So Intel's coffeelake won't run on z270 boards des

>>62341253
>Ryzen 2 when?
Late 2018.

File: 1415270369168.jpg (25KB, 500x581px) Image search: [Google]

25KB, 500x581px

>>62341253
>Don't want to buy another board
>Guess I'll buy another board

>>62341623
In all fairness if OP buys an AMD board now it's guaranteed to last a lot longer than a brand new Coffee Lake board.

>>62341253
Where in the article did he get the
>Requires 300-Series Chipset
from?

>>62341253

There is nothing wrong with buying a new motherboard every year.

>>62341253
They offered a BIOS update so Kaby lake could run on z170 boards. I refuse to REEEEE about this until they refuse to give z270 the same for Coffee lake.

>>62341253
>intel ditching socket/chipset
Woah, what a surprise!

>>62341623
>op wants to upgrade
>has to buy a board either way
>but buying AMD is guaranteed to make future upgrades easier

File: 1481255675599.gif (3KB, 124x128px) Image search: [Google]

3KB, 124x128px

>>62341663

File: (((linus sebastian))).jpg (147KB, 1280x720px) Image search: [Google]

147KB, 1280x720px

>>62341253
GIVE US MORE MONEY FOR VIDEOS GOY.

BUY INTEL GOY

File: 1449814878100.png (362KB, 700x700px) Image search: [Google]

362KB, 700x700px

>Intel
>sockets are named after how many pins are in it
>completely transparent

>amd
>sockets have weird names like AM3 or AM4
>no one really knows what's going on
is it me or does this sound a bit fishy?

>>62345454
AM2 - DDR2, AM3 - DDR3, AM4 - DDR4.
Yes, it's THAT simple.

>>62341253
>>62341596
Is that not Ryzen+
And Ryzen 2 is not until 2019

>>62345483
7LP enters HVM H2 2018. They can murder Intel late 2018 if they really want to.

>>62341253
>200$

I don't understand

can someone explain to me why people spend more than the bare minimum on a mobo?

it's just cardboard with a bunch of wires

>>62345499

overclockers

>>62345497
High volume manufacturing?

>>62345454
Why do I give a fuck about how many pins there are? How is that relevant to me as a customer? I have no idea what CPU goes to what socket without googling.
I'm an intelfag, but I seriously prefer AMD naming scheme.

>>62345521
Ye. PDKs for 7LP are available right the fuck now, and it enters risk production H1 2018.

>>62345534
still just pointing out it's a bit sketchy...
they dont want to tell you how many pins, so you cant help but ask, why not?

>>62345499
see >>62345513
also there are features on more expensive mobos that are missing on the low end stuff. You might not need them, but that doesn't mean nobody does.

>>62345513

are you niggaz for real

masterrace is dead

every AAA game is a port, you can't overclock in enterprise,

but you need a 7GHz i9 to run your eroges.

>>62345556

like what?

I'm not being facetious.

B350 and 1700 now.
X370 later if you feel like overclocking more agressively or think you'll get better RAM frequency and timings that way.
Zen+/Zen2 wheneverthefuck. Turn B350 board and 1700 into home server.

>>62345555
>they dont want to tell you how many pins
It's not like it's confidential, you can look it up on wikipedia if you're interested.
But I still don't get why would you, I don't care if a CPU has 3 or 3000 pins as long it does what it should.

File: 1369066721617.png (33KB, 545x500px) Image search: [Google]

33KB, 545x500px

>>62345567
/g/ - Vidya

File: Screen Shot 2017-09-06 at 4.55.28 AM.png (389KB, 488x584px) Image search: [Google]

389KB, 488x584px

>>62345499
Uh are you retarded?

Some additional features like extra USB ports, extra M.2 slots, built-in WiFi adapter, maybe Thunderbolt 3 if you want it, etc. etc.

I usually pay the extra $80 to get more feature packed mobos because I usually use those extra features. And some mobos nowadays come with 10GbE LAN built in too if you get the "deluxe" versions.

>>62345578
More ports for example. Low end board typically have less DIMM slots, USBs, SATA ports and such.
More expensive boards also have better support for RAID in my experience.

>>62345597
>10GbE LAN built in too

future proofing, eh?

eh?

lel.

>>62345483
there is no zen+ every single ryzen with production number 17xx has already fixed quite a lot of problems plus being on b1

File: 1482929245984.jpg (51KB, 720x757px) Image search: [Google]

51KB, 720x757px

>>62345639
It's useful for high speed local networks. I'd kill for a 10Gbps network in my house, but I'd have to upgrade literally every piece of gear I have.
Not everything is for Internet, faggot.

>>62341253
A MOTHERBOARD CHANGE A DAY KEEPS THE NAZIS AT BAY

|
|>
|
|3
|
|

>>62345535
Assuming everything goes according to schedule. Delays due to unforeseen problems aren't exactly uncommon in this business

>>62341623
well, OP speaks for the future probably.. He obviously cant trust Intel anymore.

>>62345648
I don't need problems fixed, I need clockspeeds >4ghz. Zen2 is too far away.

>>62345863
7LP is already a quarter earlier compared to older roadmaps.
And I doubt there'll be any problems fabbing peanuts-sized Zeppelin2.0's.

File: Pepe_rare.png (128KB, 1000x1000px) Image search: [Google]

128KB, 1000x1000px

>mfw I'm on Z68 for several years now
Why do you fags keep relentlessly buying new platforms? Whats wrong with a Sandy Bridge in 2011+6?

>>62345904
Isn't Zen2 just a die shrink?

>>62345927
No, it's both tick and tock.

>>62345927
They say it will reach 5ghz. Whether that's true or not remains to be seen.

>>62345935
Any fun new features?
Are they expected to show up with new AM4 chipsets?

File: 1281034302132.jpg (28KB, 400x300px) Image search: [Google]

28KB, 400x300px

>>62345939

>>62345555
Lmao you are retarded

>>62345927
>>62345935
performance wise i wouldn't expect it to be behaving much different- they claim a slight IPC gain, but that probably only applies to very specific workloads, i.e. clock for clock it'll be the same as regular Zen.

So just imagine Zen2 as being Zen, but pushing much higher clocks.

File: 1440139758980.jpg (50KB, 500x375px) Image search: [Google]

50KB, 500x375px

>>62345939
>IPC improvements over Zen
>fuckton of cores
>5ghz

>>62345910
also it's being developed by IBM engineers which gives a bit more confidence

>>62345939
5ghz is just the optimal clock
the optimal clock for 14nm lpp is 3ghz yet ryzen can go over 4ghz
ryzen 2 could even go beyond 5ghz, but that will all depend on the architecture

>>62341253
Intel shills are starting to reach apple levels.

>>62345974
Basically sounds like Zen on steroids, would be GREAT. It's a slight disappointment Ryzen is stuck at fairly low clockspeeds or we could've had a taste of it already.

>>62345963
>Zen2 as being Zen, but pushing much higher clocks.
it's all it need to be faster than skylake refresh number 4

File: 1502857521197.jpg (23KB, 640x559px) Image search: [Google]

23KB, 640x559px

>>62341253
>I spent 200$ on my board too

suck shit faggot lmao

another incel goy getting exactly what he deserved

>>62345639
10GbE is pretty affordable now.

How the fuck is that future proofing if you transfer a lot of big files across your network you idiot?

10GbE LAN cards are $150-200 and takes up a PCIe slot...so if it comes built in, why not?

>>62341623
It's thinking long term. The longevity of AMD's sockets and chipsets is a matter of public record.

File: wY7KF2V.jpg (109KB, 389x402px) Image search: [Google]

109KB, 389x402px

when are ram prices going to come down?

>>62346087
It's not the 10gig interfaces that'll kill your wallet, it's the goddamn switches. Especially if you need a lot of your ports for all the networked shit you were able to build up because 1gig is virtually free.

>>62345476
Never thought of it that way.
That's pretty clever

>>62346213
Soon, just like HDD prices. H-haha...

X380 when?

>>62346247
HDD price fixing will be their demise as solid state reaches parity for price/capacity.

File: yaM73UC.jpg (49KB, 445x401px) Image search: [Google]

49KB, 445x401px

>>62346266
When are SSD prices going to come down?

>>62346266
That'll be a while though, there's still a 10x difference in $/GB.

>>62346291
SSD prices are like the only thing that have kept steadily decreasing over the past 5 years or so.

>>62346310
lowest price I can find for 500gb is $150. Hasn't budged downwards since last year.

>>62346291
How would he even blow on that thing through the grill over his mouth?

>>62346087
it still uses the pci lanes lol

>>62345555

Well they tell you what type of RAM you need for it. For anyone that is a beginner on pc building, that's more important than how many pins it has I'd say.

Fucking ddr4 drams have doubled since summer of 2016

File: 1405272892673.png (52KB, 421x248px) Image search: [Google]

52KB, 421x248px

>>62341253
What if they actually offer some new features that require a new chipset?

can someone fucking tell me if a can put a 6700k chip on a z370?

it's unfuckingbelievable that none of the "tech" "news" "sites" have even considered this

>>62345608
>less DIMM slots
FEWER DIMM slots, you retarded mongoloid

>>62347054
they don't, they already announced everything there is to announce
it's called refresh+2 cores and lesser clock even in their own slides
release skylake 3rd time strategy
5th if you count i9 and xeons

>>62345925
this
there is absolutely no reason to upgrade if you're on sandy bridge with an i5/i7

sata 3 might be a good reason, so ivy would have made it even less of a need

the only reason i got a 6700k last year is because i've been sitting on a 45nm L5420 on a ddr2 board

>>62347099
https://en.oxforddictionaries.com/usage/less-or-fewer

Thanks grammar nazi anon, now I know.

>>62347108
>sata 3 might be a good reason
sandy supports sata 3
I moved to 1600x from 2600 because gaming, believe it or not but my 390 runs much better now
if all you do is office work there is not reason to upgrade from it, sure

I don't mind upgrading for the sake of a platform.
Things like M.2, USB 3.0 (to say nothing of USB-C) are missing on my Z68. X370 is supposed to have features like integrated SD controllers and Thunderbolt I heard.

>>62347200
People have issues with upgrading from z270, not 68.

>>62347227
Oh sure, they got screwed if they can't upgrade.
But for me, and I guess people upgrading from old systems, it's not an issue.

>>62345476
There are also AM3 CPUs that work on AM2 motherboards.

>>62341663
retarded

>>62345476
AM2, AM2+, AM3, AM3+(95W and 125W+ boards) and AM4. Fixed.

why is it anything intel ever does comes as being something only a lying vile snake would do?

>>62347399
Yeah but I take it there are no AM2 motherboards with DDR3 and vice versa. No matter where the CPUs fit.

>>62347474
Because the company is vile an corrupt to the core.

>>62345597
>idiotic
I've got B150M-PLUS for a 60$ with M.2 slot
>And some mobos nowadays come with 10GbE LAN
so you're gonna spend extra money for a technology that would be mass prodeuced by the time you replace that motherboard 3 times?
>retarded

>>62341663
Good goy.

>>62347241
So Intel only fucked their most loyal consumers who bought their latest products from this year. How nice of them.

>>62347637
intel fucked no one
your point is retarded

>>62347712
I meant fucked over.

I'm wondering if Skylake and Kaby Lake chips would fit in a Coffee Lake motherboard.

My motherboard shat itself before just past the warranty mark and I don't want to invest money in another Z170 / Z270 board.

Doesn't seem to be any info out on this at all.

>>62341253
What's wrong with the actual Ryzen?

>>62347749
how?

>>62347712
>hey, goy, buy this top-end expensive motherboard so you can upgrade later on, tee-hehe
>what do you mean only half a year has passed? your board is already obsolete, just buy a new one

>>62347791
>buy this top-end expensive motherboard
there's your problem
buy the stuff actually you need
The general user doesn't do that..
They buy the mobo who'll do the job for the next 5 years, which is basically the 60$ one
The question is why do you need top-end expensive shit when you don't use the half of the shit in it?

>>62347781
The only reason Coffe lake was released this early and in this first was because of pressure from ryzen. Intel could have done this anytime in the past but they decided to hold back extra cores/performance because they didn't have to offer anything better. And by offering this better option now invalidates their entire product line that came out a half year ago. How do call it not a fuck over if they try to hold back technology as much as they can for more money and when they are forced to improve they also refuses that their customers can easily move forward as well, instead they jew out for even more money?

>>62347823
i really can't comprehend the mental gymnastics you're going through
are you actually defend a meaningless chipset change that brings nothing new to the table while fucking people that are the backbone of your industry in the ass?

>The general user doesn't do that
are you actually saying people don't buy overclockable processors and Z boards to go with them?
>They buy the mobo who'll do the job for the next 5 years, which is basically the 60$ one
how out of touch with reality are you?
>The question is why do you need top-end expensive shit when you don't use the half of the shit in it?
because Intel's jewish practices REQUIRE you to buy an expensive motherboard if you want to use your CPU properly?

>>62347823
>waaah why do you buy the high end stuff you don't need it no one needs that kind of stuff
Then why does it even exist? And yes, people need it especially if it's in an affordable price range. Which is what AMD offers. What you say is textbook Intel shilling response. Good job man, you are quite transparent, but at least you can do your job.

Is there even any point in buying a CPU if you have at least a Haswell? Any improvements in performance are tiny.

>>62345925
>>62347108
I did try, but my Z77 MPower died right after warranty and took the i7 with it. Ain't my problem, and never buying MSI again at least for a few years. This might not have happened if I got the Sabertooth Z77 instead(those were the only 2 options nearing the end of Ivy Bridge in my region)

>>62347852
>how out of touch with reality are you?
I'm right in the industry.. Everyday I see example backing up that
>to use your CPU properly?
cpu load is 60% running a full-working OS and a 1-2 virtual machines at the same time
B150M-PLUS + 6100
>to use your CPU properly

>>62347872
That entirely depends on what do you do on your pc. Telling from your satisfaction with an outdated Intel cpu, you most likely play video games exclusively.

>>62347857
>And yes, people need it especially if
need it for what?

File: my sides.png (173KB, 640x400px) Image search: [Google]

173KB, 640x400px

>>62347897
>officelady tier i3's should be enough for anyone!

>>62347823
>buy a shit $60 motherboard
>dies 6 month later, RMA pattern continues throughout warranty
You do not need the best motherboard, but the sweet spot is definitely NOT $60.

>>62347399
Yes, Phenom II IMC had a DDR2 mode.

>>62347897
Haha, holy shit. Locked i3's are the kind of crap you find in Dell prebuilts running in hospitals and govt. buildings that their secretaries type on.

>>62347906
Actually I mostly use my PC for CAD.

>>62347912
Is it that hard to imagine that people run programs parallel, or record a video and encode another video at the same time? And these are low end normalfag activities, you don't have to be a special snowflake for it.
I don't get this whole logic, if we don't need 150% of the current cpu performance right now, then why develop and manufacture stronger cpus? When the time comes when we can't use our pcs to jack shit because they are too weak, we start developing new technologies then. Is this what you want? Stall advancement for ten years every five years because HURR I CAN BROWSE THE INTERNET JUST FINE DURR? Really?

>>62347954
It's really amazing how all these neo-luddites started crawling out of the woodwork as soon as AMD started actually releasing competitive CPUs again and forcing Intel to start moving.

>>62345963
Oh no, Zen2 will have at least 10% IPC bump on average and NO meme SIMDs (AMD has GPUs, you know. Vega20 is 1H 2018).

>>62347897
>to use your CPU properly
yes, to overclock the chip you paid premium price for, while splurging another extra for compatible motherboard

>>62347938
>dies 6 month later
kek

>>62347954
or maybe for your specific kind of you you'll buy the hardware to back it up
is it so hard to imagine that HW is not top-shit-tier-brand-new-mainboard

>>62347897
K E K
E
K
>>62347947
At least the good thing about i3 prebuilts offered to business is the rather-decent support package. With a custom-built you get none of that. Good luck taking weeks RMAing your B150M+

>>62347976
>i'm gonna overclock
>I'm gonna watch some stats on 3d mark and jerk off to that

>>62347983
>he have not heard of Biostar
>he doesn't know low end Biostar boards LITERALLY dies in 6 months

>>62347995
>Good luck taking weeks RMAing your B150M+
don't warry about that

>>62345806
i see what you did there

>>62348001
WHY THE FUCK WOULD YOU DO THE THING YOU PAID EXTRA FOR RIGHT
WHO IN THEIR RIGHT MIND WOULD DO THAT HUH

>>62347969
Just so you know, before it became the absolute cesspool it is, I was into speedrunning that made me record game footage then encode it which took hours. This was ten years ago, it's not a new trend, but people had to deal with it because there was no option, thanks to Intel. Go fuck yourself.

>>62348011
>claims to work in industry
>offices can go weeks with reduced productivity
How many SMBs have you pushed to bankruptcy?

>>62348040
>0

>>62348056
>admitting you are not actually in the industry
Finally you are being honest.

>>62348056
that sounds like the exact amount of ones you worked at

>>62348007
>LITERALLY dies in 6 months
you have at least 1y of warranty

>>62348076
>>62348085
sure..

>>62348089
downtime is more of an issue in corporate enviroments

>>62348098
you won't buy biostar for a corporate enviroment

>>62345476
So AM1 is for DDR1?

>>62348107
but that's exactly what the retard arguing here is advocating

>>62348130
No, this one was an anomaly (just like FM line).

>>62348144
omfg
r u retarded ?

>>62348159
not as retarded as the dumbass who thinks a $60 low-end board is perfect for everyone

>>62348107
You won't even buy custom-builts for a corporate environment. You need a support contract or else you are in hot soup once something fails.
>>62348159
You won't buy any custom builts either. Stop assuming the custom built you built for your 7 year old niece is a sign that you are in the industry, it is hilariously embarrassing.
>>62348089
You do get 3 years with Biostar, but you will probably have to visit them 6 times through the warranty period.

>>62348198
how the fuck you came to the idea that custom-builds are ok for a corp env??

>>62348213
>>62347897
>Samefag attempting to do damage control

>>62348238
that's me you fucking idiot
I'm asking where the fuck I said it's ok to buy custom builds for a corp env ?

>>62345639
My old pc (upgrade evert 3 years) becomes my media pc. Some point soon I will begin upgrading my house Lan to 10gbs.

File: [HorribleSubs] Aho Girl - 08 [720p].mkv_snapshot_09.37_[2017.08.23_20.17.22].png (992KB, 1280x720px) Image search: [Google]

992KB, 1280x720px

>>62347823
>$60 bioshit board
>lasting for 5 years

>>62345963
Zen2 when?

>>62348287
>GENERAL USER
are you blind or retarded ?

>>62348311
General users use laptops and tablets. Try harder Biostar shill.

>>62348311
The kind of general user you think of only needs a smartphone. I don't consider myself even a power user and I plan to upgrade my 1.5 year old rig.

>>62348311
'general user' can be either an old lady sending email or a gaymurr who spent $3000 on the PC of his dreams
you can't just generalize like that

>>62347897
Why wouldn't you overclock? It gives a 10-20% clock speed boost for the cost of spending $50-$100 more on a k cpu and a better board. In return you get years of better performance. That is a hell of a lot of value for the follar.

>>62348378
>upgrade my 1.5 year old rig.
why
>>62348409
>gaymurr
is exactly that user who would do just fine with 60$ mobo but buys 200$ one
>>62348456
because I can't load my CPU over 60%
why would I ?

>>62348477
Why are you even on /g/ when you're a low end user?

>>62348456
>pay extra for mobo and cpu to overclock
Amd doesn't have these Jewish tricks

>>62348477
>because I can't load my CPU over 60%
i'm sure that's the case for everyone out there anon
sending those emails must be very taxing

>>62348549
it's the case for 80% of the users
you can do 60% with a 3 VMs running at the same time + bunch of oth sw.
that's not small task

>>62348007
Funny, the Tpower X79 i have has been running for going on 3 years now. And back in ye olde P45 days their P45 board was literally the best possible board for Core2Duo overclocking.

>>62348580
okay, spill your beans, what exactly DO you do?
saying
>hurr 3 VMs durr
doesn't really tell anyone much

>>62348546
>Why are you even on /g/ when you're a low end user?
so /g/ is for high-end users?
since when ?
/g/ is full of anime fags, dicks, pedos and faggots wanting attention with *fetch threads
/g/ from long time ago is no about tech

>>62347897
>cpu load is 60% running a full-working OS and a 1-2 virtual machines at the same time

>60% of cpu used up just from 3OSes
>he doesn't do anything else with his cpu other than shitpost

>>62347146
native sata 3, on the chipset. non-native sata is shit

>>62348583
Learn to read, I am talking about $60 BIOSTAR boards with 3 VRM phases and D-PAK MOSFETs. High end Biostar boards can be actually decent (at least on paper), for example some of their GANK series a few generations back have IR Powlrstages in the VRMs

>>62341253
>Ryzen 2 when?
You™ Just® Wait©

>>62348604
Yet non native Sata and usb3 has worked well for 5 years. These are not good reasons to upgrade from Sandy. Ryzen finally was.

File: 1499440022263.jpg (162KB, 969x581px) Image search: [Google]

162KB, 969x581px

>>62348631
Yes goy, yes. Patience is for morons. Impulsiveness is king. Buy our new z270 platform motherboards and CPUs before they become obsolete next year!!!

>>62348593
Those screenlets doesn't view threads like this. Look at their specs, many of them are running C2D and Celerons. Reminder that you are running an i3 for crying out loud. Your desperate attempt to extrapolate them and overgeneralize /g/ is really pitiful.

>>62348589
linux host os
one for lightroom\illustrator
one for 3d max
and one for work ad/sccm/ and some oth shit

>>62348677
I'm still using my AMD Athlon 64 X2 4800+ and GeForce 9600GT from 2007.

>>62348682
so you are like a secret society ? hiding between the "regular" threads?
can you not hear how retarded you sound ?

File: d5c.jpg (36KB, 297x266px) Image search: [Google]

36KB, 297x266px

>>62348702
Bad goy, BAD.

>>62348708
>he doesn't realize the people in screenlet threads are the same few, many with many screenshots of their different systems
>he doesn't realize the people discussing performance, coding etc steer clear from those threads, and they were even requested to be banned by many
>no argument, resorts to ad hominem as last resort
>being THIS desperate

Not buying Intel is inherently anti semitic and should be outlawed

>>62347200
>USB 3
Yiu bought the wrong mobo then.
>m.2
Is just a way to hide away SSDs so your guts look more rice. Nothing is stopping you from getting the same speed.
>usb-c
Pointless for desktops, there's nothing to use it for and it's the same speed as usb3 at best

>thunderbolt
Pointless for non apple desktops, you have pci, usb3 and Sata which is cheaper and better.

>>62348690
what kind of a job makes you use those things in parallel?
also
>lightroom on i3
>while doing anything else
my condolences
i unfortunately know that feeling and i'm glad i'm not on a garbage bin pc anymore

>>62348767
>what kind of a job makes you use those things in parallel?
non, I got side projects and hobbies you know

>>62348749
Gas yourself.

>>62348779
and somehow you can't fathom that people might have hobbies that require more powerful computers that won't work with $60 motherboards

astonishing

>>62348767
actually I've don't have any problems with LR under VM or i3

>>62348791
>you can't fathom that people might have hobbies that require more powerful computers
examples ?

>>62348761
This desu desu, not buying Intel should be illegal.

>>62348806
video games
streaming things
video editing
homelab
compiling things bigger than hello world

>>62348836
so you are saying if you buy a 200$ mainboard those programs are gonna run faster??
really?

>>62348867
>hobbies that require more powerful computers
E. G. More powerful cpu, gpu, more ram
>derp muh $50 mobo

>>62348867
i'm not sure why you pretend overclocking is something people don't do

>>62348891
do you need i9 cpu and quadro for that ?

>>62348934
because people don't do it

>>62347872
The sole reason people are excited for Coffee is an overclockable 6 core on their consumer-grade chipset instead of HEDT. They should have done this with Skylake, it's absurd that they're finally moving to 6 cores.

>>62348934
when you buy a car do you chip tune it ?

>>62348867
>>62348948
>/g/ on CPU's
>"no one cares about overclocking CPU's irl"
>/g/ on GPU's
>"make sure you undervolt your AMD GPU goy, it's perfectly normal to expect consumers to tweak voltage settings on their $500 premium product!"

>>62348948
people did it in the days of Pentium 2 and they do it now
especially since it comes down to literally clicking three or four times in UEFI
>>62348982
wow, and i thought food analogies were bad

File: strawman.png (199KB, 332x312px) Image search: [Google]

199KB, 332x312px

>>62349022

1380 BGA masterrace

>>62349084
>wow, and i thought food analogies were bad
why? the engine can take it, the breaks and suspension can take it.. why don't do it? On your overpriced car? Won't you want to have more bang for the buck?

>>62348867
Jesus christ, retard, at the very least, realize that the pricier boards are built with traces that can handle more power without burning up. Your stubborness is making you look like an ass.

>>62349084
>people did it in the days of Pentium 2 and they do it now
especially since it comes down to literally clicking three or four times in UEFI
Do you talk to people outside internet ?
I mean like real people? Those who walk outside? Do you know how many of them care about overclock? ask them

>>62349157
>are built with traces that can handle more power without burning up
that's nice 200$ justification

>>62349127
oh, sorry, i forgot that car manufacturers release two versions of the same car, one that you can fine-tune and the other you can't

i forgot that Civics come in shiny boxes with a "ECU TUNING CAPABLE - UNLOCKED" sticker

oh wait, this isn't the case. But it is, if we're talking CPUs


oh, and just to spite you even more - people whose hobby is tinkering with cars do exactly the thing you just wrote about

>>62349195
people whose hobby is tinkering with computers are not the general user

>>62349214
and your idea of 'tinkering' is doing anything other than checking email
even prebuilts come with unlocked CPUs nowadays
on top of that, overclocking is encouraged on ANY Ryzen CPU

you're literally the case of
>nuuuh i don't need it, so nobody does

>>62349183
Enjoy your D-PAK MOSFET iron choke 3 Phase VRM that fails every few months

>>62348477
>why
Because I was stupid enough too fall for Intel and the i5 memes. First time in life, never again.

>>62349183
It is when you run two gpu's and an overclocked processor. I work as a tech for a manufacturer and see PLC's shit the bed left and right from power surges during bad weather. Because of this, I buy the most rugged components I can afford for my personal machines.

>>62349264
not everybody lives in a 3rd world country

>>62349264
>during bad weather
so your hi-end mobo won't get electrocuted but it would if you had a standard one?

>>62347078
This is really incredible.
Knowing Intel they're probably not compatible. Remember with kaby lake and the whole "bios update" thing where you had to stick a skylake.
What a shitfest.

And this chipsets seem to be very different from articles like in the OP. But they should say it explicitly.

>>62349315
Well in many cases yes, especially Asus boards

if zen2 could be released at 4.5ghz stock, it will be over for intel forever. only problem with ryzen is the clock wall. IPC is vastly superior than intel at the same clocks.

>>62349314
Neither do I. I work in the largest city in my state. Home of Dyn and Autodesk. Power surges happen for lots of reasons outside of shitty infrastructure.

>>62349342
you said bad weather
let's say standard thunderstorm
lightening - 1 billion volts and contains between 10,000 to 200,000 amps
and your hi-end super protected motherboard would remain intact ?

>>62349360
the design is for 5ghz. it will murder intel

>>62349415
My RCD would have tripped BEFORE that happens.
>inb4 american fusebox

>>62349384
No they don't.

>>62349415
Utilty poles are protected from this sort of thing thru grounds and surge arrestors. But, a sudden surge, or even a transformer on your street that is old and failing, may output less voltage and cause issues.

>>62341663
l>
l
l3

>>62349415
>a lightning strike has 10TW of power

A chipset change next year keeps the goyim in fear.

>>62349461
Ever hear of a bird shorting out a transformer? It's happened three times where I work. Literally nothing they can do about it.

>>62349510
>His transformers are not UNDERGROUND
WHEW

>>62349510
They can not have shitty, above-ground infrastructure.

>>62349510
>>62349384
>>62349264
lmao amerifats always make me laugh

>>62349527
>>62349525
It's easier to repair or even cool if it's in the open air. Why can't they put a cage over it is beyond me though.

>>62349360
AFAIK IPC is very similar and we could assume that it's the same for most comparisons.
>>62349417
5 GHz is server goal. 32 Epyc runs 2,2 GHz stock, 3,2 turbo. Imagine 32 core Zen2 going 5 GHz on air.

>>62345497
They won't murder anything because by that time Intel will have its new arch too.

Even with the current Ryzens they aren't murdering anything. AMD CPUs just have a better per-core price if you absolutely need more than 4 cores in your CPU.

And it's also very possible that Zen 2 CPU might require a new motherboard too even if they use the same socket.

>>62345454
This sounds like a shillpost, the only info it gives you is the amount of pins, amds sockets have a generational name, what does the amount of pins have to do with transparancy anyways?

File: 1480178904547.png (58KB, 229x228px) Image search: [Google]

58KB, 229x228px

>>62346453
its motherfucking DOOM. don't question his ability.

>>62350577
The only thing Intel will have late 2018 is 10nm Icelake though. It won't be as impressive as Coffee Lake's clockspeed although it will probably be much more power efficient and maybe a bit cooler. Intel has nothing new until 2020.
>And it's also very possible that Zen 2 CPU might require a new motherboard too even if they use the same socket.
That is a good point, AMD said the AM4 SOCKET will last until 2020, they said nothing about chipset compatibility.

>>62341253
>I spent 200$ on my board too
My z170 motherboard costs $400 and it isnt even "le gaymer epic rgb edition". I bought a ws board with the intention of putting in an enthusiast cpu when then get released.
How do you think I feel seeing all of these new 1151 cpus get released and be totally useless to me?

>>62341636
amd boards use the same socket for a decade dont they?

>>62351050
Sockets don't mean much. Just look at the coffee lake situation.

File: 1449258351081.jpg (63KB, 768x614px) Image search: [Google]

63KB, 768x614px

>>62341663

>>62351030
If you wanted a workstation board why the fuck didn't you just get X79 or X99? A $400 mobo is clearly HEDT territory, you're a fucking retard for paying $400 for Z170.

>>62341709
That still has tons of fucking issues with comparability and whatnot

>>62341709
>New Intel upgrade pack for 1151 socket motherboards!
>Can be bought separately for any Coffe Lake cpu.
>Contains a Skylake and a Kaby Lake dummy cpu (doesn't function as a real cpu, only for bios upgrading on z170/270 motherboards)
>Now for only $800!

>>62341253
New chipset a year keeps the goyim in fear

>>62351933
I think bringing back stuff like the Pentium Overdrive would be a neat idea actually. Kinda pointless since the differences between CPU's each gen isn't that amazing but still.

>>62351369
I was pissed off with all of my "old" shit with caps failing left and right, hmdi ports dying out, hdds clicking and shit. I wanted "new" stuff, not something thats been sitting on a shelf for the last 7 years.
As for my $400 motherboard, I lined up 2 sales and a mail in rebate and got it for about $260.
A part of me regrets not going with x99, but at that point id be buying a cpu that costs thousands and in the price range Id want a dual socket motherboard and then I just start upping the price a few bucks here and a few bucks there until my total is over 5k
In the end I came out with a decent skylake build for under $1000 so I guess the limit worked out good for me. Also I get that nice single core performance.

>>62348761
Fuck of /pol/

>>62352126
oy hes blowin our cova

>>62351984
Fuck off nazi it doesn't even rhyme

>>62341709
Yeah except everyone knew from the beginning that Kabylake could work on Z170 boards.
They've explicitly stated here that it will not work.

>>62348287
I bought a 50€ mobo that lastem me thru 2004-2011.
Suck a dick.

>>62349235
No it doesnt.

>>62352495
You got lucky.

>>62348948
Hey fuckface, I bought an AM3+ motherboard with beefy heatsinks on the high quality MOSFETs so they wouldn't catch fire when I'm pushing my CPU past 4.5GHz.

https://www.youtube.com/watch?v=UgqlhbrfJF4

>>62350650
>>62350577
Intels new arch wont happen till 2020 at the absolute earliest, with a if shit doesn't hit the fan of 2021, and a shit hit the fan of 2022-2023

with all of the talent at intel jumping ship, and the last time they made a new architecture being itanium (remember the core of an i7 is an i3 that is a suped up pentium pro)

And lets not joke about this, intel has nothing year over year since sandy bridge and the last real jump that wasnt process related was haswell. intels 10nm process is a step backwards and they are crossing their fingers that the 10nm+ will be at least on par with 14nm+++

Meanwhile, amd is getting a process deved by ibm, that is on track or ahead of schedule, that targets 5ghz as the norm, not the delidded water cooled outlier, this is the put the heatsink that you payed 20$ for on the thing and call it a day run speed.

amd already technically can do 20% more per clock then intel, at least on paper, and all intel really has to counter is a widely used compiler and clock speed, with the 8000's a 6 core and 6 core 12 threads that are both around 100$ too expensive to really compete.

The moment that amd gets 7nm, they could realistically put the 32 core server cpu in a laptop and have the shit run 2.5-3ghz and shit on intel for the same power.

Shit is getting real interesting.
Cant wait for drivers on vega, im out of the gpu market till early 18 and everything in amds gpu makes me think its the better long term gpu, see deus ex 2gb maxed out benchmark where the gpu used system ram and the hbm as cache. this would let me stay on the gpu long term without hitting the hard bottleneck that vram use to be.

>>62345927
>>62345939
>>62345963
>>62347971

what we know, is amd will likely be going 1:1 with infinifabric, which would bring the performance up quite a bit for normal consumer uses
I believe they are also putting in a bigger fpu pipeline,
and likely the few hardware fixes for issues present in the first run.

the infinifabric going 1:1 could uptick 7-15% depending on application,
But the big one is the ibm node, which is targeting 5ghz for normal use

>>62347777
nothing, the only thing that is 'wrong' is in situations where you are core limited, it cant make up for it in clocks, that's soon to change, but first run ryzens are still great.

>>62352082
If I remember right didnt the 6 core cpus cost 400~ at this point.

>>62354841
Hbcc is probably handy for vega11 with 4gb vram

>>62355714
We don't know VRAM config for Vega11 yet.
Maybe it's 2 stacks of low-cost HBM (remember that AMD is the biggest ally of DRAM cartel).

>>62355745
Doesn't change the fact lower vram gpus profit from it the most. This way they don't even need the extra stacks. 4gb + hbcc gonna be enough for 1080p gaming

>>62355846
1 4-Hi stack is a waste of perfectly good interposer.
Anyway i sincerely hope they'll finish the drivers for the next major Crimson update.

>>62355745
>>62355714

as much as I want vega just so I won't hit a vram cap before the gpu is useless, chrome is eating 2gb of vram right now, and because its fucking nvidia, I can't just kill the process without killing the browser for some retarded reason, I wouldn't take a gpu with less than 8gb of ram now, even if the ram technically doesn't matter, It still better to have it then not.

>>62355846
2gb vram on a game that eats over 4gb (lets say up to 6, I don't think I ever hit a limit in deus ex on my 6gb card) that would mean that it's at least good for 3 times the needed vram, in other words if you have 4gb, you effectively have 12gb, I just dont want to rely on that function, and would rather just use that functionality as a life extending feature rather than the moment I buy it its getting used.

>>62355056
Probably. I had been living under a rock clinging to old af hardware like the core 2 duo Ive had since I was still a kid. I was so far out of the loop that its not even funny.
The 6700k can run at 4.6 ghz with no problem, when it comes to single core performance does something like more l3 cache really make up for a clock speed thats over 1ghz slower?

>>62341253

It will be keyed differently though. It is a "different" socket 1151.

>>62354841

Current core chips have little with common with the Pentium Pro. They are both are build on x86 ISA and have the basic stuff (FPU, APU, x86-32 registers). Outside of that limited scope, that have nothing in common.

File: 1411177440158.png (68KB, 500x500px) Image search: [Google]

68KB, 500x500px

Coffee Lake is gonna be so much faster it requires new chipset
please buy our CPUs goyim

>>62356170
They were still iterated from P6.
And every time Intel tried something BESIDES iterating on P6 the result was bad.
Really-really bad (NetBurst), we wasted billions of R&D money for nothing bad (Itanic) and why the fuck would you pay x86 tax for GPU bad (Larrabee).

>>62356022
from my understanding, amd can get 5 things done per clock, intel can only get 4 things done

However how many programs were made to really use that right now?

now, when you are talking about running over 4ghz, you are hitting a weird issue, programs that use 1-2 cores typically are fine no matter how slow the cpu is, so there is little benefit, and programs that use more then 4 cores, are able to take advantage of amds more cores so clock speed is again irrelevant till you hit something big. for games, the average game is 10 fps faster on intel then amd even with itels stupid fucking high oc's golden samples can get, and we are talking about the difference of 100-110 or 110 to 120 here, nothing substantial.

outside of outliers like arma or the 3 other sim games that next to no one really plays but rely entirely on single core, the clock speed in games is almost moot.

However for the price that a 6700k comes out, amd gives you an 8 core that will hit at least 3.6 if not 3.8ghz on their packaged cooler, which gives you enough performance to play 60fps even if you keep it at 3.2ghz. What you get in return for the slower speed, is 8 cores, which load balance better the 4, so while in game you may max 100% of 4 cores, on intel those same 4 cores still have background applications to work with while on amd those 4 cores can be 100% isolated

>>62356236
>from my understanding, amd can get 5 things done per clock, intel can only get 4 things done
Yes, Zen has 5-wide decode compared to Core's 4-wide.
But decode is not everything.

>>62356170
here is my understanding.
Intel had the pentium pro which was amazing
then they iterated on that for pentium 3
realizing netburst was a dead end, they use the pentium 3 for core 2
they got some ibm ip which made Nehalem possible to build up from core 2
They picked the the low hanging fruit and got sandybridge very minor iterations and instruction set additions since.

all of this built ontop of pentium pros uarch.

>>62356227

Core family has a ton of stuff from Netburst (all of the good parts of it).

Calling it a suped-up Pentium Pro is being disingenuous at best.

>>62356249
no, its not, but it is something.

>>62356274
The good parts being uOp cache and register sizes. And these were introduced in Nehalem/SB respectively.
Aaaaaaand that's it (Netburst was bad).
>Calling it a suped-up Pentium Pro is being disingenuous at best.
It is one. It's not as elegant as Zen, but it works, and works nicely.

>>62356303

Netburst's awesome prefetcher units, quad-pump FSB and longer pipe staging (That is what allowed Core 2 to clock as high as Pentium 4 when P6 was capped at 1.6Ghz)

In retrospect, Netburst did a lot of things right.

>>62356913
>longer pipe staging
Pipelining is not a super speshul sikrit feature of NetBurst.
Pipelining is inherent to all CPUs.
>quad-pump FSB
>2003-onwards
>FSB
Egh.
>In retrospect, Netburst did a lot of things right.
And Zen inhereted Bulldozer's branch predictor and power management (from Carrizo). Bulldozer was still bad.