it was initially released, Planetside 2 ran like a three-legged dog even on some higher end rigs. It's much better after a few rounds of optimizations by the developers, with more scheduled for the next month or two.
I dual boot Linux boot on the same machine for my day job, 5 days a week all year. For this purpose it has actually been getting faster with time as the environment I run matures and gets optimized.
As good as it is now, I remember struggling to keep up with a two year old machine in 2003.
I trolled around Newegg looking for upper-middle tier components with a higher quantity of good reviews. A lot of the times you won't see a lot of the recently released stuff with useful reviews, so some of the parts were actually circa 2009'ish instead of being latest and greatest (2010).
I didn't splurge for a super expensive case, and my power supply wasn't modular (making it pretty cheap). i7 with a decent mobo. Went AMD for the GPU since (at the time) they were the best bang for buck. Got some cheaper G.SKILL 1600mhz DDR3 RAM (which has worked awesomely for me) for next to nothing, and I was ready to roll.
2010 $1000 rig running all 2013 AAA titles at the highest settings sounds unlikely to me…
EDIT: don't know my full specs offhand; i5, gtx 460, either 4 or 8 gigs RAM (4 I think), a 120gb SSD and 2G HD
I don't doubt the OPs rig is powerful enough to play modern games, but I seriously question the "any modern game at highest settings" statement.
FWIW, I built a rig a great rig in 2010 which I still use from time to time. i7 920, 12GB DDR3, SSD, SLI GTX 460 2GB.
Maybe the OP doesnt play some of the most demanding games?
I play some pretty demanding games, but I was able to get a lot of performance per dollar building custom. I knew where it was OK to spend more/less and did a lot of research/shopping around.
Happily ran BF4 beta on High/Ultra 2x MSAA at 1080P.
I go with generally mid range components with my gaming machine. Even then, I upgrade every 3 generations for CPU and every other generation for video card. CPU performance doesn't impact gaming as much as it used to.
This gives me reasonable performance in most games around high to ultra on a 1920x1080 monitor.
I still use an AMD processor on my gaming/development rig, because upgrading to a better Intel CPU would have cost me nearly twice as much (I already had an AMD motherboard, so that helped). I don't notice a bottleneck in most games.
I think the best strategy is to get mid-range components and just upgrade more often. You get way more bang for your buck, and you can always sell the old components on eBay or whatever.
If you're doing multi-monitor or >1080p resolutions, then you might need to get something better than the 580 however.
AAA games mostly target the console. Look at GTA5, which isn't even out on the PC. Most AAA games will run on a PS3, which came out in 2006, and has 512MB of RAM (combined system / graphics).
That said, there's a point of diminishing returns - making games look much more realistic will take obscene amounts of resources.
Planetside 2 is a weird example, though. I don't think there will be many games that have a 1-year+ delay from landing on PC before they hit consoles.
In fact the only thing I really want a faster machine for is some of the latest emulation techniques (Higan) and a vague desire to play around with some virtualization odds and ends.
Like another poster said, with "game" terms replacing "data": " But, if you can run a [game] and it's [good framerate], it just means you should really be doing better [gaming] over more [pixels]."
A lot of folks reading sensationalist articles about the PC market decline are making the conclusion that nobody likes or uses PC's anymore since sales are declining. The author is pointing out that it's a poor conclusion to make since there are other factors contributing to the decline, including the fact that the usable lifespan of today's PC's is longer than it used to be.
And that's a huge difference for the application developers so it's not only about being smart, that's a very relevant question.
[1] https://www.google.com/search?q=pc+will+disapear&ie=utf-8&oe...
Personally, I upgrade incrementally, and I still use my PC on a regular basis. The machine I have now is a hodge-podge of parts from different ERAs. I have an Intel Q6600 but DDR3 RAM, and a modern, quite beefy graphics card that I bought when it was in the upper-echelons in early 2013. It runs most modern games pretty well. I have an SSD for most software but also three big HDDs, one of which I've had since my first build in 2004.
Back in Q1 2010 I got an Intel Core i7 980X which benchmarked at 8911 according to http://www.cpubenchmark.net/cpu.php?cpu=Intel+Core+i7+X+980+...
Now in Q2 2013 (3 years later) the very top of the line processor available, an Intel Xeon E5-2690 v2, is only twice as fast at 16164: http://www.cpubenchmark.net/cpu.php?cpu=Intel+Xeon+E5-2690+v...
It used to be that things got faster at a much faster rate. And until this new E5-2690 v2 was released, the fastest CPU was only 14000 or so, which is less than 2x as fast.
The C2D to i7 jump was pretty significant as well.
I have water cooled 2600k at 5GHz running 24/7. Bigger GPU and more cores brings additional complexity and more heat. No other CPU can handle this without blowing up.
If you plan on overclocking your CPU it's probably a good idea to get a slightly better CPU cooler than the standard one anyway.
Depending on what you're doing with CPUs there have been improvements, even at the high end between Nehalem and Haswell (on the Intel side). Anandtech's data shows a nearly 2x improvement between those generations for certain tasks.
But the reality is that for other tasks (e.g. the winrar compression benchmark) the speedup is minimal compared to previous years (see Haswell vs Prescott). In many cases where there is a speedup, performance is already so good that you wouldn't notice.
It's pretty crazy to see that in the consumer world, not even games are demanding so much performance from our CPUs.
the PC market isn't dead, it is slowly receding and it won't stop. It's because of the new alternatives, and assuming finite budge, when you get one of the alternatives, which cost roughly around a consumer-level laptop, you don't have enough for another PC that you don't need.
The article to me seems extremely narrow in both its oversight and scope. People don't care about processing power not because it's a marketing gimmick, but because they don't care. People who do care are the ones who know enough to care, and they will always be minority.
There was a time when you felt like a new PC was obsolete the second you took it out of the box. But that was because we were just scratching the surface of what we could do with new hardware. We're now at a point where it's hard to find consumer and business applications for all the spare hardware that you can afford.
Mobile adoption has been so quick because everyone is buying devices for the first time (tablets), or there is an incentivized two-year replacement cycle (phones). But I'm still using an original iPad that works just fine, and a 3 year old cell phone with no reason to upgrade. Eventually, I think we'll start to see the same leveling off in mobile as well.
1) I don't need to buy a new PC every two years anymore 2) Someone should make a tablet with slots so it can be upgraded like a PC
Wouldn't that be a Windows 8 convertible laptop with a touchscreen?[1][2] Once you put expansion options on it you start compromising mobility.
[1] http://www.futureshop.ca/en-ca/product/dell-dell-xps-12-5-ul...
[2] http://www.futureshop.ca/en-ca/product/hewlett-packard-hp-sp...
Look at some ipad teardown for instance and watch how everything is packed together. Not to mention that these days everything is in SoCs: you can't upgrade the GPU and CPU separately. Even the RAM can be stacked on the SoC package and if it's not it's soldered right next to it.
Contrast that with the innards of a desktop PC which is mostly empty space.
That's a natural step because connectors are expensive, power hungry and bulk. As computers get inexpensive, upgrading makes less sense, so it makes less sense to pay more for an upgreadable system.
If any, what is dead is the software need for the Moore's law
The limiting factor is if your computer's feedback loop is tighter than your brain's perception loop. If you can type a letter and the letter appears, your computer is fast enough for word processing. But, if you can run a data analysis job and it's done before you release the "enter" key, it just means you should really be doing better analyses over more data. Certain use cases grow like goldfish to the limits of their environment.
That's the problem in this type of "PCs are fine, people just don't have to upgrade" argument. (And I've made it myself, before I really worked it through and took stock of just how far tablets have already come.)
The use cases satisfied by an old PC could all be satisfied just as well by today's tablets. Yes, the software support for keyboards, external displays and server-type services isn't quite there. But that's solely a software limitation. Not a limitation of the hardware platform. It will be addressed and more PC sales will disappear.
And the actual remaining PC-justifying use cases are precisely those where an old core2duo isn't "good enough" and simply buying some more RAM or even an SSD isn't going to obviate the need for new hardware for 5+ years.
So inasmuch as people still need PCs, they have to show up in the sales charts. And inasmuch as we don't, mobile devices are going to eat that market as the software evolves.
There's really no way around it.
Frankly, I think android and/or iOS are one good upgrade-cycle away from decimating laptop sales. If either or both really focus on the 'docked' use case -- better support for external displays and keyboard support to facilitate top-flight bulk text entry/editing -- laptops are dead to the general public.
If a student can so much as use a keyboard case and tablet to write a term paper or code up some homework as easily as on a laptop, it's over. And there's no hardware preventing that. It's software. If Microsoft wasn't so inept, they'd have been way out ahead on this.
And, frankly, between the potential of the docked use case and "BYOD" policies, mobile devices could seriously eat into the enterprise desktop market as well.
You build a device cradle, that connects to an external display, power and a keyboard/trackpad, and enables the OS and apps to automatically sense that connection and shift their interfaces accordingly and you'll see how few people truly need a PC anymore.
I think docks have some hurdles to jump in the consumer market, though. They're unsexy - I think most people associate them with boring enterprise computers and boring enterprise work, and they tend to be big and ugly. They're expensive too, considering that most people view them as nothing but plastic with a few connectors and wires inside.
If someone can figure out how to jump those hurdles, though, and make docks sexy to the consumer market, beige-box sales will plummet. Make them smaller, easier, cheaper. Make them less proprietary - at the very least, a Company XYZ consumer dock should be compatible with all/most of of Company XYZ's portable offerings. Make them wireless and automatic (OK, I realize this conflicts with "cheaper")! Let me drop my Win8 tablet on my desk and immediately have it display to my dual monitors and let me use it with my mouse, keyboard, printer, big external drive and wired network connection. Let me press a button and have it display to my TV or play to my big speakers. Have your cake and eat it too.
I mention Win8 specifically because the whole idea of it is that it's a tablet and a desktop OS in one. Why on earth is there no Microsoft-produced drop-in docking solution for the Surface so it can actually be both?! The consumer potential is crazy - got a crusty old desktop PC laying around? Toss it, keep the peripherals, buy a Surface+dock and you've got a new tablet/laptop and desktop. You can "dock" a Surface as-is with two cables (a micro HDMI and a USB hub with all your desktop peripherals wired in), but that leaves out wired network (maybe important), certain peripherals like speakers (maybe important), and power (critical), and even two cables is two too many.
Most people who say they need a PC don't need a beige box, they need a keyboard, mouse and monitors on a desk where they can work. The form factor of the box that the CPU, disk and memory come in doesn't matter when you're at the desk, so it might as well take the form of something you can take with you and use when you're not at the desk: a laptop, tablet or phone.
My tablet can connect to my wifi when I come home and sit at my desk. There's no reason at all why it can't connect to my mouse+keyboard at the same moment; so someone needs to solve the technical problems of it being able to connect to my large monitor wirelessly, and we're set.
just plain ole mathematical modeling / machine learning, and associated duct taping the tubes.
I am also going to be releasing the start of a "Numerical Haskell" tool chain sometime (hopefully this fall, but life happens and all that jazz)
The naive obvious "dot product" matrix mult of two Row Major matrices is 100-1000x slower than somewhat fancier layouts, or even simply transposing the right hand matrix can make a significant difference, let alone more fancy things.
Often the biggest throughput bottleneck for CPU bound algorithms in a numerical setting is the quality of the memory locality (because the CPU can chew through data faster than you can feed it). Its actually really really hard to get C / C++ to help you write code with suitably fancy layouts that are easy to use.
Amusingly, I also think most auto vectorization approaches to SIMD actually miss the best way to use SIMD registers! I've actually some cute small matrix kernels where by using the AVX SIMD registers as a "L0" cache, I get a 1.5x perf boost!
Still I don't see the connection to Haskell, can you elaborate ?
well: 1) i've been slowly working on a numerical computing / data analysis substrate for haskell for over a year now.
2) the haskell c ffi is probably the nicest c ffi you'll ever see. Also pretty fast, basically just a known jump/funcall! And no marshaling overhead too!
3) theres a lot of current and pending over the next year work to make it easy to write HPC grade code using haskell. Some approaches involve interesting libs for runtime code gen (the llvm-general lib for haskell is AMAZING).
Theres also a number of ways where ghc haskell will likely get some great performance improvements for numerical code over the next 1-2 years! (i've a few ideas for improving the native code gen / SIMD support over the next year that I hope to experiment with as time permits)
I am also going to be releasing the start of a "Numerical Haskell" tool chain sometime (hopefully this fall, but life happens and all that jazz)
for example: random forest, gradient boosting, gam, etc -- you will typically do parameter searches and the models you get are as good as your willingness to wait. Good software will run at a significant fraction of memory bus speed and the faster that bus goes the better your models will be.
eg: most CPU memory traffic is in "cachelines" size chunks so you're best off trying to organize information so you can use all bandwidth! I've a few ideas on this i'm trying to bake into an array/matrix library i hope to release soon. :)
Regular users don't know or care about memory management, they don't even close old windows or tabs, its about convenience. That's not a problem in mobile where the need is the mother of invention so mem management is automatic and chrome reopens the tabs you had by itself, but in a desktop environment (specially windows) one wrong click and the session restore in chrome wipes your previous session.
But it was cheap, cheaper than an unlocked iphone and it gets the job done so its ok for her.
I suspect a few people on HN will be reluctant to see their role in this arrangement. ducks
We built really powerful computers and decided the best way to use them was to run web apps consisting of a poor performance scripting language with a poor performance visual-rendering language and half the people who build with it seem to think that anything done in a web browser is free.
"Modern" CSS for even a simple site is pretty silly. So many horribly inefficient rules, sometimes dozens of levels of overwriting properties, etc. And gobs upon gobs of Javascript that does very little but is constantly running, checking input boxes, animating every tiny little detail, doing things that can be done without Javascript, etc.
I have better system performance when running a Windows 7 VM, a semi-bulky image editor, or compiling the kernel than I do with a few bulky webapps in Firefox. And this is on a desktop built just 4 years ago.
I have a light-weight Linux setup that uses 80MB of RAM after logging in. It has 1GB of RAM. I can't run GMail and Facebook Firefox and browse the Internet at the same time with minimal open tabs. It's sad.
I know that some newer frameworks take advantage of GPU features (actually I used to work as a dev in the Windows org) but guess what, GDI is still faster.
As a data point: Safari is using 600+MB now with just HN and GitHub open. It's using more memory than all the other apps (editor, terminal, mail, Rdio, Skype) combined. 600MB is not much by today's standards, but comparatively, is ridiculously wasteful. It's a damn CD.
Hope the Mavericks update improves this a bit since I'm short on RAM on this machine (4GB).
My MBA has 4GB and I constantly forget, because it is enough, and I am constantly surprised that it doesn't have 8GB.
All modern web browsers save and restore sessions across restarts.
I always up-to-date Firefox, so hopefully the memory leaks are minimal. I reboot my Linux machine every month-ish and don't exit Firefox unless necessary. I always keep a few "app" tabs pinned, only a couple of them are heavy. I have 4GB of RAM and turned swap off for an experiment. Sure enough, after a couple weeks of browsing and keeping about 25 tabs open, Firefox's memory consumption would creep up and up and eventually it would crash.
I can only guess/hope that there are memory leaks involved.
Why should they? If a tab's "dormant", the browser can just quiesce JS timeouts and let the tab's memory get swapped out. Not a problem anymore.
That most (all?) web browsers don't do this currently isn't the user's fault.
Faster CPUs, more memory, and faster storage are always welcome. I look forward to the day when Eclipse and other IDEs really start taking advantage of GPU stream processors for indexing and validation.
The PC market has relied on end users - consumers and business users - for it's growth engine for decades, and that appears to be drying up. One of the reasons for that is outlined in the article, for most use cases we don't need faster.
I fit into a lot of the special cases here: Developer, gamer, amateur photographer with many gigabytes of RAW files and even I don't feel the need to upgrade systems like I used to. Now it is about an every 3-4 year thing whereas previously it was yearly or more.
--- Why does nobody talk about them? Because nobody wants them, that’s why. Imagine somebody brings you a personal desktop computer here at South By, they’re like bringing it in on a trolley.
“Look, this device is personal. It computes and it’s totally personal, just for you, and you alone. It doesn’t talk to the internet. No sociality. You can’t share any of the content with anybody. Because it’s just for you, it’s private. It’s yours. You can compute with it. Nobody will know! You can process text, and draw stuff, and do your accounts. It’s got a spreadsheet. No modem, no broadband, no Cloud, no Facebook, Google, Amazon, no wireless. This is a dream machine. Because it’s personal and it computes. And it sits on the desk. You personally compute with it. You can even write your own software for it. It faithfully executes all your commands.”
So — if somebody tried to give you this device, this one I just made the pitch for, a genuinely Personal Computer, it’s just for you — Would you take it?
Even for free?
Would you even bend over and pick it up?
Isn’t it basically the cliff house in Walnut Canyon? Isn’t it the stone box?
“Look, I have my own little stone box here in this canyon! I can grow my own beans and corn. I harvest some prickly pear. I’m super advanced here.”
I really think I’m going to outlive the personal computer. And why not? I outlived the fax machine. I did. I was alive when people thought it was amazing to have a fax machine. Now I’m alive, and people think it’s amazing to still have a fax machine.
Why not the personal computer? Why shouldn’t it vanish like the cliff people vanished? Why shouldn’t it vanish like Steve Jobs vanished?
It’s not that we return to the status quo ante: don’t get me wrong. It’s not that once we had a nomad life, then we live in high-tech stone dwellings, and we return to chase the bison like we did before.
No: we return into a different kind of nomad life. A kind of Alan Kay world, where computation has vanished into the walls and ceiling, as he said many, many years ago.
Then we look back in nostalgia at the Personal Computer world. It’s not that we were forced out of our stone boxes in the canyon. We weren’t driven away by force. We just mysteriously left. It was like the waning of the moon.
They were too limiting, somehow. They computed, but they just didn’t do enough for us. They seemed like a fantastic way forward, but somehow they were actually getting in the way of our experience.
All these machines that tore us away from lived experience, and made us stare into the square screens or hunch over the keyboards, covered with their arcane, petroglyph symbols. Control Dingbat That, backslash R M this. We never really understood that. Not really. ---
[1]: http://www.wired.com/beyond_the_beyond/2013/04/text-of-sxsw2...
Because while basic computation is a universal commodity, what is implemented on top of it certainly isn't - a piece of software always functions as someone's agent. When the systems you end up relying on are entirely defined by someone else, the only thing that represents your will is your mind, and it is effectively executing a complex and ill-defined protocol against always-diligent computers.
You've done the computational equivalent of declining a lawyer.
But it doesn't seem like a big deal, since you're only compromising a little at any given time. But the software is always changing in ways that benefit its controllers while your expectations are mostly based on the capabilities that they've presented. So the progress you perceive is entirely in their desired paradigm. Features that would benefit you but at the expense of Google/Apple/etc are never explored, because you aren't the user of their software - you're its working set!
I can forgive the old-timers who were conditioned by broadcast media to see the world in hierarchal take-it-or-leave terms and don't understand what they're losing by sharecropping in walled gardens. And I can mostly forgive the unclued herd that just buys whatever is advertised.
But for everybody who knows the power of personal computers yet pretends webapps and locked appliances are actual progress, either out of personal laziness, cognitive dissonance, or longstanding need for social acceptance: shame on you for abandoning that self-determination you tasted the first time you truly experienced computing.
It's not just the data itself, but really about protocols used to access the data. Protocols mediate between parties, and by choosing to download a binary blob simply to check your email, you've given up any true bargaining power in that exchange. You still have some autonomy by hacking the blob (userscript injection, etc), but you're only building on unstable ground.
It's that when tablets hit the scene, people realized they don't need their PC for 90% of what they do on a "computer". Email, social networking, shopping, music, video etc.
Us old geeks who swap hardware, play PC games, tweak OS settings and generally use yesterday's general purpose PC will be the ones remaining who keep buying new hardware and complete machines.
The general public meanwhile will only buy a PC if their tablet/smartphone/phablet needs expand beyond those platforms.
The market will shrink but it will turn more "pro". The quicker MS evolves into a modern IBM the better.
That's exactly what I see everywhere. People still need a PC for the 10% of the tasks they can't do on a tablet. So, they'll keep a good enough PC around all the times.
What is a different situation from a few years ago, when people were buying a PC for each family member. Now, it's a tablet for each person, a PC for each home, and spare space at the desks.
The rest is a matter of taste, I personally prefer PCs (well, Macs) but I can definitely understand people preferring tablets, if only because they are so much cheaper, and generally so much easier to use (to my astonishment my daughter figured out how to use my iPad to watch videos on YouTube while she was still 0 years old, before she could even speak...).
When it comes to work I need a computer. And event that might change in the future [2].
[1] http://www.apple.com/creativity-apps/ios/
[2] http://thebinaryapp.com/
Easy copy pasting, selecting specific parts of a document... basically anything requiring precision and speed. You can add a keyboard and a mouse to it I suppose, but at that point you may as well have got a laptop.
My grandfather (89 years old today) switched to the iPad. He was one of the pioneer of computer usage in France (in the steal industry)... at a time when programs were punched on cards! He retired before the mouse/keyboard thingy became popular, and never quite managed to fully grasp it when he got a computer ten years ago. He is not switching back, the iPad is way easier for him.
My parents went full iPad without knowing it: after getting one a year ago they realized that they just aren't using their laptops anymore. The iPad is just much nicer to browse the internet. In short the iPad is better for them 80% of the time, and worse 20% of the time. It can only get better with time, as we get better at making touch interfaces.
Don't get me wrong, I'm not advocating anything, I just think that's where the market is heading, and it's not coming back to the PC world I (and probably you) grew up in. I for one will keep using my laptop... but I know I'm in the minority.
So yes, I'd say that there's a large market of people who would rather shell out $X00 for a little bit of convenience rather than $Y00 for something that will be totally unfit for their application. "it's not coming back to the PC world I grew up in" is a truism, but no indicator of the future market shares of PCs and tablets. Also, is there any basis to the claim that you are in the minority?
It's really nice when some build process takes less time because of better hardware. Also, try running some upcoming games on an old PC. Obviously the need for some hardware depends on what you are planning to do.
For them the choice is between keeping an old computer, on which they already know how to send emails and compose documents, and buying a new tablet on which they will need to learn a whole new set of UI idioms. In addition, physical keyboards and larger screens (and no "ipad neck") make laptops and desktop computers more attractive.
Dell, Lenovo Toshiba and HP all report sales of laptops are increasing; 3.5% in the last quarter.
Death doesn't sound too bad.
http://money.cnn.com/2013/10/10/technology/mac-pc-sales/
These days I just don't see that. Graphics cards seem to improve by 30-50% each generation, and because so many games are tied to consoles now, they often aren't even taking advantage of what's available. With multicore processors and the collapse of the GHZ race, there's no easy selling point as far as speed, and much less visible improvement (now all that useless crap can be offloaded to the second core!) and most consumers will never need more than two cores. Crysis felt like the last gasp of the old, engine-focused type of game that made you think "man, I really should upgrade to play this"... and that was released in 07. Without significant and obvious performance improvements, and software to take advantage, why bother upgrading?
A lot of people don't want to cook, so are happy with smartphones and tablets.
Why buy a desktop or laptop when an iPad will do everything you need for a fraction of the price? That's what people mean when they sound the death knell for the PC.
People who cheerfully proclaim that PCs are dead forger that PCs aren't just devices, they also attained a certain level of cultural significance. IF the death of PCs also means the death of PC culture (which involves things like game modding, hobby website making and so on), then the death of PCs is a really, really bad thing.
I really don’t see why everyone should use a computer, given the wealth of possibilities out there.
Also, I’m pretty sure the death of the PC doesn’t mean any of what you are insinuating. It will be more like the death of horse riding after the advent of the car. (If I want to go horse riding there is a club that offers that not five minutes from where I’m living. Horse riding is dead – but that doesn’t mean it’s impossible or even hard to go horse riding today.) Only that PCs will probably be an order of magnitude more relevant than horses are today and, while not always as relevant as in the past in certain contexts (at home), they will still be relevant in others (some work, academia, …).
I’m still pretty confident in the prediction that the PC at home will die. (Which will not mean that no one will have one at home. Just far less people than today.) I’m also pretty sure that the PC at work and in academia will not die.
Plenty of people are too busy with other aspects of their lives — doing things which may, for all we know, be of great cultural significance — to spend time being a producer in the digital sphere as well.
Some people devote their lives to cooking for others; others devote their lives to other pursuits, and only ever consume food produced by others.
In any case, I think you're underestimating the number of people who never cook, or cook very infrequently. In your average family household, one person typically cooks the vast majority of the meals.
To clarify, by cooking I mean real cooking — beginning with raw ingredients, going through numerous stages of preparation requiring some degree of skill, etc.
Thank goodness for tablets with full XWindows support to my desktop and the university supercomputer. I like broken metaphors.
To me a tablet is a cramped working space (small screen, limited memory), difficult to use (requiring additional utensils like a keyboard, mouse to make certain tasks bearable), with limited storage space (no cupboards), limited processing power (more like a camp stove than an oven), and only usable in short bursts.
Edit: In any case, the analogy is as much about the time, effort and skill required to cook as it is about having the relevant equipment at your disposal.
A 64gb model of the iPad costs $700 (because 48gb of storage should cost $200 to pad those juicy margins).
I bought an amazing desktop from HP last year on a black friday sale for $779. For what's in it, you couldn't have assembled it from Newegg at that price.
In another generation or two the typical Chromebook will be superior to the iPad on performance, while being half the price.
You should buy a desktop or laptop because you get drastically more computing power at the same price.
Today, the calendar says it's time for me to upgrade again. Yet the pain of obsolescence of a five-year-old laptop in 2013 just isn't the same as in 2008: USB 3.0? What new applications is it enabling? Anything I need Thunderbolt for? Not yet. New Intel architectures and SSDs at least promise less waiting in everyday use... but I'm hardly unproductive with my old machine.
The Post-PC devices[1] (tablets / smartphones) are it for the majority of folks from here on out. They are easier to own since the upgrade path is heading to buy new device and type in my password to have all my stuff load on it. If I want to watch something on the big screen, I just put a device on my TV. Need to type, add a keyboard.
The scary part of all this is that some of the culture of the post-PC devices are infecting the PCs. We see the restrictions on Windows 8.x with the RT framework (both x86/ARM), all ARM machine requirements, and secure boot. We see the OS X 10.8+ with gatekeeper, sandboxing, and app store requirements with iCloud.
The PC culture was defined by hobbyists before the consumers came. The post-PC world is defined by security over flexibility. Honestly, 99% of the folks are happier this way. They want their stuff to work and not be a worry, and if getting rid of the hobbyist does that then fine. PC security is still a joke and viruses are still a daily part of life even if switching the OS would mitigate some of the problems.
I truly wish someone was set to keep building something for the hobbyist[2], but I am a bit scared at the prospects.
1) Yes, I'm one of those that mark the post-PC devices as starting with the iPhone in 2007. It brought the parts we see together: tactile UI, communications, PC-like web browsing, and ecosystem (having inherited the iPods).
2) I sometimes wonder what the world would be like if the HP-16c had kept evolving.
I really don't understand your concern.
Hobbists have a wider selection of computing tools than ever before (altough, that statement was true at any time since the 50's). We have the entire arduino ecosystem for hardware hobbists, throwaway PCs like the Raspberry Pi for embebbing real computers everywhere, several different standards of desktop-capable parts for more powerfull systems, and the server ecosystem for the real beefy ones.
Most of those computer types aren't even able to run Windows or OSX. iCloud and Secureboot won't make them go away.
I don't think that's quite true. We had Heath kits and a lot more variety of computers from the late 70's to the early 90's. There is no under $200 computer sold at major retailers like there was in the 80's.
There are LOTS of under-$570 computers sold at major retailers today.
$570 is a lot father out of reach today for many than $200 in the 80's.
It's unfair to say that they don't get cheaper considering that a Mac Mini is an entirely different class of machine from an Apple II, in so many ways that it's ridiculous to even try to list them here.
As for your statement about $570 vs $200... what do you base that on?
* Well, there is a perfectly rational reason for that, and it is not really a problem for hobbists. But that's the fact.