>2009
>still using the computer I built in 2003
>build new PC
>it's even a budget build, yet still insanely fast compared to six years before
>2017
>six years after 2011, and a decent PC from back then is still pretty fast
>if you built a top-tier PC in 2011, it's probably still running like butter today
Is Moore's Law slowing down?
slowing down?
It doesn't apply anymore
>>59891411
Depends what you do with it too, had a q6600 and went to an i5-3570k, did seem quite a bit quicker, but when I did some video transcoding it was a lot faster. I kinda want to get a new PC, but it's hard to justify it (new video card or even new SSD would probably help more it seems). Maybe if I did more video stuff it would be worth it.
>>59891411
>4 cores in 2009
>vs.
>8 cores + 10x cache in 2017
no, it's not
>>59891411
I'm currently pretty much handling 85% of my everyday workload on an 11-year-old Pentium M 760 with 1 GB of DDR2.
There's really only so much you can do to perceptibly speed up algorithms that were pretty much perfected in the 1970s and early 1980s, I don't understand why this is so surprising to people on this board. Swapping out for a chip that's 2x faster on a task that already took a few milliseconds isn't going to make much of a real impact, all that's left to improve on for home use is performance in computationally expensive and constantly evolving applications like video games.
Yes for single threaded performance, no for multithreaded.
>>59891578
for word processing you can literally just use your phone now.
it would be vastly more power efficent
>>59891411
1200$ game computer from 2011 still perfect for normal use in 2017, and light modern games.
There's a huge difference but only if you push your hardware to the max. If you do any gaming, encoding/rendering, you'll see a massive difference. But for office productivity, media consumption, and browsing, there's literally no difference.
>>59891601
Word Processing itself isn't even a hard task. They had fully WYSIWYG word processors for computes with 20Mhz CPUs and 4MB of RAM.
But what you say is right. What really is the new thing is power efficiency because everything is on battery. It's also why Intel is shitting itself over this.
I'm using a 10 year old laptop and after putting an SSD in it it never struggles except with stuff related to the web browser (which these days is the case even with top of the range machines) if I don't block scripts. On one hand I'm glad because it means I'll save a shit load on hardware.
>>59891601
Most home tasks you could probably handle on one. Hell, it's not even a recent thing, I'd say most smartphone/PDA-grade embedded chipsets made after around 2002 are probably good enough for solving most offline problems in the home, maybe even beyond.
>>59891741
Have you ever used RISC-OS on an ARM SoC? It's disgusting how fast and responsive it is. The OS is quite rudimentary since they haven't really changed much since the early 90s, but if it had some productivity software it would literally be a better offline workhorse than a machine that costs even 20x the amount.
I'm honestly starting to think that within a few years offices will start replacing desktop computers with phone and tablet docks.
>>59891411
>Is Moore's Law slowing down?
No. It's dead. It died about 7-8 years ago.
>>59891804
It's partially what came to mind when I was writing that, but I've mostly used older stuff. I remember Acorn systems used to be pretty decent even with shitty ARM610s and StrongARMs, dunno if I'd willingly swap over to one even for offline work though.
Only time will tell if the phone docking thing ever amounts to something, I don't really feel like it will. Most people I personally know are fine with still using desktops, laptops and other traditional systems. They think of their phones and tablets more as companions and toys than real computers.
SSDs removed the one major bottleneck left in modern PCs, now even an i3-2130 from 2011 can run like greased lightning. Core 2 Quads are even hanging in there. Add a cheap discrete video card if you really need it.
>>59891411
Now try watching a 10bit hevc video on it.
ha? gotcha!
>>59894161
Nice me.me
>>59891411
>>if you built a top-tier PC in 2011, it's probably still running like butter today
>have top-tier pc from 2011
>struggles with bloated websites
>>59891601
>2017
>not using a RasPi for office werk
>>59891411
Moore's law only worked for a small period of time. It stopped being accurate a decade ago or something.
I built my computer in 2010 and a Phenom II X2 is too slow for 2017.
>tfw I fell for the core unlocking meme and it didn't work
>>59891411
>>six years after 2011, and a decent PC from back then is still pretty fast
Maybe for /g/, a group of people who do fuck all with their computer other than fix their Linux installs all day long.
Try running even a basic arch visualization render at modern industry expected standards and see how "awesome" your 2011 shitbox is.
Guess you haven't used an OS on an SSD, then.
>>59894552
he's obviously talking about consumer use you sperg
>>59891411
>Is Moore's Law slowing down?
Nah, but Wirth's law is speeding up (because of all managed languages VMs nested in other VMs, it's quite insane how many levels of re-interpretation happens to draw a pixel in a modern web browser)) so it has started canceling out hardware gains.
>C2Q Q6600 @ 3.4ghz
>GA-EP43-UD3L
>HD 7850
>64gb for OS/Photoshit/Lightroom/Vegas
>1tb Blue for music, work files, a little bit of games, random shit, other software
>8gb DDR2
feelsgoodman.jpg
thing runs just fine. however, Vegas is a bitch to work in sometimes. i play older games, so idk how new ones run on an old piece of shit like this. but i am /comfy/ asf senpaitachi
>>59894614
Computers become twice faster every 18 months, software becomes twice slower every 15 months.
(c) Bill Gates, 1799
>>59894638
>not putting a chink xeon inside your shitbox to boost for the last run
>>59891411
PC's have been way more powerful than most people need for a long time now.
>>59894672
Basically true since 2000 in terms of what the software SHOULD require, true since about 2010 in terms of everything else.
>>59894274
Stop using Firefox.
>>59894274
Probably not top tier then, mine is from 2009 and doesn't struggle at all with shit like that.
>>59894274
google had to spend millions to make js slightly less shit on chrome, you can't blame your hardware for shitty interpreted languages written by the lowest tier programmers interpreted by (usually) shitty engines
>>59891411
AMD didn't give Intel any proper competition for over half a decade until Ryzen, that's why progress has seemingly slowed. Moore's law will hold up for another 10-15 years prob
went from a q6600 to a 2500k, it was a decent boost in performance in some areas but the performance difference going from core2duo to intel's i series was severe
2500k's motherboard crapped out (never buy asus) so I upgraded to a 5820k, outside of the insane memory bandwidth that x99 offers and the extra cores + hyperthreading the performance gain was minimal, I'd have regretted going for a 4 core i5 without hyperthreading to say the least
the q6600 rig is still going strong with an updated graphics card and still plays fine on sub 1080p resolutions (which is what most people were playing on in 2007 anyway)
>>59894552
Fucking this
>>59894638
That GPU is ~6% faster than a 750 ti, but CPU isn't that great. You would probably end up CPU bottlenecked. Maybe do what >>59894665 said and get a chink'd xeon to throw in there.
>>59891610
2600k is still more than enough for gaming, on the other hand my 670 is showing its age, probably just that it only has 2gb vram when cards now have more vram than I have system RAM
>>59894274
I bet you bought an apu you nerd, or are running without drivers or something you dumb fuck
I have a phenom x4 945 4 GB ram and gtx460 system
Problems weren't
Who here /2500/?
Haven't even though about OCing it yet. All of the games i play are old. Working just fine with a Sandy Bridge and a r7 360.
>>59897113
2500k's overclock so easily it's a crime not to see how far you can push it. Best done with an aftermarket cooler of course