>every faggot and their dog cries at Intel that IPC has barely

It's on par with and even better in IPC than Kaby Lake, go ride on Intel's dick elsewhere.

>>59091035
>implying intel can even beat sandy bridge

AMD has made a 53%[1] improvement from piledriver

[1] https://www.youtube.com/watch?v=1v44wWAOHn8

>>59091035
(You)

File: 1487801576360.png (234KB, 800x612px) Image search: [Google]

234KB, 800x612px

>>59091035

>>59091035
Poor quality bait. You know better than anyone Ryzen is at the very least on Skylake IPC.

>>59091047
Now, I'm not biased, and I go for whichever is better, but from all the pre-release benchmarks I've seen, your statement is an outright lie and you should be ashamed of speaking on something that you have no knowledge of.
Shame on you.
Shame.

>>59091047
>Intel has no competition for several years
>Focuses on energy savings over IPC
>Competition on IPC arrives
>Drop prices by $50 across the board
>Wow nothing happened
gg amd

>>59091073
What a man!

>>59091105
Fuck off.

>>59091073
>this shit cpu is better than that shit cpu
Incredible.

>>59091131
Post one of the benchmarks

I believe you're not knowledgeable enough about CPUs to interpret the results kekols

>>59091035
leaked review that got downloaded before embargo :
https://www.youtube.com/watch?v=YWEHs_R5t9s

>>59091131
Do some more research, there are a lot of sources saying it's on par with at least Skylake.

>>59091135
They'll need to drop by more than $50 for the price to performance to be at Ryzen's level.

>>59091153
whats a good cpu then?

>59091153 (You)

That would put the IPC in skylake range

File: 1487801576360.png (245KB, 800x612px) Image search: [Google]

245KB, 800x612px

>>59091148

File: intel-post-ryzen.png (246KB, 800x612px) Image search: [Google]

246KB, 800x612px

>>59091035
>Being this much of an intel fanboy

smells like poo in here

I don't care what the benchmarks say, I could not sleep at night with an amd cpu in my rig

The FX would have been good if they didn't have single core performance worse than a Pentium, only way Ryzen is going to be competitive is if they managed to address that.

File: 1487909176qq861.png (245KB, 800x612px) Image search: [Google]

245KB, 800x612px

>>59091035
>>59091148

File: 1487814827640.png (495KB, 1070x601px) Image search: [Google]

495KB, 1070x601px

>>59091337
>I don't care what the benchmarks say, I could not sleep at night with an amd cpu in my rig

pls b bait

>>59091035
>>AMD drop a new processor with IPC to match Sandy Bridge, a six year old architecture
you missed the mark by about 5 generations.

>>59091105
>>59091220
>>59091374
pathetic

I'm going to Ryzen and AMD when I upgrade because I'm still super furious about how fucking shitty Intel's Z77 board is. I can't change C-states on an OVERCLOCKING board. Why did I buy this trash...

>>59091337
Intel fan boy here. It's been funny watching AMD crash and burn.
We seriously can't let AMD get the codes.

Not gonna lie former #TeamRed comrade here but this is fucking hilarious watching AMD crash and burn. In all seriousness though we can't let these goyim undercut our market!

>>59091393
Got a chuckle from me actually

File: 35jw9pg.jpg (24KB, 640x613px) Image search: [Google]

24KB, 640x613px

The big question is how will Ryzen age? And did AMD blow it's whole load at once and will now ride this architecture for the next 5+ years right back into oblivion while Intel is building a new 7nm process fab in Arizona?

So Intel and AMD are on par now. They're BOTH at Sandy Bridge IPC now, you since there's been no actual; IPC* improvement.

*SYSMark 2014 doesn't count.

File: wryyyyyy.png (245KB, 800x612px) Image search: [Google]

245KB, 800x612px

>>59091393

>>59091433
The sucessor will be Zen+ which is just a die shrink on a 10nm node, sometime in 2018 IIRC. They however haven't revealed later plans yet

>>59091433
AMD is going to absolutely obliterate the market once their new modular RAM-less APU designs arrive in 2022
You missed the boat to invest while they were at $1 a share but it's not too late yet.

why is amd releasing the rx580 and not a rx490?

>>59091473
Thanks Raja

Former AMD fan here. It's been hilarious watching them crash and burn. Glad I'm on the Intel side now. #teamblue

>>59091386
Who's holding the gun?

Slightly better performance, dubious benchmarks, suspiciously cheap... whatever.
The point is that Intel is a respectable household brand name, and if it costs me a few extra bucks to sleep soundly at night and know that I got my family and myself a quality product from good hands than so be it.

>>59091499
Kek

>>59091587
This.

If you buy AMD like a retard youll never sleep once you realize at any moment when youre sleeping the PC could just erupt into flames. Not to mention you wont sleep from the massive headache everyone in the family will give you for buying a slow 2010 tier processor that cant do anything.

This is why I just recently upgraded to a 4690 from my old 3790, its really a night and day difference compared to AMD.

File: intelvsryzen.png (263KB, 800x600px) Image search: [Google]

263KB, 800x600px

>>59091393

>>59091669
wtf is this autism

I swear computer normie memes are the worst garbage in existence

>>59091464
7nm.

>>59091162

that's insane being the lowest-end ryzen chip

File: dec0ae0d35adf54881a35b0981a223f0.png (1MB, 999x1199px)

1MB, 999x1199px

STOP BUYING INTEL AND START PRE ORDERING RYZEN NOW

Man Intel must be spending a fortuane in shilling. There are three threads on /g/ that are anti-Ryzen.

File: 1487825833939.png (1MB, 700x960px) Image search: [Google]

1MB, 700x960px

>>59091666


nice trips, but TDP is lower than comparable intel chips....

>>59091135
Ryzen it's outdoing Intel on power consumption as well senpai

apparently they overlock like shit. i'm waiting for intels next 6 core with soldered ihs before i upgrade. we'll see what's what with zen+ vs whatever intel comes up with next. a 6 core soldered kaby lake at 5ghz would be sexy.

>>59091464
Zen+ isn't a die shrink, it's Zen refined, with lot's of shit that didn't make Zen due to area and time problems
Zen+ will have AVX512 according to the Zen lead engineer, most improvements will be FP-wise iirc
It's even possible that Zen+ might be made on a refined 14 nm process, and not on 7nm
>>59091433
They promised a 10+% increase with Zen+, with Intel literally having nothing to go against this they have guaranteed dominance
And then come the APU's with Vega, and the Exascale HPC APU's that will literally rape everything in existance
The fist 1000 PF computer will run on AMD APU's, mark my words
>>59091909
The 4.8 Ghz 4c8t is already a fucking housefire according to it's TDP, and no, a soldered IHS doesn't lower TDP
You're asking to burn not just your house, but the entire neighborhood, with a 5 Ghz 6c Kaby Lake

>>59091176
it doesn't exist because everyone is retarded except me.

File: question.jpg (80KB, 1024x768px) Image search: [Google]

80KB, 1024x768px

Will the chipset have the same or similar features as Z270?

File: Falcon_den.gif (952KB, 300x168px)

952KB, 300x168px

>mfw my 3570k @ 5ghz is still perfect and will still be for years and you are all cucks and consumerist retards crying over brands and wasting mom bucks

When are the 6 core versions coming out? I haven't seen anything except pricing for those.

>need ln2 to get it to barely overclock to just 5ghz
Holy shit that's horrible...

>>59091718
at least they have memes. Shills like you dont even bother creating OC.

>>59092319
"Q2"

Really disappointing

>doubting Jim Keller the Intel Feller
rip in shit Intel

File: reallymakesmethink.png (632KB, 1389x1527px) Image search: [Google]

632KB, 1389x1527px

>>59092363
>Shills like you dont even bother creating OC.

think again

here's some OC I made :^)

>>59092416
a compilation of memes is barely OC, but good job shill.

why do you think these are base 3.6ghz and 4.1ghz boost,gonna laugh my ass off when they barely reach 4.4 there's a reason barely anybody is showing temps and voltages

>>59092539
It's doing 95W TDP at base 3.6 and 4.1 boost you stupid shill, this is way better than Intel needing 140W to do the same thing

>>59091525
Another intel shill fag.

File: 5985698656.png (23KB, 500x294px) Image search: [Google]

23KB, 500x294px

>>59092597
because it has literally half the pcie lanes,literally useless for massive workloads,not a shill btw just some random guy on /g/ late at night

>>59092706
You literally cannot saturate those PCIe lanes with just the CPU stupid fuck, if you need more you wouldn't be looking at the HEDT parts
But at Naples, with has up to 4 times as many lanes, and that's just from the CPU
Fucking shill
Also, Intel's massive TDP doesn't come from the uncore but from their retardedly big FPU, lrn something for once stupid niggger

>>59091035
>intel has been using the same architecture of sandy bridge for a half a decade.
>being surprised people are hyped when Amd make on a new architecture that equals sandy bridge revision 7.

>>59093064
>intel has been using the same architecture of sandy bridge for a half a decade.
It actually dates back to the Pentium Pro, in 1993, with the short exception that was Netburst
Shit's still P6

>>59092706

24 lanes vs 16.

Plenty for a desktop.

And if you have a "massive workload" that still isn't being met by a desktop part, then why are you cheaping out on a cutrate competitor that's shooting above it's price class instead of going after the heavyweight champ to do your workload?

>>59092706
>>59093125
I'm talking about a situation where your gonna have ssds in raid and cards in sli or crossfire for 4k rendering with a quadro or two so 24 lanes is gonna saturate quickly reason I'm comparing is because it's being compared to 6900 which is better suited for that

>>59091035
Seriously this shit is pathetic

AMD can't even beat Intel's last gen processors

File: blow-your-freaking-mind.png (818KB, 627x741px) Image search: [Google]

818KB, 627x741px

>Intel had no competition for 6yrs
>Focus on shitting energy savings on mobile proc
>Focus on growing iGPU business
>Decent laptops have discrete GPUs
>Nobody wants a netbook with i5/i7
>Nobody likes i3 performance
>AMD awakens and analrapes i5/i7 line
>Intel has to resort to OEM bribing
>AMD revamps mobile procs and server procs.
>Intel shills scrambling for FUD talking points
>2020 Jim Keller inaugurated as the 46th President of the U.S.
>2020 - week 2: Keller, via executive order, begins the mass conviction of all gypsies, semites and non-Japanese asians.
>Intel stock is delisted from the NASDAQ, and top execs are jailed.
>Aside from a few criminal shills, most everyone has replaced their tick-tock toys with government mandated-minimum 1600 or greater.
>In his 4th presidential term, Keller has optimized the fabrication process to 1nm and all citizens have Ryzens embedded under the skin.
>Singularity has been achieved.

gg Intel

>>59093181
We can't let AMD get the nuclear codes.

>>59091162
3ghz 65 watt CPU gets beat by 4.2+ ghz 91 watt CPU in a game with dogshit multithreading.
WHO WOULD HAVE GUESSED?

DELET THIS

[spoiler]just came in here to say this meme[/spoiler]

File: 1487452561666.jpg (26KB, 192x171px) Image search: [Google]

26KB, 192x171px

kikes in full damage control

File: From vid.png (64KB, 499x498px) Image search: [Google]

64KB, 499x498px

>>59091162
>>59091764
>>59095848
Looks like a 8XXX sitting in the socket tho.

>>59091433
>Intel is building a new 7nm process fab in Arizona?
Feels good to be Arizonan desu. We gave the world Terry Davis and now this.

>>59091718
>wtf is this autism
Probably /r/ayymd.

>>59092121
>3570K
>5GHz
Probably 3 degrees away from a literal housefire, if that ever goes beyond POST. Stop lying on a nepalese foot fetish imageboard, kid.

File: 15125.png (314KB, 616x643px) Image search: [Google]

314KB, 616x643px

but but my popular vote...

>tfw going to upgrade from my now ageing i5 4670k and a GTX 980 to Vega and Ryzen

feels good rooting for the underdogs

>>59096188
But pretty soon Intel will be the underdogs desu.

>>59091073
>AMD has made a 53% improvement from piledriver

We've had no real reviews of the chips yet people are pre-ordering like there's no tomorrow.

Not saying it's going to be bad but people will be disappointed with how they actually perform compared to Intels. It's going to be Bulldozer all over again, so much hype before release, with record high pre-orders then review time comes and buyers remorse sets in.

>>59096188
4670k is more than capable for gaming still
>>59096202
wrong!

>>59092416
i like it, keep it up.

>>59091415
How many shekels did they promise you Shalom?

>>59091035
YOKES on you fagot, even ancient software runs on at least 2 threads if youre using windows vista or superior.
So just IPC alone isnt going to make a huge difference.
You need, better l1, l2 and l3 conectivity (which ryzen has), better branch predictors (again, ryzen has), a more streamed pipe line (ryzen again), lower misprediction penalty (ryzen) etc etc.

So yeah, currently AMD is ahead wheter you like it or not.
I would assume intel has some neat trick up in their sleeve, the predictors were always better but then again maybe they just squandered their shekels in SJW programs like diversity shit and all...

OP is quite upset lol

>>59091035
>IPC
Why do you retards keep using words you don't understand?

>>59091047
Source?

>>59091035
Op has a point here

>>59091035
tech enthusiasts tend to be more invested in AMD's success because it's more meaningful than anything Intel can do at this point.

literally every Kaby Lake processor is on sale. Intel is going back to the drawing board after ignoring the x86 market for half a decade. they're probably going to have no choice but to revise their ultra-fragmentation pricing strategy.

Intel winning means stagnation. AMD winning means both companies get better.

File: 1485470510178.png (380KB, 826x958px) Image search: [Google]

380KB, 826x958px

>>59092070
pls build me a cpu, anon

where the fuck are the R5's ?

>>59096237
I don't believe that to be the case.
Although I've never been in preference of AMD and always picked intel because "it just werks", I say this is AMD's last shot at a good processor or their company might end up pretty bad (so they're giving it their all).
We just have to wait and see.

>>59096188
Same Specs here buddy, ordered the 1800x

>>59091135
>ryzen is 95w
>comparable intel is 140w

really makes me think

>>59091047
>even better in IPC

Again, people on the internet talking out of their stinking mouth, spreading bullshit.

>>59092121
>5ghz
And here I am with my 4.3ghz and Evo 212
I'm good tho, would rather not have a house fire, not worth it.

File: IMG_2044.jpg (46KB, 1212x300px) Image search: [Google]

46KB, 1212x300px

Keep crying, Shlomo

I cannot wait for the reviews to see the buttblasted people doing damage control.

The reviewers feel like beta testers because everything is not matured at all, which is one of the reason why they release the mainstream Ryzen 5 and Ryzen 3 series.

Early adopters and especially those who pre-order hardware without any reviews, are braindead people.

>>59098799
maximum overclocked $390 cpu beats non overclockked $330 cpu

holy fuck!

>>59091162
funny you should mention that
https://www.youtube.com/watch?v=2Fy87gUVisQ&feature=youtu.be

lets see how the 95w does

>>59098799

Stop quoting that lying donkey fucker, you gullible faggot.

https://wccftech.com/radeon-rx-480-gaming-performance-review-leaked

>AMD RX 480 Gaming Benchmarks Leaked, Rivals Fury & 980

>>59091084
/thread

Ahahaha, extremly picky with RAM confirmed by reviewer.

Enjoy your 300$ mainboard and 200$ 16GB to get the best out of your CPU.

Many people receiving their new Ryzen will be in for a nice surprise with RAM compatibility.

>>59098877
sauces?

>>59098877
>as it stands the amd code has restricted ram tuning options which means many ram kits at launch will not be compatible. This is the same for our competitors also.
What we recommend is the following:
If fully populating a system with 4 dimms (2dpc), use memory up to a max of 2400mhz.
If using 1dpc (2 dimms) ensure they are installed in a2/b2 and use memory up to max of 3200mhz.

The indication i have received from hq is that amd has focused all their efforts on cpu performance so far and will release updated code in 1~2 months when we expect improved ddr4 compatibility and performance."

Right from ASUS

Early adopters and pre-order people = fanboys to the point at which they enjoy getting shat on

>>59098877
the current top preforming board (for overclocking) that we know of (ROG Crosshair VI Hero) is 250, and most of the other X370 boards are well under 200 and not far off with overclocking anyway.

>>59098932
>250

Really depends on the country.

270 here

But I rest my case with it that is all good they are unlocked, but you will have to spend a lot of money getting a decent mainboard and ram.

I am sure it will be all better with Zen+.
Intel had the same issues, it takes some time for the codes to mature.

This on the other hand really does bother me now, because I want to build a system.

I guess I will go with an Intel now and switch to Zen+ if it's gonna be a drast improvement.

File: IMG_1726.jpg (52KB, 640x360px) Image search: [Google]

52KB, 640x360px

AMD is better for intellectual activities like multicore rendering, Intel is for brainlet gaymers

>>59098861
>Both CPUS are at stock speeds with stock coolers
>7700k
>stock cooler

It doesn't come with a cooler.

>>59099082
thats your problem.. neither does 1700x

>>59098993
yeah it's a shame, AMD is going to lose out in markets that tax them more/ cost more to ship to.

>>59099103
The video is bogus, then.

>>59092121
Whats it like having a CPU eject 200 F worth of fire?

>>59099119
the video is so legit as the other one afterall we do know that the x is faster than the 7700k

>>59098545
its 95w when overclocked ?

>>59099118
That DinoPC guy mentions that the 1700 is 40 pounds cheaper than the 7700K.

Again, not here and surely in many more countries.

1700 and the 7700K costs exactly the same here.

>>59099154
>we do know that the x is faster than the 7700k

No we don't until reviews next week.

>>59099223
i highly doubt we will see the x floating around on benches..

mark my word for it

>>59098877
>Enjoy your 300$ mainboard and 200$ 16GB to get the best out of your CPU.

Mobos are already out, dont try to project just because 2011-3 mobos cost that much.

>2400mhz ram is sooo rare!

>>59099118
>AMD is going to lose out in markets that tax them more/ cost more to ship to.

Not really, the markup is the same for AMD and intel. Ryzen is just a bit more marked up in EU and others because it's new. All in all the markup isnt bad at all.

Only 70€ markup on the 1800x, not bad. Would have expected 100-120€, that's 20-25%.

rx480 8gb is still 300€ here

File: RAM.png (25KB, 1281x593px) Image search: [Google]

25KB, 1281x593px

>1866 MHz only if all RAM slots are filled
Defend this.

>>59099267
you are missing the point. you can run intel cpu's great on budget parts. Why is someone going to get one of the r5's when they can just build with intel for cheaper and less problems.

like any of these new AMD cpu's are irrelevant if they can be sold at a better price than the Intel option.

>>59098861
no actual photos of setups, just random non matching gameplay. fuck right off.

>>59098784
4.5GHz here on 212. Shits good.

I wanted to upgrade cooling but the more and more I research the more and more watercooling looks like a meme. Problem is I don't want to hang a shit ton of metal off my motherboard. The 212 is already kind of heavy.

>>59099181
is intel overclocked?

File: 1468532343132.png (70KB, 400x275px) Image search: [Google]

70KB, 400x275px

>AMD
>Company people know only for making cheap, shitty budget parts for poorfag gamers
>Suddenly makes processor nobody can afford and needs crazy expensive mobo and RAM

It's like when the autistic kid at school puts on shades because he think he'll look cooler.

>>59099414
I am asking. I don't OC my stuff. Does the watts increase if you overclock ?

>>59099392
>>59098784
>>59092121
So, we are not upgrading from the 3570k?
Guess I'll just get a 1070 and be set

>>59099445
the cheaper little brothers are coming out later this year.

>>59091035
It's more that all of them are unlocked, they start at 4c/4t, and end up at 8c/16t for half the price of what jewtel wants to charge

>>59099445
>>Company people know only for making cheap, shitty budget parts for poorfag gamers
you need to be 18

>>59099445
And yet it's STILL cheaper than equivalent intel.

>>59099363
Lot of cheap mobos are already announced. Im saying you have no proof of there being anything wrong with them.
And the r5 series is already outperforming everything on the 1511 socket for less.

File: 1398081830279.jpg (77KB, 1000x635px) Image search: [Google]

77KB, 1000x635px

>>59099492
I don't know about you, but for me there's no need. At 4.5GHz the 3570K crunches everything I throw at it. I just want it to run cooler and quieter and WC sounds neat. The bottleneck for gayman is my graphics card. For everything else it would indirectly be my generation/platform, i.e no NVMe, certain USB ports starting to malfunction, VRM heatsink probably needs to be replaced etc.

Upgrading now is painful as fuck because it automatically means a motherboard and RAM upgrade.

>>59099661
Since I need a GPU upgrade also I'm basically needing to build a whole new PC.
Fuck that noise. I'll get a 1070 soon-ish and wait it out until next year to see what's up and not just jump on the hype trains.

Yes the power drawn and given off will increase as you increase voltage and clock speed.
V=IR

>>59098841
The point is that Ryzen beats Intel in IPC clock for clock.

>>59099661
>What are Ryzen 5 and Ryzen 3

>>59099661
2500k here, I'm mostly the same. There's nothing wrong with it, runs games good enough but I want to multitask harder and even do more demanding tasks. I still have a gtx760, I will upgrate to ryzen before I upgrade GPU. Modern games are so shit.

The new PC thing is what exites me the most almost, I'll have a powerful home server.

>>59099623
AMD cannot beat Intel in a price war, they don't have the financial capital to do it. AMD is banking on preorders to stay afloat. I guarantee by Christmas 2017 Intel will have better value builds compared to every Ryzen offering.

>>59091587
> suspiciously cheap

This is what intel fags actually believe!

>>59099858
Of course they could, but they probably don't want to, for a few reasons:

- Intel would look like jews if they suddenly halved the price on their CPUs. This would reveal them too much.

- There's just nothing that competes with r5 and r7 on 1511. The 2011-3 mobos cost so much, even if they lowered the price of the 2011-3 cpus to 1/3-half the mobos would still be $150 more.

- Intel would actually have to sell at a loss, their giant IGPUs cost them too much. Maybe if their next line had no GPU, but that's not the case.

File: 1374525541335.jpg (210KB, 600x600px) Image search: [Google]

210KB, 600x600px

>>59099789
>Upgrading now is painful as fuck because it automatically means a motherboard and RAM upgrade.

This applies regardless of whether you go AMD or Intel. There's no need for me to upgrade to anything regardless of its value proposition. Plus, next year there'll be something better on the market (Zen+, apparently).

>>59099846
I also get that "omg build new PC NAO" feeling, but when I look at actual cost its ludicrous and unnecessary. I'd rather put away some money every month in an inflation-beating notice deposit account and build a fuck fancy system guilt-free at the end of it, instead of trying to perfectly budget everything to the last cent.

IDK about you tho because you want more multitask performance, which I assume means more cores/threads.

>>59100069
you only need to upgrade RAM and motherboards once every 6 years with AMD though

>>59100069
>. I'd rather put away some money every month in an inflation-beating notice deposit account and build a fuck fancy system guilt-free at the end of it, instead of trying to perfectly budget everything to the last cent.

Yeah that makes sense, but I have a lot of money at the moment. I'll probably go with the 1700x, because it looks like it's 120€ more for 200mhz.

Looks like I'll be spending 1-1.1k€ with no gpu. I could go up to, say 2k no problem it's just about what seems price efficient.

>>59100092
I hope they replace it earlier this time.
DDR4 would be horribly slow in 2020.

A ram change a year keeps the goyim in fear

>>59099661
>Intel poorfag crybaby needs to justify his purchase
Fucking kek what a sudden turn

>>59100349
I built this thing in 2012. I would say my purchase is damn well justified. There is no need to upgrade. All I do is college and Overwatch, and it does those two things really well.

>>59100594
replace sonic with tails and your meme is PERFECT

>>59091035
Poor intel, they're really desperated.

>>59093171

Your hypothetical is still full of faggotry.

Why are you using a cutrate desktop part instead of getting what you need?

Naples is coming in 32 core, and 16 core. And reportedly a MASSIVE 500mb L3$ that's shared.

Now stop arguing like a bitch and use some reality.

>>59091035
who cares what the product is, it's amazing to watch /g/ suck a big corporation's cock this much while simultaneously whining about their also big competitor, working themselves into a lather over the privilege of mindlessly consuming someone else's product to get those sweet frame numbers in children's games

/v/ migrants are the worst

If you scaled a Pentium 4 down to 14nm could it actually hit the clockspeeds it was supposed to?

>AMD does what Intel don't.
>BLAST PROCESSING

>>59095848
we weren't looking at the same benchmark.

intel has the higher highs and the higher minimum, but notice the average. that did not scale like you would expect, at all. this screams that the game is using 100% or close to of up to 4 threads, and background applications are fucking with the intel side, making it dip lower more often.

>>59097911
they pissed away several billion dollars on poor acquisitions and giving processors away for free.

intel has more money then they knew what to do with and pissed it away.

>>59099448
depends on where the sweet spot is. usually you get 100-300mhz for almost free, but then it goes up exponentially, amd seems to be close to that, and the 6900 doesn't go over 4.5ghz without some non consumer/enthusiast levels of cooling.

>>59101880
Broadwell-E doesn't go over 4.4, period.
Haswell-E overclocked better, at least had a higher average of SKUs hitting higher clocks.

Israel is in full damage control mode and Babby Lake buyers having a meltdown with either buyers remorse or post-purchase rationalization. It literally can't get any better.

>>59091131
the issue is the only benchmarks we have are of amd at non equal clocks, of coarse kaby will get more shit done, but it looks like when you make clocks equal, amd beats them by a fair margin.

4 core are going to shit on intel hard if they don't have real ipc improvements, and 6 core most likely will beat out i7's

>>59101920
silicon lottery stopped stocking i7E's so I can't really tell you what they got it to clock to, but i'm sure they had a 4.5 sku

>>59101947
>my power level is ryzen

>>59101952
Must be the mother of all golden samples, I have no doubt Ryzen will have a few of those as well.
Question is who will get them.

>>59101978
amd is likely binning for servers, but we will know that when reviews come out and we see manual overclocks come out for the 1700 1700x and 1800x

>>59098799
stop using that fake graph and do reality check for you amd fagboi

>OCUK staffer “Gibbo” reports successfully overclocking a retail Ryzen 7 1700 processor with an ASUS Crosshair VI Hero motherboard to 4.05GHz on all 8 cores. With this overclock the CPU would effectively be running 450MHz higher than the Ryzen 7 1800X flagship, which is very impressive considering the huge price delta between the two chips as well as the low 65W TDP of the 1700. “Gibbo” estimates, based on his results, that the 1800X should be able to hit 4.3GHz on all 8 cores.

>We just tested a 1700, it hit 4.0GHz stable in everything, but ONLY in the Crosshair mainboard, the lower-end boards it was hovering around 3.80GHz as the VRM’s were cooking with extra voltage. It however was maxing around 4050MHz, so I’d say 1700 can do 3.9-4.1GHz, of course the 1800X will probably do 4.1-4.3 as no doubt better binned, but if your clocking the motherboard has a big impact on the overclock and so far Asus Crosshair and Asrock Taichi seem the best two.

>4.5ghz NEVER EVER

>>59102059
Why were VRMs cooking at 4.0 when VRMs aren't anywhere close to cooking with overclocked BDW-E and Piledriver? You know those things that sip more than 200W?

>>59102114
Could be low amp VRMs or the 1700 being really leaky, usually nothing good comes out of forcing low power ICs like the 1700 to guzzle voltage down like a hooker.

>>59102166
The opposite too, significantly undervolting some chips leads to some bizarre behavior.

File: 71UBB4uVZKL._SL1000_.jpg (162KB, 982x1000px) Image search: [Google]

162KB, 982x1000px

>>59101947

>>59102059
You do realise that most i7-6900Ks only clock to 4.2-4.3 GHz, right? Ryzen will almost certainly be identical.

The real question is how high the 4 and 6 core SKUs clock.

>>59102059
Thank god it is because of shitty mobos and not the cpu itself.
Those are 8 core CPUs, I believe that 4ghz is alreadly a lot, remember that 6900K is a 3.2ghz base 3.7ghz turbo you can overclock i to 4,3ghz sure, but at 1.4v it draws over 200W and goes close to 90°C operation.

I believe that the R5 will reach much higher clocks due to it having less cores and so less energy needed and better heat dissipation.
So, for me, reaching 4Ghz in a 8 core CPU with same IPC as intel for 1/3rd of the price.

Again, a 7700K is a FOUR CORE CPU, you can reach way higher clock on it but will draw more power and run way hotter than a Ryzen with lower clock gaining a negligible head on performance.

File: 1442542765279.jpg (65KB, 720x720px) Image search: [Google]

65KB, 720x720px

>>59102166
I wonder what voltage they were running at versus what the max voltage is for Ryzen.

Everything so far is suggesting that Ryzen is a poor overclocker. Not good.

>>59102367
>Thank god it is because of shitty mobos and not the cpu itself.
It's also because the 1700 is the lowest of the lowest binning.
The 1800x is the highest of the highest.
The max clock will certainly be much, much higher.
If the 1700 can reach 4 GHz, i can only wait to see how much can the 1800x reach

File: chuckle.gif (3MB, 286x258px) Image search: [Google]

3MB, 286x258px

>>59099661
>that pic
>mfw

>>59102431
It probably won't go over 4.3-4.4 without a golden sample, but still, I personally don't think it's worth the extra $200 for 300MHz, a 1700 and 4.0-4.1 will be great and most importantly, cheap.

File: letöltés (7).jpg (5KB, 249x202px) Image search: [Google]

5KB, 249x202px

>>59091035
>mfw when AMD is finally putting pressure on Intel
>mfw my i5 4690 is going to be low end

>>59102431
the 1800x maxes out 4.1 ghz with all cores.

The only chips that they've tested are the workstation chips. I mean how many people are going to be crunching 4k video all the time. Most of the people here are just going to be playing video games and we have no idea how this would fare with 8 cores and 16 threads due to optimization.

But realistically let's see how games handle the mid tier chips that people of /g/ will actually buy.

>>59102506
Neat, where is this info from?

>>59102431
Don't expect more than 4.2-4.3 GHz max overclock on 8 cores.

>>59102525
The AMD release event. Many people covered it but it should be on linustechtips.

>>59102523
More cores is absolutely fine for gaming.

https://i.4cdn.org/g/1487948983134.png

>>59102462
>is going to be
It was already low end a year ago

>>59102540
No overclocking info has been at the AMD event.

>>59102540
That was the max with XFR on stock amd wraith spire (95w) cooler, not manual overclocking nor max XFR reachable

What interests me is that new Ryzen overclocking tool, looks very detailed and specific, apparently capable of turning off cores that aren't in use so that would be real amazing for gaming.

>>59102594
It is an extension to the Wattman overclocking tool of the Radeon software.

>>59102594
Intel should have made something like this years ago.

>>59092416
>>59092427
So that chart looks like a meme to you?

>>59102620
Intel should have been bit less jewish considering the wads of cash they had and giving people some cheaper high end stuff but nah.

>>59102567
thank you for the explanation. I miss the old days of normal clock speeds and anything was just overclocking.

>>59102544

Even then you see a noticeable plateau after 6 cores.

>>59102523
It will be good for games that need coarz..... like flight sims

>intel just does a price drop on coffee lake
>AMD doesn't get anymore sales outside of preorders
>AMD eternally btfo for another decade

>>59096343
Thanks

I need to bolster it a bit more with intel and kek stuff I think

or completely remove the kek thing and make it a collage of bad things happening to intel and nvidia lately

>>59102782
Intel usually has a habit of price drops even when its getting its ass churned, yeah.

>>59102782
Intel will have to completely get rid of the the k cpus to compete

They're fucked

https://www.youtube.com/watch?v=W4sfQ6mxAPM

>>59102558
Fucking this, Intel made CPUs that should've been garbage years ago still cost hundreds used because of their ((((sales tactics))))

>>59102811
To keep OEMs from switching

but consumers will never see those prices

>>59091433
>Intel is building a new 7nm process fab in Arizona?
It won't be ready in time. Meanwhile, GloFo already has 7nm in the works.

>>59091047

how can you say that? ryzen isn't even released yet.

>>59102820
Nobody cares about K chips. Those are not significant for total sales.

>>59102949
Leaked benchmarks all match up.

>>59102968
you are an idiot shitload of people buys K chips

>>59102988
All is noise until we get proper reviews. I'm exited that the highest end Ryzen is only $500,but I'm still skeptical. How does it overclock? If cinebench single threaded scores were identical, then how worse is it really on single threaded programs? I'll probably be buying the 1700X maybe even the 1800X on Q2 if it delivers. So far, everything appears to be fantastic.

>>59102968
There will be no reason to even think about a non-k chip now that AMD gives you unlocked clock multiplier on everything

>>59103065
Cinnebench single core isn't anywhere close to real life serial code.
For starters real life serial code isn't 100% FP like cinnebench

>>59103202
Isn't zens integer performance even better??!?

>>59103202
>synthetic benchmark
Lmao
>synthetic single thread benchmark
Wahahahaa

People thinking CB == game performance are retarded.

>>59103222
Yes, you can see in the frontend design that AMD didn't have time to make the FPU larger.

>>59103247
>Didn't have time
More likely they cut it's size down to increase yield and decrease die area. How do you think it costs so little?

>>59093347
>implying Koreans aren't best azns

>>59103459
>implying gooks aren't completely incompetent and the ones to blame for the two biggest tech flops from the last year: the LG bootloops and the Galaxy Note housefires

>>59103283
No, it's because of time and complexity, a bigger FPU wouldn't even add 1.5mm^2 to each core.
And they're cheap because AMD isn't aiming at 60%+ margins like Intel.

File: 1473016597588.jpg (254KB, 500x652px) Image search: [Google]

254KB, 500x652px

Ryzen is a work of art, a work of love, a work of progress

>>59102422
>an 8 core cant reach 5GHz
>wow

>>59102523
When 4K games become common thing those Ryzen will shine too. It is alreadly shown that the 1800X at 3.6 gave a few more fps on battlefield 1 with a Titan X. If games don't use more than 4 cores, then we will be stuck on 4k @30fps.

>>59103459
>not superior nihonto

Nigga I'll fight you

File: bhjg.jpg (94KB, 440x294px) Image search: [Google]

94KB, 440x294px

>amd keks paying 14% more for a 7700k that has 9% less single thread performance and only 80% more multi threading

amdkek poorfags btfo again, how's that minecraft TNT fps retards? lmao i thought so

>>59104270
5960x could do it.. on a heavily matured architecture and process. Something Zen doesn't have yet. Zen+ is going to be the end of Intel.

>>59104367
Vaporeons are not food! wtf D:

>>59104449
pls at least i posted the cropped version

File: i7-2600KPower-ConsumptionAnalysis.png (21KB, 481x366px) Image search: [Google]

21KB, 481x366px

>>59099448
power draw increases exponentially with voltage and frequency. it's 65W at 3GHz and 95W at 3.6GHz if we go by AMD's TDP figures. note that TDP relates to the thermal solution required to keep the chip below a certain temperature threshold, not the actual power draw of the chip at listed specs, so it could be slightly less or slightly more.

i7-2600K was a 95W chip at 3.4GHz, but at 5GHz it'd pull 200W+ depending on voltage. same with AMD's flagship 83xx chips (which is why the lower clocked ones are advertised as 125W and the 9590 is 220W).

pic related

>>59104520
Voltage also decrease as node size decreases. Ancient chips pulled 5+ volts for example. You're also assuming Ryzen as an architecture is just as efficient as Sandy Bridge if you think tdp is going to scale the same way at higher frequencies. idk mang.

Will a low end ryzen chip be okay for a gaming PC? I want to build something cheap but decent.

>>59104767
have to wait and see later this year. The R7's are absolutely not for value gaming builds. Need about a $1750-$2000 gaming PC to justify buying a Ryzen R7 CPU.

>>59104865
You can build a $1000 PC with
>Ryzen 1700
>16 GiB RAM
>RX 480 8 GiB

>>59104865
>need a $1750 dollar computer to justify a $330 dollar cpu
what

>>59104734
not saying that at all. power draw will always increase exponentially with voltage and frequency. AMD's own specs list the R7 1700 as 65W TDP and 1800X as 95W TDP. beyond stock specs, who knows. maybe it'll turn out to chug 150W at 4.2Ghz, or maybe it'll only use 120W. unless someone wants to break the NDA and show us, we'll just have to wait.

>>59105070
>power draw will always increase exponentially
I thought it only increases quadratically.

File: intelmeme.jpg (57KB, 1160x800px) Image search: [Google]

57KB, 1160x800px

>>59102422

>>59105070
6900K pushes over 200W overclocked above 4.0Ghz

>>59105100
i'm retarded

>>59104767
Not sure about R3 but R5 will probably be pretty cheap and run games well in 1080p.

>>59104367
What you say doesnt even make any sense. 7700k is a 4 core processor, by far AMD has better SMT so it is over 200% beter multi threading than a 7700K.

File: cd82f7c18ed1bb96dac4000325d0cdb2.jpg (286KB, 600x800px) Image search: [Google]

286KB, 600x800px

>>59105163
>implying i didnt say 80% _MORE_
at least you tried

>>59102567
>4.1 Ghz all cores at 95W
straight up insanity when did I jump into this timeline

>>59098877
>600 USD due to insane demand on the 1800X
>300 USD mainboard
>200 USD RAM
1100 USD for the whole package
Meanwhile, the 6900K which is slower than the 1800X starts at 1100 USD, just for the CPU, and OC boards are ridiculously expensive as well

>>59104946
>pairing a high end cpu with a mid range gpu

allrighty then. oh wait, people actually do this

>>59099445
The top end FX's retailed at around 300 dollars more than the top end Intel parts back in 2003, which was around 1200 dollars
gb2/v/

>>59106125
Some people don't actually need a powerful GPU.

>>59105742
enjoy being water bended by the other vaporeons

>>59102059
>which is very impressive considering the huge price delta between the two chips as well as the low 65W TDP of the 1700
>low 65W TDP
They know the TDP difference is purely due to different clocks, right?

>>59102523
>The only chips that they've tested are the workstation chips.
They're priced comparably to Intel's way lower offerings, and even if your manchildren games run on less than 4 cores you can just disable the other cores (which the CPU probably would do by itself) and get much higher clocks on those 4 cores

Dear stupid shills,

Stop fucking promoting garbage chips that pretend 99 slow cores are worth anything, you harm AMD itself in the long run.

If you actually do succeed to shill for them, you will literally make Intel a lot of money because they will actually improve.

Hence, go fuck yourselves, nobody needs you, AMD doesn't need you.

>>59104367
Im sure ryzen r5 can btfo i7 when it arrives

>>59102620
They did but, like anything that isn't a desktop or server x86 chip, it's shit
Intel can't into anything but x86, and only on traditional markets
>>59102782
>dropping a 1000+ dollars SKU down to 400 dollars
wew lad sure it will happen
Intel can't compete since they can't bin every die like the madmen at AMD can
>>59103283
The Zen lead engineer said in an interview that a lot of improvements didn't make it into Zen's first iteration due to area and time problems, pointing mostly to FPU stuff like AVX512
>>59104520
You can extrapolate Zen's power draw at certain frequencies from their old slides, according to them it follows the same curve as Excavator
>>59104767
The lowest end Ryzen should be enough for pretty much any current game, bar a few RTS's
>>59106234
>desperation

>>59106155
the person was wondering about gaming pcs

>>59091035
Shill more.

>>59106306
>>desperation
You are the one quoting the entire board, spending your life in here trying to convince us nigger.

>>59106452
You're the one who's going to be jobless pretty soon my street shitter

>>59091587
>few extra bucks
More like double the price bud

>>59106172
jokes on you this vaporeon is being mulched by the owner under duress; they don't have a choice and there will be no justice for any of the vaporeons killing them and they're too stupid to know who was really behind it

>>59099445
AMD has had years it was ahead of intel and this will be another time it has that

>>59091135
>intel
>dropping prices
lmao

>>59103059

No they don't. Most volume is low tier i3s and i5s.

>>59106125
>pairing a high end cpu with a mid range gpu
It's the right way to go since GPUs lose value faster.

An ancient Q6600 from 2007 is still enough for modern games but a GTX 295 released 2 years later is unable to start modern games.

>>59091433
>Intel is building a new 7nm process fab in Arizona
lol, they haven't even gotten 10nm working.

>>59111744
>GTX 295 released 2 years later is unable to start modern games

Are you talking about DX11 only games? There are literally a handful, and right now only Crysis 3 comes to mind.

>>59092121
>consumerist
>retards
Back to Cuba with you

>>59106125
replacing a graphics card is 10 times easier than replacing a cpu.

>>59091131
You sound like some womens march retard.

>>59093347
>non-japanese asians
Who's Lisa Su, amirite?

>>59098841
>overclocked

ipc doesn't have to do with overclocking you fucking shill

>all the people that bought 6700k-7700k trying to convince themselves they didn't throw their money away

enjoy your denial

>>59099858
>amd cannot brat intel in a price war
>offers octacore 800$ cheaper

how much do they pay you per post?

>>59112725
5 cents :/

>>59112749
wumao

>>59091035
AMD fucked up and went for MOAR SORES instead of focusing on IPC back then. Now they'll have both, so hopefully we'll see better IPC and more cores in the future.

>>59104946
>>59104946
>rx 480

better buy a cheaper cpu and a decent gpu if you're gonna play games on it

Meanwhile im sat here extremely content with my 4790k which I have had for sometime and dont intend changing for the forseeable future.
I can clock the tits off the thing if I want to and itll do over and keep up with the best of them.
See now amd may (until results are released im on the fence still) have actually caught up. Fucking took their time though.

>>59106234
how much do you get paid per post?

>>59112641
Canadian.

>>59111880
>GloFo, Samsung and SMTC are alreadly playing with 7nm, to be launched in 2018

>>59113263
Intel confirmed for second rate cucks.

'It's upsetting to have to say, but should be said. As an IT Director I have already been approached by an Intel representative to discuss the "special promotions" and "Incentive rebates" that they are willing to offer me for purchasing exclusively Intel powered servers. I was not well received when I explained that I would not be purchasing anything until I see the benchmarks of the new Ryzen (naples) core and compare the cost-to-performance ratio. It was at that point they made it very clear that they would go to great lengths from a financial standpoint to incentivize me to NOT purchase AMD chips.

These are the tactics that were used back in the day that AMD had feared would be used again. Intel offered "loyalty rebates" and other clever promotions to hide the fact that they were paying customers to lock out the competition. I sincerely hope that other IT Directors and all in IT in general do what's best for the industry and choose based off of what's best for the consumer and their environment. The rebates are nice and all, but if we continue to hurt AMD for a small rebate, we destroy AMD's ability to compete and enable Intel to continue their "Tick-Tock" releases and oppressive pricing.

Be aware folks, Intel is scared and is already looking to use their weight and $'s to NOT out innovate, but to obstruct competitions ability to enter the marketplace. Why else would they be contacting and attempting to "Incentivize" customers well before the launch of AMD's server chips in Q2-Q3? '

FUCK INTEL!

>>59113416

Seems a bit hard to believe given that you're talking about servers here... Ryzen isn't a server processor and isn't being marketed or sold as such.

>>59113416
>Ryzen performs similarly to competing Intel CPUs
>main difference is Ryzen is substantially less expensive
>Intel offers discounts

Why is this a bad thing?

>>59113416
>>59113530

File: 135538415.jpg (102KB, 345x497px) Image search: [Google]

102KB, 345x497px

>>59091148
>>59091393
God bless America

>>59113459
the last time they did it they violated anti trust laws and had to pay AMD more than a billion in damages.
They are going to do it again this time because the fines are a lot smaller than the lost revenue of AMD.

Green GPU Green CPU