Ryzen 5 lineup IS HERE

intel better lube up.

File: Br78sel.png (201KB, 1028x552px) Image search: [Google]

201KB, 1028x552px

Intel on suicide watch.

worthless when a G4560 does the job at 70$ desu

I'll wait for reviews before getting hyped this time, it should be good either way.

>>59413166
Then get the Ryzen 3

>>59413148
1500x looks like a viable CPU for a decent midranger.

File: 1488664232472.png (84KB, 653x726px) Image search: [Google]

84KB, 653x726px

>>59413166
>g-guys just get our 2 cores CPU, i-it's the same

>>59413185
call me if we´ll ever get 4-4.5 ghz out of the box with >4 cores

File: DBEURmm.png (91KB, 653x726px) Image search: [Google]

91KB, 653x726px

>>59413171
only if it'll be cheaper / better at a close price
>>59413185
>w-we h-have r-real c-cores a-at s-same p-performance b-but T-HEY ARE P-PHYSICAL!

>>59413162
>3 times as many threads
>only 1.69 times performance in a heavily multithreaded benchmark

Until AMD beats the G4560 I don't care.

*Blocks ears*

Lalalalalalalalalalalalalala

>>59413166
>2017
>2 Cores
How will I run VMs

File: tumblr_ojlfeiYr7Y1s307p6o2_r1_500.gif (2MB, 500x281px) Image search: [Google]

2MB, 500x281px

>discount poo

An OC'ed R5 1600 seems to be the best value for applications.

For games obviously the only choices remain G4560 and i5-7600K.

>>59413235
>tumblr
pathetic

Gaymur here, is there any reason I should upgrade my i5 4460 any time soon

Mainboards still MIA

>>59413253
What are you talking about?

>>59413253

wtf is a mainboard?

>>59413267
A motherboard.

So do I get an R5 or wait for R3?

>>59413232
threads

>>59413274
-> >>59413205

>>59413246
nope, unless you go my route. buy 1400 overlock that shit and wait for Zen +

>>59413213
threads != cores

>>59413148
Cheapest ones with the minimum amount of cores you want and overclock, how is this difficult?

File: promo289257259.jpg (164KB, 1920x1080px) Image search: [Google]

164KB, 1920x1080px

>>59413166
>Before Ryzen 7 was released: "Lol who needs anything more than an i5 amirite"
>After Ryzen 7 was released: "Hurr durr you're all fucking retarded if you don't get a 7700k for video games so you can get one extra FPS compared to every other CPU"
>After Ryzen 5/3 is released: "Lol why would you need anything other than a G4560 for video games?"
Really fires up those brain cells

>>59413300
amd always losing xd

>>59413300
Everyone knows you don't need more than one core @2GHz for games.

>>59413308
Yes i need to dig my old Prescott from it's grave.

>>59413304
Not really, Ryzen 3 and 5 will render Intels entire line obsolete. Hell, Intel did that to themselves when they released the G4560.

too bad this poo don't OC well

>>59413318
i3 line*, I'm a retard

>>59413318
AMD is "their new line will kill em" since 2008
Intel is "we kill em right now" since 2008
git gud

>>59413345
>doesn't know what happened before 2008.
You know, let's pretend that the original Core Duo wasn't just like Ryzen. An untested, unproven processor when it came out. This is what's happening now.

For years, Intel tried to combat the 64X2 and pathetically destroyed the Pentium Brand with Netburst.


Hell if the 64X2 didn't happen, we probably would've just tried to push everything on raw single-core performance.

>i3, R3
>i5, R5
>i7, R7
This is deliberate, right

>>59413369
Yes.

File: intel.jpg (151KB, 1176x1200px) Image search: [Google]

151KB, 1176x1200px

>>59413300

>>59413345
Turns out it's 2017 now and not 2008, dual core CPUs that cost more than $100 serves no purpose whatsoever in a desktop.

>>59413384
>muh temporary performance increase for muh games

>>59413363
the past and the future predictions dont matter, intel is king now, until amd becomes king then amd is shit
>>59413384
dual core cpu's with threads definitely serve more purpose than 100W processors, you should get a time machine and get back to the 90's

Looks like I'll be getting a 1500x this year.

>>59413428
>the past and the future predictions dont matter, intel is king now, until amd becomes king then amd is shit
I hate to say it, but you sound like some fanboy. We should like that Ryzen is bringing the pain for the mainstream offerings. This would push intel to release something that is badly needed. A mainstream quad-core offering.

>>59413410
Are you ok? Where in that post did I mention games? Buying an i3 is even more idiotic if you're going to use it for more things than just games since you're going to need at least a quad core for any remotely demanding multitasking and a G4560 performs pretty much identically as the i3 for half the price.

>>59413428
Every Pentium and i3 Kaby Lake is 2C/4T you downie

I caren't one bit. I'm getting a 1700 when I get my tax returns (Expected to come in May) and that's final.

File: 1488691675731.jpg (119KB, 915x678px) Image search: [Google]

119KB, 915x678px

>>59413148

So now that the complete Lineup is official, what would be the informed patricians choice?

I'm still undecided if i should get a 1700 or a 1600X

>>59413477
1700. OC that shit to 4.0 and call it a day.

>>59413477
Buy the 1400 and overclock to 8c/16T 3,9GHz, it's the same chip

>>59413477
I really would love to see benchmarks for all of those.

Do you think we'll start seeing any anytime soon? I don't want having to wait a whole month.

>>59413477
1600X unless you absolutely need 8 cores

>>59413464
How does forcing intel to get better make AMD any better than a motivator? Their CPU's will still be shit
>>59413468
You are taking this shit too seriously my man, I suggest leaving the PC and taking some deep breaths, it's just an argument, no need to be so insecure

>>59413540
(you), Ramesh.

>>59413557
I prefer rupees but thank you

Why quads are so low clocked?

>>59413471
same here. 1700 is a sweet as fuck deal and will proably last me until the end of the AM4 platform.

Realistically speaking how shit will they be in vidya

>>59413580
Binning

AMD produces lots of Ryzen 8 core CPUs. Some turn out better than orthers. The best get sold as R7 1800X, the second best as R7 1700X and so on. The worst of worst get 4 cores disabled and sold as R5 1400.

>>59413602
Very good after devs/microcuks will start optimizing for Ryzen. Windows and gaems are basically unaware of Ryzen topology.

>>59413205
So it doesn't support 2x PCI-E 3.0 16x?

>>59413638
What purpose would this have?

>those stock clocks/turbos

God dammit this confirms my worst fear: the 6/4 core parts aren't going to clock much if any better than the 8 core parts. I hope the process matures and in time we get better clocking potential like with polaris and the FX chips.
Also a little disappointed to see not only do the 4c/8t parts have the lowest clocks, they're a bit more expensive than I'd hoped.

Guess I'lll see how things improve as all the woes inherent to such fresh tech get sorted out before I buy.

I also really wish the high end wraith were a pack in with the 1500x or 1600x. The old one matched or bested a lot of pricey slim ITX coolers and that alone could make these things an awesome buy.

>>59413477

It's 6/12 not 8/8

Please redpill me on this: If an 8 Core 1800x at 4 GHz can't beat an 7700k in games, why should an 6 Core at 3 GHz be able to ?

>>59413656
Multiple graphics cards?

I mean, a PCI-E 3.0 16x link is probably difficult to saturate, combined with scaling, but nonetheless, I'd imagine we might see a performance limitation in very high end setups with that.

File: 1488425529478.png (307KB, 915x678px) Image search: [Google]

307KB, 915x678px

>>59413166
Hi corelet

>>59413674
>Also a little disappointed to see not only do the 4c/8t parts have the lowest clocks, they're a bit more expensive than I'd hoped.
If they are competitive with the 7500 and 7600 the pricing is what you'd expect.

File: image.jpg (57KB, 630x449px) Image search: [Google]

57KB, 630x449px

>>59413687
>If an 8 Core 1800x at 4 GHz can't beat an 7700k in games
but it can, and it does when there are no software bugs gimping its performance a whopping two weeks after its release.

>>59413477
See this
>>59413713

>>59413745
>the only guys to do some ACTUAL testing are some germans
Really makes you think.

>>59413768
>some germans
those are the germans though, also PC is pretty big platform in germany so they have to be honest or they'll get carried out on pikes

>>59413369
It's deliberate but also somewhat misleading as in most cases core counts are doubled. They should call the R3 R5, R5 R7 and R7 R9. Oh well.

>>59413768
Germans hate jews (Intel)?

File: good goy.png (604KB, 800x523px) Image search: [Google]

604KB, 800x523px

>>59413428
Intel shills are this desperate.
>W-we're still on top in marketshare!
Half the reason they have that marketshare is scummy backroom deals to keep AMD's superior product out of OEM PCs.

I really wish the marketing weren't so heavily targeted at gamers. These chips will be great for workstations, but the failure of Bulldozer and bechmarks with 7700k's at 5ghz will mean these chips will forever carry a stimga of "MOAR COARZ." Is gaming really the only remaining market for high end desktop PCs?

No doubt these things will take the server market by storm, though. Intel has no answer to those MCMs. It ain't gonna happen overnight, but that's obviously what AMD was ACTUALLY targeting. That and laptops. Dat TDP man. Can't wait to see how the APUs fare.

Saldy Vega is a similar story. AMD wants dat GPGPU money, but they'll obviously still market it to gaymurs.

>>59413814
the point is pricing
core count is cherry on marketing top

so when proverbial pleb sees i5 4c/4t vs r5 6c/12t what would he chose? this time it's actually better deal though

>>59413847
>i5 4c/4t vs r5 6c/12t
Forgot the i5 isn't unlocked, while the r5 is.

>>59413846
this time they are under 250, so i5 is automatically out of picture because even 1700 beats it
6 core is ideal for gaming, it would probably tax ccx issue less since ton of games support 6 thread but not that many support 8+
also at 4Ghz it's basically 1800x at games for 250!

>>59413870
also it's going to be even cheaper than z270+i5K
i7K vs 1700 is barely cheaper this one is way cheaper

>>59413743
They're not astronomically more expensive than I'd expected, but I was stupidly optimistic they might offer a ~$130-150 4c/8t chip with the wraith cooler. $169 bottom dollar isn't really much higher, but still.

Inb4
>why do you need a high performance cooler on a 65w chip, you entitled cunt?

I'm more interested in the noise reduction. Was also hoping to see boosts above the 8c parts, suggesting better OC potential.

That 1500X looks juicy as fuck
>90% performance of intels' $350 chips

So, what the highest clock for stability for 6c ? Using wraith/high end air cooler ?

>>59413880
Have you seen how many retards are calling the 1700 overpriced because the now similarly priced 7700k does better in gaming? This is obviously a completely better offering than FX ever was, but bringing so much attention to gaming only brings more attention to the one area these chips don't utterly smash Intel at

How would 1400 compare to the current Intel cpus? Which one would be equal roughly speaking

File: 1482336217657.jpg (17KB, 542x540px) Image search: [Google]

17KB, 542x540px

>4 cores/8 threads
>3.5ghz

>>59413975
>Still can toe to toe with 4.2ghz
Really makes you think

any bets on the 1400's performance?

im hoping its in the 4690 ball park

>>59413235
>tumblr

File: allies.jpg (145KB, 1500x1054px) Image search: [Google]

145KB, 1500x1054px

Wew Lads.Great pricing on the 1600X there. I fell for the Ryzen 1700 meme when I managed to pick one up a couple of days ago for $398 AUD (Absolute bargain in Australia).

I'll just run it on the stock cooler for a year until I can upgrade to a good Ryzen optimised air cooler. Hopefully until then the stock cooler does me perfectly for a while.

>>59413988
lmao sure amdrone, expect 66% of the performance of an 7700k at best

Too bad their upcoming GPUs don't look so promising cause I was ready for a full AMD build

>>59414001
where did you get it

>>59413246
I have 4460 too. no need to upgrade. Its fine for a couple more years.

>>59413988
>Toe to toe
citation needed

>>59413996
in what?

https://www.youtube.com/watch?v=DQilK2dOJTg

in games it will be something like this compared to 1700
it's ZEN cores, there is only one zen core and it goes in every chip
that's if the L3 same as 1700 though, 8MB cache for 4c is a lot

>>59414019
http://www.ebay.com.au/itm/AMD-Ryzen-7-1700-Processor-16MB-Cache-3-0-GHz-AM4-8-Core-16-Thread-Desktop-CPU-/332146902775?hash=item4d557f86f7:g:j4QAAOSwTuJYvhWP

20% Discount Code "CYBER" takes it down to under $400.

hmm idk, how much of an upgrade from a fx8350 is this? does is actually work with emulation? thanks fx8350 was a pos for emulation. and can it run the android sdk emulator on windows?

>>59413155
It should lube it's dick for AMD's asshole?

>>59414046
it's upgrade even for sandy, how well do you think it does against FX?

>>59413745
How about you show us all of these other examples where the 7700K is systemtically beating Ryzen?

>>59413148
Still no reason to update my 4690K @4.6Ghz

>>59414061
"how much" is what i mean. like are their daw and dgraphics program benchmarks yet? not for gaming and tossing up between us hich to buy. might get the
>>59414043
this guy got

>>59414043
ty friend

>>59414032
very interdasting

>>59413972
R5 1500x == i7 6700k

>>59414046
Ryzen is still not fast enough for Cemu.
You're best of getting the i5-7600K and a good cooler so you can OC to 5+ GHz.

>>59414091
I Think the same seller is also selling the ASRock X370 Fatal1ty K4 Motherboard for $190 AUD (Post-Discount) which is about $30 less than retail price, if you're interested. I picked up that one as well.

>>59414123
dual core pentium is better than i5 for CEMU it's single thread emulator

>>59414123
I have i7-2600K and it's more than enough for CEmu, stop shilling.

>>59414007
Kill yourself, shill.

>>59414123
>Ryzen with two CCX latency is still not fast enough for Cemu.
ftfy

>>59413509
nah they like to build hype and despair, same with R7 lineup

>>59413148
If the R5 1500X can't beat a i7 7700K then it's pointless that AMD even tries.

p.s Loos and poos don't need to respond to m, I don't want to hear excuses, it's equal ground, 4C8T if AMD can't beat Intels 4C8T then it's totally shit.

>>59413817
he knows

>>59413616
Kek

>>59414154
:^)

>>59414160
Oy vey

>>59413713
multi-core,multi-socket systems have their place if you need terabytes of ram or have some workload that is parallelizable but not embarrassingly parallel, i.e. where latency between different servers would cost you performance. granted, those are niche uses, but they exist.

>>59414123
>Cemu is still not optimized enough
ftfy

>>59414154
Kill yourself.

>>59413296
If xfr actually overclocked and not just slight boost, there would be a good question as to do you want to do it yourself, or just pay to have it done for you potentially better than you can but as its shit, cheapest and overclock to 3.8 at 1.35 volts and test up from there till 1.4

>>59413477
>I'm still undecided if i should get a 1700 or a 1600X

either go 1600 or 1400, xfr is useless in its current incarnation, and you can clock any of those cpus up to 3.8 at minimum, with a severe silicon lottery loss being 3.7

>>59413509
we will see a 4 core cpu benchmark, because if they are all in one ccx or across 2 is a big factor if its good or not.

>>59414297
>or across 2
Don't even say something like that.

>>59413697
sub 3% performance loss goinf from 16x to 8x

>>59413148
That 1500X better overclock like a dream

>>59414077
>130 fps vs 150 fps that means right is better!
fuck that I'm more interested in ryzen's lower workloads, this is a sign of a CPU with room left for improvement by applications, it's like if people forgot about their first quad processor or something, everything used just 25% or 50% of the CPU but look now how Intel's 6700k is already maxed out in most video games.

>>59414319
we don't know how they binned it, are there cpus that are so fucked they legitimately had 4 core skus? did they just take a 6 core sku with one bad ccx and just turn the whole thing off?

with the 6 one we know at least 2 on another cxc are there, but the 4 core... that's up in the air and may be objectively the worst cpu you could buy, with the 1600 being the best if you don't want 8 cores, set all the system shit to run on 2 cores on one ccx, set all games to run on 4 cores on another ccx,

>>59413166
2c/4t was obselete years ago. AMD is finally forcing this to market, Intel was more than happy selling garbage to retarded shills like you. That CPU should cost $40 and come in a goddamn lunch box.

>>59414369
just a follow up kind of, we won't see good 4 cores as a base, we know its all one ccx till apus.

If this is accurate
https://www.youtube.com/watch?v=BORHnYLLgyY

Imagine the difference betwen a 1500X with two good cores in two CCX' and a 1500X with four good cores in one CCX.

>>59414297
>>59414319
see
>>59413150
>the processors are all 8-core models with a CCX or two cores disabled

>>59414369
>set all the system shit to run on 2 cores on one ccx, set all games to run on 4 cores on another ccx,
Has anyone made a program to do this yet? Seems like it'd be pretty valuable for current Zen owners.

File: image.jpg (122KB, 632x758px) Image search: [Google]

122KB, 632x758px

>>59414077
how about you pick your own cherries, bitch?

here's the overall result, and the Ryzen is only bound to improve futher whereas the i7 already played all its jokers.

>>59414372
yes and no.
4 threads won't be obsolete till nothing functions on less than 4 threads, and i'm not sure this will ever happen.

It is possible that in 10 or so years time, speed of single cores makes a breakthrough with a new technology, or that someone makes a really good fuckin process, and they only make cpus with 8 cores 16 threads being seen as the poor people shit like pentiums are currently seen, but current 4 thread processors won't be obsolete anytime soon.

>>59414046
Even the cheapest Ryzen will be leaps and bounds ahead of a 8350

>>59414123
If all you want to do is run CEMU and Dolphin, then a 7350k, or a 7600k, both seem to love AVX performance, which is Skylake's strong point.

>>59414394
speculation based on TDP alone.

>>59414123
Just buy a fucking wii u for the price of the fucking cpu jesus

>>59414396
I have wondered this in the past, but back then, the only way to tell the os what process goes where was all manually and because the fucking thing opens new processes all the time, like chrome, would need constant babysitting.

Though i'm curious if something has come out since I last looked into it. would love to set something up so all boot processes function on 1/2 cores, then chrome can have its 1-2 core containment core when I do something intensive and the program I want to have the most free room can have 4-6 cores depending on if it gains from 2 ccxes or if it looses.

>>59413300
>>59413687
>>59413745
>>59414032
>Playing videogaymes
>>>/v/

Ryzen is truly appealing for productive tasks given the lower tdp than bulldozer. Intel CPUs are overpriced.
it's such a shame that ayymd's marketing is so gaymur-centric.

Why would R5 be faster than R7?

>>59414464
>ayymd's marketing is so gaymur-centric.
it's not, Ryzen was promoted as being for everyone and everything. it's just the gaymurs who are by far the loudest crowd.

File: 1468922895102.png (572KB, 600x580px) Image search: [Google]

572KB, 600x580px

http://www.zolkorn.com/en/amd-ryzen-7-1800x-vs-intel-core-i7-7700k-mhz-by-mhz-core-by-core-en/view-all/

>pajeets and the whitecucks are crying their shitty cpu can't deliver core to core at same ground

>>59414468
Lower latency between cores if it's a single core complex. Also, if those ASRock leaks are true then the 1500X would have 768KB of L2 per core instead of 512KB of L2 per core.

>>59413148
This is one of the greatest business accomplishments in years. AMD just set themselves back up as a legitimate competitor to Intel CPUs.

I hope developers start optimising for Ryzen in the near future. If I had time to play games or edit videos I'd definitely get one of the new chips.

>>59414477
>Haswell IPC with double the threads for about half the cost of intel offerings
seems like a win to me desu

>>59414477
>mostly very similar results with the two largest deviations showing opposite advantages
AMD FINISHED!!!!!!

also terrible writing.

>>59414477
Eat a bullet already.

File: 1425342417559.png (568KB, 960x720px) Image search: [Google]

568KB, 960x720px

>>59414420
>Low-end CPU is enough for me!

>>59414464
if it has 8MB L3 low end xeons are dead.

>>59413148
>$219
How much of an improvement over i7 3770?

Curious to see how this will translate to aud.

>>59414691
We don't know if it's using 1 CCX or 2. If it's 2, it's prob. going to be the weakest of the entire Ryzen lineup.

>same chip as the 1800x
>just different clocks and different cores disabled due to flaws in fab process
>these idiots expecting single core perf to magically increase with cores disabled

Yall are delusional. AMD gave it their best shot and missed. Their stock is already dropping and by august after vega launches, and flops, it'll be back to the trenches with them as usual, rebranding and hyping up worthless shit just like they've been doing for years.

>>59414742
AMD's stock is fluctuating no more than the rest of the tech market.
Fuck, nVidia dropped $30 in 3 months. If you're going to be toting the death of anyone, it'd be them.

>>59414726
The 4c and 2c will probably be the only ones with 1.

>>59414726
Weakest at what?

Not going to lie, I will be using it to play video games, but that's only part of my use and not the primary.

>>59414742
>these idiots expecting single core perf to magically increase with cores disabled
Why wouldn't it? The 1800x performs better when you disable 4 of its cores.

File: 1470712905701.png (5KB, 727x432px) Image search: [Google]

5KB, 727x432px

AMD btfo

>>59414837
This is bullshit. Everyone knows bacon takes longer to cook than sausage.

>the way it's meant to be grilled(tm)

>>59414861
If you want black bacon
>blacked

>>59414877
Well-done bacon is patrician-tier.

I wonder if anyone ever successfully cooked something on a Fermi GPU testbench. Fuck now I'm hungry.

I will wait for Ryzen 3.

5 and 7 are shit and not worth it.

>>59414904
scrambled eggs on prescot were popular at one point

>>59414077
The 7700K gets crushed in thread-heavy workloads.

>>59414917
R3 is APU

>>59414934
1 CCX

>>59414940
>R5
>not 1 CCX

>>59414946
The 4 core part, maybe. We don't know yet.

>>59414477
>At most a 15% difference
>Exactly the same joke of a performance difference tons of people are mad at Intel for

>>59414809
so the 1500X it will be

>>59413817
KEK

>>59413246
Only if you want more cores.

>>59414983
They could have just done something stupid like disabling 2 cores on each. We'll see though.

>>59413768
deutsche genauigkeit ;^)

germans do everything 100%, doesn't matter what

>>59414940
>>59414946
>>59414983
>>59415009

This is the thing. Before we know this all our conjecture is worthless.

>>59415011
Yes Germany now make another actually decent Anno game please.

>>59415009
don't say this, i need that savings for vega

hope they fix that shit and i could go for the 6c/12t

i want a goodam apu with hbm and a new micro itx form factor with no goddam shit on it.

>>59414161
Intel does this as well.

>>59415045
Could just buy a 7500. I will depending on how the 1500x performs.

>>59413148
>there's no way AMD made any significant IPC improvements in the span of a few weeks
>those chips don't actually clock higher
>they're probably not gonna OC much higher either
>the 6 core models will most likely be dual CCX chips with two cores disabled and thus suffer from the same context switching issues
>they're not actually cheaper than equivalent i5 chips (at least until their SMT implementation starts working in the real world giving them a real advantage)

I have a huge hard on for the R7 line but this is disappointing

what am I missing?

Does anybody know how the 1700x/1800x will do emulation wise?

>>59413235
Who's that semen demon?

>>59414691
It'll probably beat the 3770. Unless you need the extra cores it's not going to really benefit you.

>>59415100
8Mb L3 cache for 4 cores, and 16Mb for 6 cores.

Low end xeons are dead, everyone got them for big cache.

>>59414325
If even, I've seen benchmarks of 1080's losing less than 1% performance on 8x Vs. 16x. Pretty much within the margin of error. PICE 3.0 has some pretty monstrous speeds per lane.

>>59415101
ps3 apparently needs MROE cores, everything else is passable but not the best
nobody did actual benchmarks for some reason, there is a few but they are old day one irrelevant with new BIOS updates

>>59415101
should perform on par with the stronger i5 chips unless there's some Intel specific optimizations going on

>>59415119
fair enough, but that's an extremely niche requirement

>>59415141
>extremely niche requirement
games like L3, for everything office like it has more than enough IPC

>>59414472
>it's not, Ryzen was promoted as being for everyone and everything. it's just the gaymurs who are by far the loudest crowd.


Except it's not, stop being one of those /vg/ gayming idiots who can't even understand a product presentation

>>59413814
It's strange they used the same names as for the GPUs though

>>59415165
if games like cache then why the fuck is Ryzen behind 90% of the time?

>>59415203
cache gets dumped into RAM due to OS doing something wrong, apparently scheduling is only small part of it, there is something more

File: 3dd4fc4284d0[1].jpg (84KB, 654x859px) Image search: [Google]

84KB, 654x859px

>>59415203
Ryzen has ridiculous L3 latency due to having two CCX.
https://www.techpowerup.com/231268/amds-ryzen-cache-analyzed-improvements-improveable-ccx-compromises

File: 1470872997186.jpg (25KB, 435x129px) Image search: [Google]

25KB, 435x129px

>>59415174

>>59414742
You said it yourself: Different clocks. A higher frequency means higher single core performance.

>>59414243
>What is server virtualization

Considering a very lower price there a whole market getting amd chips to virtualize servers

>>59415267
except the R5 chips don't actually clock higher

>>59415340
>4Ghz is not enough

>>59413148
>paying more for literal miniscule clock speed boosts and nothing else

Are amd retarded or what?

>>59415360
You only pay for it when Intel does it.

>>59415396
What?

>>59415224
>>>59415203
>cache gets dumped into RAM due to OS doing something wrong, apparently scheduling is only small part of it, there is something more
Then until AMD figures out what the fuck is going on, then I would wait until the next cycle.

Plus buying day 1/v1 hardware for a new chipset is just sad times and something I have sworn off.

My i7 7700k at 4.9GHz doesn't have this problem.

>>59415528
>his cuck 4c cpu doesnt even 5GHz

Like seriously how can AMD beat the G4560?

>>59415528
Yeah, I'm sure it's the only reason you survived the winter.

>>59415360
Intel has done the same thing for years, nobody bats an eye.

AMD does the same, and everybody loses their minds!

I'm don't know if O should get the 1400 or the 1600
1600 will be quite more future proofed, but I can replace the 1400 in the future

>>59415573
Intel has stretched their userbases asses so wide they don't even notice when they are getting fucked anymore. It's not surprising.

>>59415549
It's at 1.27v and I only have a shitty air cooler :^)

>>59415626
>lying on the internet
sad!

>>59415646
Shills are given top silicon.

>>59413512
Why X version?

>>59415646
It needs 1.33v for 5GHz, though. My air cooler can't keep up at this voltage (reaches 90c in stress tests). I'll probably upgrade to water soon.

>>59415704
binning

>>59415360
Both Intel and AMD have been doing that literally forever.

>>59415738
So it will oc 100 mhz higher? You think its worth it?

>>59413621
The 4c probably won't have any issues with legacy software since it seems a whole CCX will be disabled
>>59413674
We already knew the sweet spots frequency wise for Ryzen, they are already quite outside them at 4 GHz
>>59413936
Unless they're quite bad odds is that the lowest end will OC as well as the R7's
So there's no point in getting anything aside from the 1400 and the 1600 if you can OC

>>59415573
>buy Intel non-k cpu
>different price points for different clock speeds
>incentive to pay extra for extra clock speeds

>buy ryzen
>different price points for different clock speeds despite being unlocked
>buy low end model and overclock it to surpass higher end model
>no incentive to pay more for miniscule clock speed differences which can be made up via overclocking
>no extra bonus features of the higher end models with the clock speed boosts

It's almost like amd want to lose sales. At least limit ram speeds or some shit on the lower end models so there is an actual inventive to buy a higher end model. It might be jew-y but it's necessary for business. Segmenting exists in business for a reason.

>>59415775
We don't know if the 1600 can even reach 4.0GHz, so I don't know if it would be worth it. Wait for reviews.

>>59415801
>buy low end model and overclock it to surpass higher end model

You forgot

>with a much higher voltage, usually above safe limits

Buy at least an X version if you care about overclocking.

>>59415801
>>incentive to pay extra for extra clock speeds

>being cucked for clocks speeds
>"incentive"

>>59415708
I know ivy bridge and kaby lake aren't equal but surely you shouldn't base real world temps on avx or linpack or prime?

my 3570k at 1.405v touches mid 90s in stress but stats below 78°C in Vegas or games where all cores are saturated like in bf1

>>59415844
Literally every review I've watched has shown the 1700 match or surpass the clocks of the 1800x with voltage that doesn't exceed the limits amd have recommended.

>>59415801
It's almost as if AMD wants to make a product that's good value.
Shame on them!

>>59414861
Bollocks. Bacon can be cut into tiniest strips, while sausage shape prevents good grill

>>59415887
Every review rants to sell you one
When they review Intel they want you to buy Intel and vice versa.
They make those reviews for people who are on a fence about buying certain stuff which they will probably buy after seeing review with referral link
Do you remember last time when you saw a review of some new PC hardware which would bash said hardware?

File: C6hGF7bWgAAmpVG.jpg:large.jpg (46KB, 601x360px) Image search: [Google]

46KB, 601x360px

>>59415887
AMD may be sending selected samples to reviewers. The lineup is purely based on binning, the top parts being the best ones.

>>59415897
>b-b-but Muh consumer value

This narrative doesn't last. We should be pushing amd to do what is necessary to regain marketshare as a whole rather than satisfy a small niche group of enthusiasts.

Just look at rx 480 and it's "consumer value". That narrative lasted literally 2 months and then the 1060 massively outsold it. Amd actually lost marketshare last quarter to nvidia.

Amd need marketshare and need it quickly for anything good to happen in the future. This ryzen hype train will die soon enough and everyone will go back to buying Intel soon after. I've seen this many times before. They can only prevent this if they mimick Intel by going full jew and locking the consumer down.

It's just business. My points might sound far fetched but just wait, it will happen.

>>59416006
>Just look at rx 480 and it's "consumer value". That narrative lasted literally 2 months and then the 1060 massively outsold it. Amd actually lost marketshare last quarter to nvidia.
Its almost as if one of these companies has a massive marketing budget. Nevermind the drones who buy nvidia "because it's the best" despite benchmarks showing the 480 beating the 1060. As with intel nvidias top end beats AMD and everyone jumps to the conclusion nivida is just better all around.

>>59415708
What cooler do you have?

>>59416059
AMD begin marketing retarder.
People think mmmhhh I like 1080 but is expensive,maybe 1070 mmmhhh... well 1060.

AMD don't had high end card is full retarder.

>>59416059
Tbf the 480 was only good value in America. I see on here that alot of Americans bought it but everywhere else nvidia seems to have good pricing on 1060s where amd might be slightly more expensive and I see a lot of those non Americans buy nvidia.

File: Intel_inside.jpg (250KB, 1032x774px) Image search: [Google]

250KB, 1032x774px

>I get 90°C on my 7700k
lmao

Nvidia only won on the basis of DX11 games being better 'out of the box'. Now DX12 and Vulkan are around it's going to swing the other way.

>>59415227
>Ryzen has ridiculous L3 latency
completely untrue as a blanket statement.
its latency within a CCX is HALF that of intel. between CCX it's 75% slower, so it's a tradeoff. it has more cache overall, lower latency intra-CCX cache but higher latency inter-CCX cache.

>>59416232
Infinity fabric? More like Infinity bottleneck.

>>59416229
The same was said about dx11 when amd were the first to support it. It won't last.

File: 15006224.png (164KB, 480x360px) Image search: [Google]

164KB, 480x360px

All of this doesn't mean shit until more am4 motherboards come out and are readily available

>Just keep wating!

>>59414468
Less cores that allow better single core performance and way better overclocking since the CPU will be less hot and use less energy.

>>59416389
Nobody gassed the kikes before Ryzen's release, now we have no AM4 mobos

>>59413253
>>59413259
>>59413267
>>59413274

hello old timer.
you slipped up and showed your age, it could become a problem in here though, but rest assured that some would actually be quite glad to have the voice of experience sharing anecdotes with us.

>>59416389
I believe more boards should be here in a month

That's usually how these things work, r-right

>>59414934
Even better for me.

>>59416232
>but Ryan Shrout said it's worse.
Good luck trying to put it all under context for intel retards who can't read too good.

Godspeed anon.

>>59413148

Amayzen price

>>59416006
>Just look at rx 480 and it's "consumer value". That narrative lasted literally 2 months and then the 1060 massively outsold it. Amd actually lost marketshare last quarter to nvidia.
The RX480 is still better value than the 1060 3GB, but good luck telling that to the 14 year olds who only use The Way It's Meant To Be Played™ gaymen gear

>>59416176
Nvidia is piss poor value where I live in Euorpe, the 4GB 1050 Ti costs as much as the 480 here.

>>59415234
Enthusiast = over clockers
Gamers = Gamers
Creators = Users of Auto CAD, Photoshop, Maya, Premier, Solidworks

Yea, so not just gamers.

>>59416598
Where do you live?

>>59416214
>at 5GHz
>on an air cooler
>throttle temperature is 100c
>meanwhile raisin at 3.9GHz needs 1.4v and runs at 65c+ (max allowed is 70c)
LEL

>>59416621
Sweden, the 480 costs €210 and the 1060 costs €240

>>59416652
You buy from Germany right? In Germany the 1060 is certainly cheaper than then hehe 480.

Just waiting for the benchmarks.
Once bitten, twice shy.

File: no bro.png (517KB, 1280x720px) Image search: [Google]

517KB, 1280x720px

>>59416652
>whining about a €30 difference
>GTX 1060 is worse because it loses in a few handpicked benchmarks here and there

>>59416687
Than the *

File: 1483612795662.jpg (129KB, 569x855px) Image search: [Google]

129KB, 569x855px

>>59413148
those are really good prices. r5 seems good

>>59416697
Look up what value means

>>59416697
as opposed to it marginally edges out in only a few other handpicked benchmarks here and there?

>>59416697
https://youtu.be/3Jml0uItdnE?t=1079

4GB 480 beats the 3GB 1060 and is literally 2 fewer FPS than the 6GB 1060 in an average of 19 games. You can stop posting now.

>>59416697
RX 480 is better value, both now and moving forward with DX12. GTX 1060 price premium is unjustified for what you get. It arguably should be even cheaper than the 480.

>>59413768
dude, they are autistic workaholics.

>>59416697
Price is one thing but experience is another. I know people who would pay that small amount extra for nvidia reliability.

For example my best friend bought a 290 2 years go and it was faulty out of the box (black screening, unable to hold factory overclock etc) so he returned it and was told there were no more 290s being stocked because of the incoming introduction of the 300 series but after some hard haggling they eventually stepped him up to a 290x for no extra cost and that was a great bargain till one of the 3 fans on that 290x stopped working (probably due to the fact it was most likely a returned 290x from a previous sale because the 290x's were also not being stocked) and he had to return it and they sent him a 390 as a replacement. This 390 was fine in terms of the actual gpu and hardware functioning well but this was around the time of the crimson driver release and those drivers were having issues on 3xx series gpu (maya cursor flickering, gta 5 crashes, bad pcsx2 performance etc) so my friend got fed up and bought a gtx 1070 as soon as it came out even though he would have been fine with the 390.

thanks to AMD I can now build a small form factor PC for couch gaming for about the same price as a console of shittier performance.

>>59414154
>350$ without cooler
>189$ with cooler
>similar performance in many games
sure kike

>>59416243
wait for 4000mhz motherboards

>>59416929
You could already do that years ago with an i3 and now with a G4560

Will 4/8 be one CCX or 2 CCX with deactivated physical cores ?

>>59417007
2 CCX with one disabled

>>59416850
You can't really go wrong with either. They're both within like £10 of each other and perform virtually identical (~2fps difference). It just comes down to whether you have a preference for each company or which cpu you're using. The 1060 would be better in a budget build with something like a pentium because of the more efficient drivers (source: digital foundry kaby lake pentium review).

>>59417016
Oh and my reasoning is that a 4c part only has 8MB of L3 .... the same amount as a single ccx

>>59416908
Still running an HD7870, still supported, never had issues in almost 5 years.

>>59417027
>7870

Literally why? That is the same gpu that's in the ps4 and is probably only good for 30 fps on low settings in most modern games.

I say that because my 980 a struggling to run maxed out 60 fps in some recent games like ghost recon wildlands beta.

You should upgrade.

>>59415844
I was in a thread on /v/ the other day with people arguing that they should not have overclocked their K i5 processors with the exact same clockspeed as the k model, and justified spending 50 dollars more on it with some asinine logical leaps.

I think AMD is doing the right thing. Non retards like me and you get great cheap processors, and retards like that can spend more for something they could do themselves because mental retardation.

>>59414726
>don't know if it's using 1 CCX or 2. If it's 2, it's prob. going to be the weakest of the entire Ryzen lineup.

1400/1500X are confirmed as 8MB L3, so it's most likely gonna be on just one CCX.

File: 1467825247503.gif (472KB, 640x360px) Image search: [Google]

472KB, 640x360px

>>59414418
why the fuck are these benchmarks so different than others where 7700k beats the shit out of everything? Is there one outlier game there that balances out the crappy performing ones?

>>59413235
>tumblr
>unironically posting worst DropOut

>>59417076
>I say this because my 980 is struggling
>Meanwhile my r9 390 doing just fine

It might surprise you Nvidiot but AMD actually supports their products so they don't get worse over time.

>>59413213
>i7 7700k 2 times as many threads and only 1.1 in a heavily multithreaded benchmark
I feel like AMD beat Intel in their own game.

>>59414837
>cooking bacon that fast

Terrible. Take a lap.

Bought myself a Mastercase 5 for my Ryzen build, it came with two decent 140mm fans. Was thinking of picking up another three 140mm fans to help keep my system cool.

Are these quality Noctua NF-A14-FLX's quality?

http://noctua.at/en/nf-a14-flx.html

File: value-gaming.png (7KB, 480x288px) Image search: [Google]

7KB, 480x288px

>>59417163
Because they're not actually that different.

File: 1234931504682.jpg (38KB, 600x451px) Image search: [Google]

38KB, 600x451px

>preordered x370 motherboard 3 weeks ago
>still hasn't been in stock
>return window on other untested parts about to pass

Fuuuuuuuuuuuuuuuuuuuuuuuuu

>>59413814
9 is not a prime number you dipshit.

>>59417199
Yes, but who the fuck would pay 25 bucks for a fan?

>>59417163
Ryzen is 25% faster than the 7700K in F1 2016.

>>59417190
Funny, because the 980 still gets 20+ fps more than the 390 in ghost recon.

File: Straight from reddit.jpg (191KB, 1024x679px) Image search: [Google]

191KB, 1024x679px

>>59417163
>Is there one outlier game there that balances out the crappy performing ones?
if by outlier, you mean "the common experience" and crappy performing, you mean "nearly identical".

Then yes, yes to all of that.
You can get basically the same (or better) experience on AMD's Ryzen as you can Intel's i7 kabylake. That'll probably hold true for the R5 line which is where the true value comes in.

You could get a 4c/8t for 190 with a $90 B350 board(i prefer the asrock ab350 pro4) and 8GB of DDR4 Ram for 40-50 dollars ($330 total)

An entire setup that'll perform just as well as the 7700k but for $10-$20 less.

>oh it's anotherf baseless hyping train

I assume thre R5 and R3 series CPUs with SMT will suffer the same as the 8 core CPUs and share the same bugs and problems.

Haven't people turned off a few cores their 1800X yet for testing?

Gaming performance will be even lower than the 1700.

If it was better than the Intel i5-7600K I doubt AMD wouldn't charge more.

We are talking about mainstream CPUs here.


I am holding off until 11th April.
After that I will either buy an i7-7700 or a Ryzen 1600X.

I will only go for the 1600X if it doesn't draw more power @4Ghz-6cores than the 7700.

Also gaming performance, which is my main focus, where I need the fps needs to be at the level of the 1700@3,7Ghz.
The i7 is good enough for photoshop programs and CAD software.


Otherwise it's just too inconvenient.

I don't buy on your wishfull low IQ thinking that soon™ it will smash the Intel CPUs.

>>59417246
>tfw some Australian retailer is selling them for $17 Aud

File: 1489452254881.jpg (603KB, 2000x1330px) Image search: [Google]

603KB, 2000x1330px

>>59417253
>cherrypicking: the post

>>59417163
they left cpu depending games like cs:go and quake 2 out, it's just rigged

>>59417302
>SMT

Will be patched.

>I doubt AMD wouldn't charge more

Why would they, when they can completely dominate the market with their pricing strategy?

It's really like everybody believes because Intel had a stupid strategy for the past 5+ years AMD has to follow it.

Intel can literally not compete in a price war right now because of their size. They cannot afford to sell at AMD numbers, but AMD can afford to sell at AMD numbers and make MORE money, so they are in a perfect position to compete with intel right now because Intel is a huge bloated piece of shit.

>>59417302
>R5 and R3 series CPUs with SMT will suffer the same as the 8 core CPUs

Not true because it's only a single CCX aahahahahahahAHAHAHAHAHAHHA *breathes in* HAHAHAHHAHAHA

>>59417320
>That's like 13 real money.

Oh then that's a great deal, go for it.

>>59416908
> I know people who would pay that small amount extra for nvidia reliability.

I don't say this very often but, in this case, I did stopped reading right there, indeed.
Maybe you work in marketing, legitimately and truthfully, and that's simply how you are used to writing, but goddamnit your post reeks of shilling.

>>59417338
You are a perfect example of somebody with wishful thinking and thinking it is somehow reflecting reality.

>Will be patched.

Bullshit without source.
There are still retards out there, not too few, who still believe that the Scheduler is not working on Windows 10 even though AMD themselves have examined it and have released a statement that it is working as intended.

>Why would they, when they can completely dominate the market with their pricing strategy?

Because they are not a fucking charity club as you make them to be.

>Intel can literally not compete in a price war right now because of their size. They cannot afford to sell at AMD numbers, but AMD can afford to sell at AMD numbers and make MORE money, so they are in a perfect position to compete with intel right now because Intel is a huge bloated piece of shit.

Holy shit, this statement doesn't make any sense on so many levels.

Will ignore such people and such posts alltogether in the future.
Not even worth replying to.

>>59417330
>discussion about 980 having issue in ghost recon beta
>b-b-but cherrypicking

are you stupid or what?

>>59417415
Ad-hominem

>>59416908
The only graphics card I've ever had that broke was a 9800 gt. With computer parts, sometimes things break. It doesn't matter if it's Nvidia or AMD.

>>59417397
you sure proved him wrong with all these points of refutation to prove that this pretty common case of faulty hardware and factual proof of crimson drivers causing issues on r9 390 and 380x were all lies

...oh wait

File: 1416658885645.png (15KB, 300x300px) Image search: [Google]

15KB, 300x300px

>>59417424

>>59417411
>>59417302
>"bullshit without source"
>makes tons of claims with no sources
You are a perfect example of a pessimist with worst-case thinking and thinking it somehow reflects reality

>>59413213
>1.5x the cores
>.89x the clock speed
>1.69x the performance
I'd say they did pretty damn good.

>>59417451
I proved nothing. I commented on your obnoxious way of expressing your points. It was a straight up ad hominem and nothing else. I really don't know wtf your main argument was and I really don't care.
But chances are that you are on the clock right now.

>>59417431
the only graphics card that has broken for me was an x1950 pro. i guess i can say i've had some good fortune when it comes to hardware. going by /pcbg/ and gpu threads on /v/ it seems a lot of people have faulty hardware. you'd think in 2017 that manufacturing technologies would have greatly reduced this though.

File: vivaldi_2017-03-14_23-36-54.png (24KB, 680x297px) Image search: [Google]

24KB, 680x297px

>>59413246
>Gaymur here, is there any reason I should upgrade my i5 4460 any time soon
Yes. Many games now days get up to a 40% performance improvement with hyperthreading, clock-for-clock

4c/4t are EoL and should be replaced.

>>59413237
>For games obviously the only choices remain G4560 and i5-7600K.
You miswrote G4560 and i7-7700k

7600k can't hit 144fps except for a few games now days. Even the 7700k struggles to do that in many newer games.

>>59417516
are you retarded? were not even the same person you spastic. there is nothing wrong with posting an experience youve had. people do it all the time in other threads. just because its a single case scenario which happens to go against your beloved company doesnt mean its a shill post.

get the fuck out.

>>59417411
>This statement doesn't make any sense

Intel is a bigger company than AMD. Intel operates by selling processors at a certain cost. If intel cut the cost of said processors by over half, they would no longer make that money, even if they dominated the market, and could not continue to operate as they are. As it stands now, even if they don't change the prices, if they still have 50% of the market, they would still make more money than if they entered a price war with AMD.

It's simple mathematics.

>>59417562
so, you're defending the fellow in the cubicle right accross yours. Gotcha. You're a good friend.

>>59413162
Not really. 99% of people buying mid range desktop CPUs are gamers, not video editors. The i5 still has better single threaded performance.

>>59414123
>Cemu

Nice meme there

>>59417586
let me guess, if it was a post about bad experiences with intel, nvidia, microsoft etc it would be an accurate representation of their product, right? but because it's amd its all a lie amirite?

i'm starting to think you're the shill here. grow the fuck up. your corporation doesn't care about you.

>>59417549
My i5 6500 can hit 144 fps in all the games that I own that can hit 144 fps with my GTX 1070...

>>59415844
>buy Intel
>different price points for different clock speeds despite being same CPU
SKYLAKE TO KABYLAKE

>>59417411
Simply because it's not considered to be a bug it doesn't mean that future optimizations aren't possible.

Since this a new uarch entirey, even though it came to market working from the get go, which is good, the possibility of it having further room for improvement is greater than it having already been squeezed out of every tiny bit of performance.

>>59413148
any europoor prices leaked yet?

>>59417848
Yes.

>1600X 429€
>1600 399€
>1500X 379€
>1400 359€

Though you can get them 5€ cheaper by sending "I LUV AMD" to 444-521.

>>59417906
i bet the uk prices will be all fucked up. eh, i'll probably pick up a 6600k in that case.

>>59418004
Don't make me feel bad anon, that was a joke.

>>59417201
I thought 5960x was much better
apparently it sucks according to recent intel shilling

>>59417335
>quake 2
>a game from 1997
I'm not sure whether or not you are being autistic on purpose

>>59413177
>1500x looks like a viable CPU for a decent midranger.

And even the 1400 indeed. Priced at pretty much the same as locked, 4c/4t i5 6400s and 6500s that run a touch slower.

Expect every system builder ad you see in magazines to be putting ryzen chips in their mid range builds from May onwards.

>>59417906
kike c'mon

>>59417625
so can ryzen in those games
it can't hit them where 7700K either can't hit them difference is 10 fps = 0.4ms

>>59418052
https://www.youtube.com/watch?v=oMNFoNtKCR8

you laugh, but people took this video seriously

>>59417216
>Buying your parts before everything is available to buy
retard

>>59418196
i don't get this either, 1700 sold out here, but no boards
I can't keep CPU laying around, I want to run it

File: pepepepe.jpg (12KB, 251x242px) Image search: [Google]

12KB, 251x242px

>>59417216

>>59417594
and 99% of people dont bother overclocking their cpus either

>>59413148
$219 for the 1600 Ok there it goes I need to save a little more

>>59414130
Why the fuck is the emulator for a 3 core system instead single threaded?

People are giving that fuckhead money when he's so bad at programming that he turns multithreaded games into single threaded ones?

>>59417020
I wanted to link punctures from that article myself. I think the fact that 1060 beats amd in combo with pentium shows potential for other untested scenarios. I t might be better somewhere else too

>>59418194
>Someone took that seriously
Forget the kikes
The retards are who need to be removed

As a sand nigger, I have always been very aware of Intels kikery and their factories built on stolen land, but I have still never been the AMD fanboy as it was hard to deny that Intel simply offered better products in the last decade I could never bring myself to buy their chips unless I really had to upgrade. I went from core2duo to Ivy Bridge which I have been on since.

It feels fucking good to say that AMD is a serious contender again. The1600x chip looks really promising, competes with the 5820k on performance and the i5 on price.

The only thing Intel have going for them currently is the single core performance of the k series, which is a fanboy meme the perfect example of marketing cuckery. Dont know why people even bother.

Fuck you Intel. I am enterprise IT architect, and I will remove as many Intel chips as I possibly can from the all the DC I work on.

>>59414640
i'm getting a 1700, but point still stands, 4 threads won't be obsolete till they stop selling consumer 4 threads, and all/most current 4 thread cpus are out of service, which is at least 7 years for now.

>>59415129
last pcie test i seen was 8x was a 3% loss, 4x was a 7 or 8% loss

pcie2 4x was a 12-16% loss

you use next to no bandwidth the gpu, the only reason they take up an entire slot instead of a 4x is weight bearing.

>>59417076
7870 is as good as the 1050 and rx 460.

Zen plus when :3 ?

>>59419286
2020

>>59415100

All of that sweet cache. Why has no one talked about this yet?

1600x and 1600 both have the full 16MB of cache allocated for less cores. This will bridge the gap between IPC, no?

Or is it because the cache is split between CCX, 8mb on each side, that I am dead wrong?

>>59418194

those comments are great

>you're hired- Gamers Nexus
>It felt smoother on AMD

owwwwww my sides