Based i5 2500k
Images are sometimes not shown due to bandwidth/network limitations. Refreshing the page usually helps.
You are currently reading a thread in /g/ - Technology
You are currently reading a thread in /g/ - Technology
> Launched in 2011
> Still no need to upgrade
> What is happening to Intel?
> i7 6700k get?
>>52147450
Because AMD is behind so Intel is free to release new cpus with minor gains.
No reason to go to a 6700k over a 2500k unless you have a good reason, and gaming isn't a good enough reason.
>tfw i5 2410m since 2011
>all I do is browse 4chan and twitter with either itunes or spotify open
won't have to upgrade until my laptop dies
>>52147651
>and gaming isn't a good enough reason.
What about shitposting on /g/?
>>52147450
5820k is what you should get if you're going to get an i7
>>52147450
Zen needs to be a home run, or this will spell the slow death of traditional x86 CPUs.
>>52147450
you wait patiently and forget about it.
>>52147743
>Zen
>being a home run
>>52147870
I know it's a not a big chance, but it's AMD's Hail Mary for the x86 market. without AMD, intel with stagnate x86 until ARM and light based CPUs pass it by.
>>52147450
2500k here. OC'ed to 4.6Ghz, max temps of about 70°C so I could push it further but I don't see a reason to.
I'll probably update at the end of next year because I want things like m.2 and more 4-pin fan headers.
>>52147905
How long do you think it will take ARM to become as powerful as x86?
I'm sure right now it's good enough to be used for something like a file server.
>>52148238
if you write software to take advantage of many cores at once, ARM might be viable for PC now. the issue is moving to a new architecture, something we haven't done in decades.
>>52148174
are you actually cpu bottlenecked or something ?
only thing I don't like about my 2500k is the p67 it's socketed to for being an obsolete and cheap POS
i've seen very little bottlenecking in games with it and the only thing I wish I had a faster processing would be webm encoding
>>52148238
why would arm become as powerful as x86, nobody needs more powerful arm cores
>>52147733
this
Running an i5 2400 from 2011. I grew out of games during that time, so I see no compelling reason to upgrade.
>>52148301
My mobo is a z68. I did get some bottlenecking when I installed a second 670, the GPU usage wouldn't go above 60-70%. So I bought an NH-D15 and now with an OC the GPUs sit at 99%.
I went from a 2500k to a 4790k nearly a year ago today. It absolutely was a worthy upgrade. Would have gone to a 5820k but it would have cost another $300-400 for the more expensive cpu/mobo/ddr4
>>52147870
>Not believing in Jim Keller
>>52147870
>~60% IPC increase over Piledriver
>Not a home run
Add 60% to piledriver benchmarks and you have every Intel chip beat.
>>52148238
>How long do you think it will take ARM to become as powerful as x86?
Probably a decade. ARM is almost usable for cheap Facebook machines currently. Beyond that, it's still a long ways off. At least unless you use ARM in massively parallel configurations, in which case you should have just bought AMD.
>>52148449
I went for the 4930k about 2 years ago, totally worth the money, happy with my performance, only a couple minor things I wish I would have done.
Only upgrade I can conseve at this point is a xenon, the x99 chipset just doesnt have the ohh factor to make a jump to a new motherboard.
>>52148598
yep, i feel you on that. I have my 4790k at 4.8 on a corsair h110i. With how shitty of gains skylake has shown, my 4790k is going to stay relevant just as long if not longer than my 2500k. That makes me pretty sad, honestly.
>2011
>i5-2500k
>based
>get an i7-6700k
>had to install a dvd drive to install win7
>>52147450
Intel plox... 2600k reporting in, fully passive build.
CAD work and image manipulation mainly.
I will use this bitch till it dies of diarhea in 20 years.
>>52148916
>win7
>not installing in less than 2 minutes with usb3.0 install
>>52148888
Im not a fanboy of either side but at the time Intel had the specs that I needed. I hope AMD gets their next generation together because I'm tired of hearing the next generation of computer components not being the leaps and bounds in performance they used to be.
Like AMDs k6 and intels p2 were massive changers in the market then there was 64bit but the last 7-8 years have been clock speed here, power consumption there, more cache now and then
Ehh. 2500k is starting to show it's age in stuff like Witcher 3 and GTA 5. Both games stutter while going fast through the environments.
Even on Skylake i3's this stutter doesn't happen.
>>52147450
>AMD A10 5800K
>Launched in 2012
>Still kicks ass
>No need to upgrade
>What is happening to AMD?
>>52149127
AMD's CPU division went dark a while ago, apparently they're dumping EVERYTHING they can into Zen.
>>52149153
10 bucks says Intel has a war chest of tech locked away to destroy AMD's last ditch attempt.
Would explain the slow incremental upgrades.
>>52149179
they won't destroy them, not when AMD keeps the Monopoly title away from them. They'd likely just match AMD or slightly beat them.
Is it OK to get a 6700K if I'm upgrading from an E8400?
>>52149203
Fucking AMD holding up the CPU market.
>>52149246
>E8400
Anon, I just put the E8400 in my grandmothers computer.
Go upgrade.
I got my FX 8320 in 2012 and i feel the same way. Maybe, just maybe, we dont really need such powerful processors for shitposting and vidya (and animu)?
If you were, say, using your desktop as a professional workhorse, heavy editing or rendering or whatever, youd have upgraded long ago.
>>52149249
with a bunch of twigs and twine. Bulldozer almost drove AMD to ARM.
>>52148534
I thought it was 40%? Arent we to expect performance similar to haslel?
>>52149264
I literally gave my E8400 build away last year, good on your for passing it along.
>>52147651
How about "I want one and I play games"?
>>52149401
nice boss.
>>52149401
I like it!
>>52149401
what kind of ghetto language is this
>>52149401
Now mail me your old 780Tis.
>>52149414
You don't need to justify yourself to the world, only to whoever is spending the money. If you want it and are spending your money to get it, go right ahead. That has nothing to do with being a worthwhile upgrade or not.
>>52149682
South africa probably
>>52147450
I was going to go for Intel next, but my AMD Phenom™ II 1090t Black Edition is still kickin' it! Way to go AMD!
>>52149720
Damn I remember looking at 955 BE years ago
the point of skylake isn't so much the processor itself it's the features of the chipset that are nice to have. M.2, USB type C, DDR4, etc.
i'm loving my 6700k. i recently upgraded from being stuck on laptops for several years due to my living situation not allowing a desktop. i haven't found any real use for this much CPU power yet, but I'm still okay with it because I expect the 6700k to be sufficient for many years ahead.
>>52149766
>I expect the 6700k to be sufficient for many years ahead.
Depends if Zen is a huge leap or not.
If zen goes miles ahead Intel will probably edge it out ever so slightly to keep the market moving.
Kind of sad but it appears to be the game they're playing.
>>52149766
Z97 has M.2, and DDR4 has no practical improvememts besides being 4W more efficient per stick. H170 motherboards with M.2 and two PCIe 3.0 16x slots are inexpensive though.
![Mfw+i+bought+vietnamese+quot+beats+quot+for+15+and+they+still+_477fee5bb71b44c60c551addca1901df[1].png](https://i.imgur.com/sk0C10U.png "Mfw+i+bought+vietnamese+quot+beats+quot+for+15+and+they+still+_477fee5bb71b44c60c551addca1901df[1].png")
>>52147450
>What is happening to Intel?
>>52149747
Yeah man. But to be truthful, this cpu is probably one of the greatest during it's time. I was considering getting an i5 but decided to get some new thermal paste and it's helped out a bunch. I'm sure within the next year though, I'll have to upgrade and I'll have to go with Intel.
>still on an i3 2100k
i don't give a fuck anymore.
>>52149872
Does your cpu cooler use screws and a backplate? I just got an i5 build and upon seeing the push-pin style cooler I decided to use an aftermarket one.
![tumblr_nbivf4UUON1r0stxoo1_500[1].gif](https://i.imgur.com/vjUligk.gif "tumblr_nbivf4UUON1r0stxoo1_500[1].gif")
>>52149914
>still on an i3 2100k
AMD got lazy, so intel didn't need to compete
![Intel+vs+amd_101027_5330026[1].jpg](https://i.imgur.com/4WSDHFG.jpg "Intel+vs+amd_101027_5330026[1].jpg")
Still holding true.
>>52149921
Yeah, it's got some sort of plastic backplate screwed into the motherboard. I'm still considering getting the Noctua fan cooler or the AIO liquid cooling. This was just a cheaper fix for the moment.
How come?
>>52149914
>2100k
>k
>2100k
>>52148536
>ARM is almost usable for cheap Facebook machines currently.
Uh I think it was usable for cheap facebook machines a while ago dude
>>52150050
I read that the pressure created by the pish-pin coolers aren't that great for performance and using just screws and backplate with the stock Intel one improved temps by 8-10C. I wanted something cheap, cooler and quieter than stock so the 212 evo for $20 was just right.
>>52149914
>2100 K
>>52149380
Similar or slightly worse than Haslel at single-threaded applications. But Zen could beat out an equivalent Intel based on the sheer number of cores alone. If I'm not mistaken, Zen is being focused on the gamer market because it's the only real growing money-maker in the industry. With DX12 being able to fully utilize at least four threads for games, I think Zen might do a cakewalk around similarly priced Intels... Until Intel drops the price on four-cores-or-less processors because they can.
I only just recently dumped my 1100t that's I've been using since 2010 to go towards a 4770 and Z97 board, it was cheap, and the 1100t was somehow worth $200, so it wasn't a huge cash drop. It feels like a brand-new untapped world of computing in comparison
>>52148272
Iirc mt6752 was on par with a g3260 in multi thread performance. It should be plenty for the average user.
I just wish there were more good motherboards. p67/ z68 boards are stupidly expensive, especially if you want Matx
>>52150030
AMD didn't 'get lazy', Intel very specifically took measures to cripple them so that AMD wouldn't be able to out perform them again. Its much easier to maintain their 'legitimate market domination' narrative when the competition has no money to invest in R&D.
>>52151072
Educate me? Or a reasonably unbiased link please.
A 6700k for gaming is overkill. I have been converting quite a few of my rips to h265 lately on my gaming rig (4770k, pic related) and I was bored so I fired up fallout 4, completely forgetting about the encoding. It worked fine. In fact, I couldn't even tell that I was encoding other than some increased load times (good lord I started a new game and it literally took five minutes on the load screen).
>mfw just bought an i7 6700k
jelly?
>also bought a 970
>and ddr4 :^)
i'm not even a "gamer"
>>52148391
That's weird, my 670s were fine in sli on my i7 860. It was running at 3.9ghz though.
>>52151107
>6700K
I hope you're either not intending on OC'ing ever (using stock heatsink) or are doing some sort of water cooling, that PCB bend issue makes me nervous.
>970
Why? especially if you aren't gaming.
If you wanted a nice CUDA card, you should have just got a much faster GTX770/780.
>DDR4
This is fine, you'd be silly buying DDR3 brand new now.
>>52151107
I got a 4690 non K lel
>>52151258
>stock heatsink
They don't come with a cooler.
>>52151086
https://en.wikipedia.org/wiki/Advanced_Micro_Devices,_Inc._v._Intel_Corp.
Intel used unfair tactics to artificially limit sales of AMD parts when AMD had a performance advantage to reduce returns to AMD that they could have invested in further R&D.
That's not the first or the last time that it happened either. Every time it happens Intel receives a slap on the wrist and AMD is left with a paltry amount of money that doesn't make up for years of lost time and money to spend building better chips.
They're also hardly the only company hurt by Intel's anti-competitive practices. Remember when third parties like Nvidia could produce their own motherboard chipsets for Intel hardware enabling cheaper overclocking and better integrated graphics?
Intel badly needs to have serious antitrust measures taken against them, and be broken into three companies. An x86 licensing company required to grant licenses to any legal and capable company under FRAND terms. A fab company required to produce chips for any company in a non-discriminatory manner. Finally a chip design company that would be simply one of a new number of x86 chip producers with no process advantages or licensing cheats to use to artificially gain an advantage.
>>52151292
>Intel knew the PCB bend was so bad that they opted to ship no cooler, so they could always blame the customer for attaching a cooler that didn't comply with Intel's load requirements.
|
|>
|
|
|
>>52149401
You done well, but where I live, E8400's are worth literally nothing.
Fuck, first gen i5's and i7's are dirt cheap. Second gen i5's and i7's are still kind of pricey, though. I have no idea why.
I'm still happy with my FX-8350 desu
>>52151258
i was a good boy this year and got a nice bonus, wanted a super leet computer, an upgrade from my 6 year old i5 750/275 build.
also
>muh visual studio compile time
>bending
the case i got has the mobo horizontal, i think im fine
>still using my AMD Athlon x2 5200
meh
>>52147450
6700k masterrace reporting in
>>52151472
>It's still running.
I'm surprised anon.
>still no need to upgrade
>implying that's bad
3770k here, going to use it forever. Just gonna upgrade the GPU next year when pascal hits the shelves.
>>52151504
Me too. I bought 7 years ago when I was 18 and threw together a shitty $400 PC
>>52151453
>>bending
He means the PCB of the CPU itself. Intel jewed out (again) with Skylake and made it thinner, but it probably won't be a problem unless you clamp down the cooler like a gorilla.
>>52151346
No only the K model doesn't have cooler.
>>52147450
I have a 2500k and the only reason I'm interested in upgrading is because sandy bridge doesn't have VT-d and I'd like to run linux with pci passthrough for windows guests.
I have a bad batch 3570k.
About 4ghz is the max stable overclock I can get with a Hyper 212 EVO
I may get a 6600k/6700k just for this reason, I can get them at MSRP because I live by a Microcenter.
>>52151713
Sandy had VT-d.
Even some PentiumD models had VT-d.
We told you not to buy a shitty K model, but you didn't listen.
Hell, back then there wasn't even a reason to buy a K - because the max multiplier on a non-k Sandy is 4bins above max turbo - so a 2500non-K could do 4.1ghz...
You could have had VT-d AND nice clocks.
But instead you chose memes.
You only have yourself to blame.
>>52148238
>How long do you think it will take ARM to become as powerful as x86?
They're already there.
The Snapdragon 820 could trade blows with an i5 750 I believe.
in 2020 Intel will be shitting themselves due to real competition.
>>52148536
>ARM is almost usable for cheap Facebook machines currently.
The Tegra X1 emulates Wii and fucking gamecube on Android.
We're already at casual x64 replacement tier.
>>52151788
Not that guy but 4.1GHz isn't that great compared to the 4.6GHz I have been running my 2500k at since I got it.
>>52151804
Performance doesn't matter until arm gets x86_64 real time emulation support with no noticeable performance loss
>>52151850
Adobe and shit are already working on porting apps to arm/iOS.
By 2020 workstation software will have native ports.
>>52151850
Can a RISC CPU even emulate CISC CPUs?
>>52151886
They do it all the time.
Right now.
Your CPU is doing it.
Right now.
x86 is CISC, but the only thing on your CPU that understand x86 is the decoder.
From there it's a custom RISC architecture, specific to that microarch.
>>52151886
absolutely. There's a famous name for that principle but I don't recall it off hand.
Can it emulate it with comparable performance? Now that's a different question.
>>52151973
>>52151985
I see. I've always thought that CISC can't be emulated by RISCs, not without huge performance loss anyway.
>>52148941
win7 has no drivers for usb install, retard
>>52152026
You can still, actually. It will fall back to USB 2.0 speeds, that's all.
>>52149827
z97 has m.2x4pcie? Really?
>>52147450
>2500k
>not i7 920
>not being Bloomfield master race.
Seriously, the only thing this maybe chokes on is emulation, and even then I can work around it.
I got an an i7 4790k a few months ago, was that a stupid decision?
>people still buying Skylake K processors after the recent bios updates to Z170 motherboards
>>52152055
Only ASRock, and even then, only two Z97 motherboards
>>52152022
Here's an example:
Say you have a CISC ISA that doesn't have a dedicated MULT instruction. You can emulate that by adding n times.
A RISC ISA might do something more involved like booth encoding that requires fewer cycles.
Both achieve the same result in the end and wouldn't be noticeable to the programmer at a higher level.
>>52152052
NO, it dosen't have usb drivers at all. It creates the virtual drive "x:\" during install. No usb drivers.
>>52152022
>I've always thought that CISC can't be emulated by RISCs, not without huge performance loss anyway.
That's not entirely untrue.
The decoder is modern x86 CPUs eats time, the dispatcher eats time, if the dispatcher gets something wrong and misses a cache hit, then instructions can take a really long time.
Your thoughts are right, but essentially a very '70s view of the world, back when new CPUs could just be entirely new instruction sets.
Now, CPUs have largely standardised on a few (getting less all the time) instruction sets and then you just decode whatever instruction into whatever smaller instruction.
It costs performance, but allows for compatibility - so we don't have to completely rewrite software that is increasing in complexity all the time, every few years.
Java is a the perfect example of that way of thinking vs. the 70's way.
You only have to write the software once, the JVM has to be ported to a new CPU arch once and it's likely that CPU arch isn't changing - but at the core you've got multiple levels of crazy abstraction which kills performance.
>>52152126
So if it has no USB drivers how can it read a USB drive?
>>52152184
If your motherboard does not have usb 2.0, like all skylake, for starters.
I still have it. Bitch can't go x45 multiplier for some reason with my Noctua U12P.
>>52152086
This, I got mine for $85
>>52147450
i5-2500k? Pleb. I'm running an i7-920 @3.9 GHz from 2008 and it's still going strong.
>>52151514
I'm in the same boat as you but I have a 670 currently.
3770k is still solid for my purposes and cant see the point of upraging it anytime soon.
![9r4uw2[1].png](https://i.imgur.com/RObdaiw.png "9r4uw2[1].png")
>>52152058
>>not i7 920
>>52152304
>I'm running an i7-920
Plebs
How much difference would I see if I upgraded to a haswell i5 from a c2d? I don't game, just web browsing and some occasional matlab, would there really be that much of a difference?
>>52152126
That's fucking false. I know because I've been installing USB stick Windows 7 on USB 3.0 ports.
>>52152597
I literally just tried to install w7 a couple days ago and it didnt work
I had to use an external dvd drive
>>52152597
>>52152126
Protip: you're both right.
Most USB3 controllers will fail for install as they don't properly fallback to XHCI.
However, if your controller does do graceful fallback to XHCI (ie. it can run with generic USB2 drivers) then it will work, just as a USB2 controller.
>>52152828
sorry EHCI. not XHCI.
>>52152597
speccy plz
>>52147702
KEK
>>52152828
I tried to install win 7 on a 2.0 stick in a 2.0 slot and kept getting the "required cd/dvd device driver is missing" message
had to go buy a cd drive just to install
>>52153074
also this was after upgrading to skylake, never had an issue before in the 100+ installs ive done with the drive
>I5-4460
>One year and still no reason to upgrade
>>52153074
>>52153084
Most likely you still had a USB3 controller that doesn't fallback gracefully.
Next time you install windows (if you remember) disable the USB3 controller in your bios entirely then try installing from a USB2 port.
I've done it, it's mostly a pain in the ass for 7 though.
>No where as much as a pain as trying to do a full UEFI install of win7...
Guys should I upgrade my GPU or CPU first?
I have an i5-2500k and an r9 280x currently.
I'm want an r9 fury and a skymeme but I don't know what to get first.
>>52153485
CPU if you want, but its better to wait to see what comes out for the 1151 socket.
Upgrading your GPU makes more sense, just get a GTX 980Ti.
>>52147696
Why dont you have a tablet for that?
>>52153485
Why upgrading so soon? Also CPU first since GPU's won't be around until after Summer
>still on 2600k
>motherboard port breaks in shipment, can't use video card
>this is the 2nd motherboard i've had for it
>insurance will cover the replacement motherboard
Planning on picking up a 6700k, Z170X MB, and some DDR4 ram tomorrow. Should be able to cover nearly all the cost with insurance + selling old parts. Y/n?
>>52153602
>$400 cpu
>>52153637
>last one lasted 4.5 years
>holiday money
also $370 for me, i got a discount
>>52152249
>skylake mobos don't have usb 2.0
Retard alert, wait, actually. I'm not sure if my mobo has usb 2.0.
Well, no, it doesn't. False alarm.
>>52149179
Make that 1000 bucks. I know they do.
>>52147733
sure, if you don't want to wait for the die shrink next year
oh wait, it IS next year
>broadwell e
>6,8,10 superior Intel cores
>not waiting for this 1st half 2016
Suuuure, goys. Buy buy buy obsoilete technology
I got my 2500k Christmas 2011. It's past Christmas 2015 and it's still driving the system I'm typing this post on.
GG Intel, you made a CPU so good you can't obsolete it.
>>52149714
>South Africa
>English
Choose one
They speak African there, it's basically Dutch, but a lot more retarded.
my i7 920 is still running like a boss
still gunna upgrade to broadwell-e tho
>>52147450
>Intel stagnates
>Moore's law is actually a lie
>Shill's blame AMD
>Based Keller will save us
>>52151514
>>52152370
Have you delidded, changed thermalpaste yet?
Liquid pro looks promising option.
>>52153548
I like using a keyboard and mouse, plus I originally bought it to take to uni with me (never did).
Also I'm not a tablet autist, I have my phone for android.
>>52152375
The i5-2500k is already a 3.7 GHz chip under load. The i7-920's stock clocks tops out at 2.8 GHz even with its old version of turbo boost (itll only go to 2.97 if a single core is being used). Going from 2.8 to 3.9 (not a high overclock, these chips can hit 4.2+ with a large cooler) is a massive 40% improvement, whereas going from 3.7 to 4.4 is only a 19% improvement. Those CPUBoss scores are at stock clock speeds, and the difference becomes small after overclocking. We actually lose in single threaded but win in multi threaded.
>>52153759
I'd wait until Skylake hits the enthusiast chipset. Overclockable Xeons are going to be sweet.
>>52152073
What update?
The one that allowed non-unlocked CPUs to be overclocked?
>>52154965
>Intel will never release a fully unlocked 10+core Xeon
I'm fairly certain that there is a market for it in the enterprise and business world. Hopefully K12 will be an overclockable ARM CPU that makes Intel rethink its Xeon lineup.
>>52149414
>>52147651
I have an i7 2600k and still play fairly recent games from time to time (Batman AK being the most current) and have no problems. And I have not even OC'd it.
Xeon 1231v3 still better price/performance than any Skylake.
>>52149349
For the price and so far reliability I don't see how people bitch about it and act like it's a bad line of cpu's. It's not like the performance is bad it's just less then we expected and wanted. Also like amd gpu the cpu's seem to age well with software being more optimised for multiple cores. I have no regrets going from a duel core pentium to fx6100
>mfw A10-5800K
>mfw not needing an upgrade unless I play Arma or shitty ports
FeelsGoodFrog.png
I'll wait for Zen or Cannonlake
>>52148492
>Jim Keller
He quit.....
>>52155062
>The one that allowed non-unlocked CPUs to be overclocked?
yep
>>52156023
>Not knowing how Jim Keller works
Up to >>>/v/ you go
>mfw retired old i5 2500k desktop and hooked up cheap old i3 2310m laptop because it's more than powerful enough
>>52147696
my nigga 2450m reporting in
>>52156062
>not knowing how Jim Keller works
He quit....Before launch..... At least stay for the launch....
Zen is confirmed shit.
>>52156103
Get the fuck out
>>52156167
>this mad AMDrone
Yeah he quit because AMD is a great company with excellent prospects for the future!
>>52148365
>I found that I no longer enjoyed games during that time
That's a shame anon, but a) it has nothing to do with age, the average gamer is in their 30s now and b) don't be a patronising shit.
>>52156103
He quit because he was done with Zen.
>>52156234
He quit because AMD don't have a processor after Zen to work on.
>>52156321
>He was 'done' with zen
You mean he couldn't save a sinking ship
>Because AMD don't have a processor after Zen to work on
So it's the last of the sinking ships.
>>52151309
Also, aside from the intel fortran compiler fiasco that was already mentioned (it made it so non intel CPUs would perform tasks in a slower less optimized way), they have also been caught bribing PC manufacturers NOT to use AMD, basically giving away intel chips for free (through rebates to the PC manufacturers) because they could afford it, to ensure anyone who buys a pre built never hears about amd.
Its funny, on the one hand, clapitalism is competition which in theory increases creativity and pushes the productive envelope so to speak. But its also what is holding back major tech advancements, along with breakthroughs in disease like diabetea, because unfortunately theres just so much more money to be made with things the way they are, as opposed to say, lets all dedicate our efforts to advancing as a society, for the betterment of all.
/rant
>>52147450
>still no need to upgrade
This is false. Skylake is at least 20% faster.
>>52156343
>You mean he couldn't save a sinking ship
No, I mean that there was no further architecture work to do. It takes a long time to get a design into devices dude.
Keller isn't the guy you want to get your design on silicon, keller is the guy want to get your design
>>52152129
So...kinda like an abstraction layer for the CPUs?
>>52153074
Ive installed windows 7 from a usb 2.0 slot, many times...
>>52156234
>Getting called out for shitposting
>Must be an AMDrone!
This /v/-tier cancer is exactly what's wrong with /g/ nowadays. See >>52156321
Jim Keller left companies before the launch of his creations several times, most notably AMD with the Athlon64, mostly because the major design process is done way before a product gets launched.
But of course, you're just some shitposting faggot, so you wouldn't know.
>>52156421
>>52156373
>It's shitposting
>Lead engineer can't stay till launch
Can't wait for Zen, I'll bring the popcorn.
>>52156449
Don't get me wrong, Zen is going to be shit, but keller leaving after he's done working on it is not a sign that it'll be shit.
>>52156493
Sure thing bro,key people just leave companies before product launch because it inspires confidence in the product.
>>52156521
Keller is a freelance CPU architect. He left apple before they released their keller chips.
He was at AMD for 3 years dude.
>>52156541
Yeah and now he's at samsung.
Literally months before Zen's launch.
Couldn't stay till launch.
Maybe he's trying to run away from the stench?
>>52156559
AMD weren't going to pay him to sit around twiddling his thumbs. He left because the contract he had with AMD was over and they were not going to pay him to do nothing.
>>52156609
Leave it to /g/ to question million dollar decisions of a billion dollar company
>>52156521
He joined AMD in 1998 and left 1999, designing the K7(Athlon) and K8(Athlon64). The first came out around his departure, the latter in 2003.
Both were a success.
Keller rejoined AMD in 2012 and spent three years designing Zen, and allgedly even doing the groundwork for Zen+
There no evidence yet whether Zen is gonna be good or not, but assuming it's shit from him leaving early shows that you have absolutely no clue what you're talking about.
>Inb4 I just pretended :^)
>>52156493
I'm 99% certain Zen will close up, at least somewhat, to Intel, but it's unlikely of beating them. Nonetheless, if AMD can pull off another Phenom II, they're already good enough for some competition again.
>>52156609
>He left
He quit. there's a distinction.
It wasn't that he was pushed or fired.
He quit.
>>52156609
Yeah, it's not that Samsung poached him after he sabotaged AMD so they're a cheap acquisition.
5820k X99 masterrace
They worked on better efficiency
Jobs didn't want a Intel CPU on the iPhone because it was too hot and drained too much power. The same reason why netbooks were shit, they had to use a shitty mono core CPU otherwise the battery wouldn't last
>>52149179
>Intel notice AMD brings out ZEN and it's improved performance up to haswell
>Intel pushes the "unlock button" and yells "release the K!"
>All the intel CPUs with K or not are unlocked to full power and prices goes down in stores
>AMD loses
>Intel bathes in their tears and buttfucks them
>AMD is bought by intel
>>52156817
>bought by intel
Try samsung.
>>52156559
See >>52156672
>>52156655
>Jim Keller, a well-renown engineer with a history of working on and shifting between multi-billion tech giants left AMD before the launch of its creation again, he must have fled!
>How dare you make assumptions about a multi-billion dollar company!
Kill yourself you shitposting faggot
>>52156868
>Leaves before launch
>BEFORE LAUNCH
>IT'S NOT EVEN OUT YET OR READY
>Everything is okay don't panic and buy AMD.
You first.
>>52156837
And Jim is in talks to sign up with Samsung as lead Architect
>>52151258
>that PCB bend issue makes me nervous.
that was the fault of some builders who use electronic screwdriver and bolt it down too hard.
>>52156345
disgusting SJW on display here
>>52147702
>>52156893
I know you're just baiting again, but for reference
See >>52156672 again
This time with I'll also list some sources:
http://9to5mac.com/2012/08/01/apple-cpu-guru-jim-keller-who-came-with-pa-semi-deal-departs-back-to-amd-to-lead-group-under-mark-papermaster/
http://www.anandtech.com/show/6129/apple-a4a5-designer-k8-lead-architect-jim-keller-returns-to-amd
http://www.xbitlabs.com/news/cpu/display/20120801190422_Return_of_the_King_AMD_K7_and_K8_Designer_Jim_Keller_Back_at_AMD.html
http://www.techspot.com/news/49611-apple-chip-designer-jim-keller-heads-back-to-amd.html
http://www.computing.co.uk/ctg/news/2344223/amd-to-design-new-micro-architecture-for-2015-launch-under-chip-guru-jim-keller
If you cannot comprehend this, then I feel sorry for you, as you must have either some degree of mental incapability or being a sad brand loyalty faggot who is wasting his time on shitposting in forums.
>>52155332
Same.
I have always viewed Intel's flagship processors as LMP1s. AMD's are F1s.
Look at the very old LMP1, the GT40 MkII. It can lap a track faster than most modern race cars, does about 220 MPH, Accelerates like a bat out of hell. Basically, it can thrash any modern super car, and can outrun any hypercar on a track.
And this thing was built in 1968. It's old as shit, but it's also fucking fast as shit. Just like Intel's old i7s.
The 2600k never ceases to amaze me how it still manages to chug along year after year putting up a fight.
So to the people in this thread saying "It's not a worthwhile investment if all you do is video games", you have no idea. Buying a flagship intel chip is the best futureproofing you can get.
>965BE
>no need to upgrade
Bottleneck is the worst shill ever.
>>52157206
depends on what you do with it.
>>52147450
>got an i7 2420 in 2011
>great computer
>dies in october
>forced to get a new laptop
>get i7 6700
>everything runs the same
JUST
>>52147450
new instructions
less power usage
more efficient
smaller die size
smaller transistor size
more i/o
more pci lanes
more memory lanes
ddr4
yeah other than that, nothing has changed
>>52156274
>counting people who "play" mobile games as gamers
>>52153724
>1st half 2016
Current assumption is that Broadwell-E will be presented at Computex, which means June.
By that time, you will be telling people to wait yet another six months for the Skylake-E tock.
>2011
>i5 2500k is all you need they said
>2015
>can't even have source quality twitch streams running in the back and play games at the same time
>>52164186
that's because twitch is such a piece of unoptimized bullshit
>>52164586
also it bottlenecks my 980ti. It is at 4.3Ghz and I have to increase the resolution to get more frames.
>>52147651
>What is DDR4
>>52148174
>ARM replacing CPU's in non-mobile applications
The only reason ARM CPU's will take over the market is because mobile devices are going to take over the market.
>>52153745
I thought African was like ooga booga ug ug wabooga
>>52165894
Why are low-end Intel CPUs called Pentium anyway? Didn't Pentium used to be the high-tier CPUs?
