Well, what are your thoughts? Is this the end for Intel or AMD?

Neither, they both have their niche. People will buy one or the other depending on their own preference.

You have to remember, they've been at this whole little turf battle for years now and they're both still in it.

>>59175161
fpbp

>>59175141
>1700
>not the 1700x
>base clock only 3ghz
>competing with the 4.2ghz base clock Kaby Lake i7
>65w AMD chip with 8c/16
>91w intel chp with 4c/8t

AMD is a clear winner here. Reviewers are going to have to struggle to paint Ryzen in a bad light.

gta v is the only gaym where the difference is really appreciable and most people in the market for either chip won't be getting a titan x which is pretty much what you'd need not to get a gpu bottleneck anyway. 1700 has more versatility and is probably more efficient to boot but any high end gaymer should be happy with either, at least until vulkan/dx12 really starts to take off.

F A K E N E W S

A

K

E

N

E

W

S

File: dnaeNia.jpg (317KB, 1100x688px) Image search: [Google]

317KB, 1100x688px

And here is Ryzen against an i7-6800K.

>>59175305
>still loses

Why would anyone buy it? Its worse, end of story.

>>59175362
95w vs 140w.

why does amd need twice as many cores to compete with intel?

>>59175457
why do you need to shitpost

>>59175518
i'm just asking

>>59175141
> 1700 offering roughly 7700k gaming perf, minus iGPU for pass-through virtualization and troubleshooting, plus double cores for normal workstation use.

this is a pretty easy win for AMD IMO.

>>59175141
>>59175305
Comparing the R7 1700X to the i7-7700K from the first benchmark, is the R7 1700X better?

>>59175457
lower clock speed and poor thread utilization, mostly. single thread they share almost identical IPC.

>>59175362
>half the price
>same performance
why would anyone buy it?

those speed differences are well within the error range from framerate readers.

>>59175457
Didn't you faggots say that cores didn't matter for games?

>>59175375
1700 is 65w also underclocked AF, it is a 3.7Ghz against a 95w 4.5Ghz and they are almost even showing how massive is the IPC of ryzen.

>>59175554
If the games tested were making use of all 16 thread properly, the R7 1700 would utterly destroy the 7700k.
They aren't. Also, as the 1700 has more core, even if they are unused they drag down the clocks quite significantly, making it double worse.
It's enormously impressive the R7 1700 manages to be competitive with the 7700k under those conditions.

>>59175184

Yep,
>compating a 65w cpu to a 91w cpu.

>>59175141
That comparison is bogus on power usage.
On Price tho, 1700x is more expensive than 7700k. and 1700 slightly cheaper.

The price comparison is even better if you compare with 6700k at 300$ vs 1700 at 330$ at pic related same performance.

>>59175569
In those specific games shown, at whatever settings they used, it looks like the 1700X would edge out the i7 7700k.

>>59175569
R7 1700x is clocked higher 400MHz higher base, 100MHz higher boost (because binning). It's better indeed.

>>59175606
well i would hope so it has 4 moar cores

It's the end of one of the worst eras in computing history.

Intel has to improve, because Zen is just getting started.

Let's hope the server benchmarks are even more impressive.

File: 6OdWa03.png (960KB, 939x482px) Image search: [Google]

960KB, 939x482px

XFR chart

File: certified shit wrecker.jpg (13KB, 381x286px) Image search: [Google]

13KB, 381x286px

He actually did it.
I'm enjoying this comfy before the storm.

>>59175610
Would that mean the R7 1800X beat the i7-7700K? And if so, would that mean the R5-1600X will too at a cheaper price point, since they are both clocked the same?

>>59175672
I believe so, yes.
Not every game will show performance like that, but an awful lot of modern ones do now. This is looking incredibly good for AMD.

>>59175141
>>59175305
>>59175657

Source ?

>>59175695
From Ryzen event, still under NDA.

>>59175141
It's not over for anyone, Intel is still in a super comfortable position and AMD are looking better than they have in seven years or more. Prices will come down, the market will look at little less one sided and we'll still be saying "<name> IS FINISHED AND BANKRUPT" five years from now.

>>59175657
Idle power will be really low, I wonder if it'll boost though non intensive workloads like cellphone SoCs do. Could make for insanely low power usage for spreadsheets and things like that.

>>59175703
I doubt it wasn't don't on purpose, the conference was 3 hours long and we saw less than an hour of it publicly. They likely went over R5 and R3 specs too

>>59175141
>the only considerable difference is in GTA V
>with ryzen you get 4 extra cores and 8 threads
>outside gaymen, the Ryzen wrecks the 7700K
>only 65W TDP
The choice is obvious.

>>59175703

Thanks.

>>59175141
Oh wow, the budget 8 core with really shit performance is as good as a 7700k!

>>59175865
>$329
>budget
It's entry level for 8 cores
People really seem to forget that.

>>59175305
slightly higher ipc for intel but 2 more cores for ryzen for about the same price. pretty even. ryzen will be a bit better highly multi threaded tasks and is also a bit cheaper.

>>59175305
>winning in proper dx12/vulkan implementations
>works only on win 10 anyway
I don't see the problem. AMD has a clear goal, which is next gen api dx12 and vulkan.

>>59176324
Win10 support only is a Microsoft restriction.

>>59175305

IT IS OVER!!! AMD IS FINISHED AND BANKRUPT!!!

>>59175362
>>still loses
are you blind?

>>59175141
When does the NDA for the reviewers lift? I thought that was today?

>>59176871
March 2nd

>>59176926
March 2nd as in 18 hours from now or march 3rd for us euros?

>>59176967
Usually review NDAs are some time during the morning EST time

>>59175141
just imagine what amd could do with 140W in this new architecture

>AM4 Motherboard you were going to buy suddenly out of stock
>decide to wait instead of pre-ordering just a cpu
Holy shit, looks like a dodged a bullet

>>59177372
>just imagine what amd could do with 140W in this new architecture
burn your house down?

File: amd-is-dead.png (2MB, 2560x1312px) Image search: [Google]

2MB, 2560x1312px

>>59175141
Lmao

Ryzen is fucking horrible for gaming as predicted

>>59175184
Dude are you out of your mind?

the 1700 and the 7700k are the same fucking price

The results speak for themselves, you'd have to be stupid to buy a 1700 for gaming

>>59177461
Sure thing, shill.

>>59177444
No idiot should be buying the 1700 for gaming since it's clocked so low.

If you do buy it for gaming, you better OC it.

>>59177570
Truth

>>59177444
This was already debunked. You should know by now GTA5 engine has a lot of problems, even with a 7700k, especially with more cores like 6900k. The engine just is crap for anything else than consoles. It's a weird thing too as it's x86

>>59175141
I wonder how my i5 4670 would fare against both of those cpus

>>59177461
>The results speak for themselves, you'd have to be stupid to buy a 1700 for gaming
>you'd have to be stupid to buy the thing that costs the same but gives you twice the multithreaded performance and still matches in gaming performance
>you'd have to be stupid to buy the thing that gives you literally better performance per dollar
>you'd have to be stupid to buy the thing that draws less wattage
how the fuck.

File: 1479417260496.jpg (61KB, 702x562px) Image search: [Google]

61KB, 702x562px

i jus bought 7700k on ebay for 299 free shipping and feels bredy good

>>59177419
I feel sorry for the 6900k owners then

>>59175141
Wow. So the bottom end Ryzen 8 core is about 99% as good when it comes to those 4 year old games that aren't well threaded, AND I can use it for multi-tasking, rendering, and so on? So there's literally no downside. Does it use less power, too?

That's a massive win for AMD, but I have a feeling OP was being faceticious to get people to say what I just said.
jk OP here, got u guys

On another note, why is Civ VI so shittily optimized? That's a game that begs to be highly threaded. There should be no excuse for that with how it and all its DLC sells.

>>59177743
>This was already debunked.

Really? Then why does it match up with all the other benchmarks.

>>59177911
>On another note, why is Civ VI so shittily optimized?
because firaxis cannot into code.

>>59177911
The worst thing about Civ VI is that turn change is exclusively single threaded. It would be very hard to parallelize that, but it still sucks.

>>59177819
It gets beat in gaming
It has worse performance per dollar
AMD always lies about their TDP, wait until the real power usage comes out, the 1700X has already been shown to be 28W above the TDP.

Basically the 7700K is just a better CPU for most people. I would only get a 1700 if you are really into doing very specifically optimized multi-threaded tasks.

>>59177819

File: intel.png (148KB, 763x930px) Image search: [Google]

148KB, 763x930px

>>59177980
Proof? Their TDP is much lower, so they have much more room to overclock. You can even disable cores to allow more headroom to OC if you want to on the new Ryzen chips.

Games will be moving to more parallel computing too, so they will benefit much more from the extra cores in the near future.

Intel has been legitimately BTFO, they will probably try to drop prices, which will show everyone how much they were being ripped off in the first place.

>>59177972
Not really?
There are many different things calculated each turn, right?
Have thread for each thing.
Or put each AI on their own thread. So each turn, the AI does their thing on that thread and reports back to the main thread.

There are so many ways you should be able to thread stuff in that game.

But even accounting for turn change, that shouldn't affect FPS much except when turn is charging... So it still makes no sense how bad perf is for either there.

>>59175951
It's budget by octacore standards. Remember that Intel's octacore costs 1100$.

>>59178037
Overclocking isn't about TDP at all, dude. You retards need to stop repeating that nonsense.
It's about stability, voltage leak tightness, and how much voltage it can take.

And yes, AMD does underrate their TDP on CPUs. They go by something called "ADP", or have in the past. It will surely use over 65 watts in some conditions.
TDP is just how much cooling is needed. Hence the "T" standing for "thermal" and not "watts".
That said, I think it will still use less power than Kabylake in many cases.

I need to build a 4K editing (Premiere/AE) machine, from what I understand Ryzen will be better at this than Intel right?

>>59178043
Again, Firaxis can't into code. XCOM2 also runs horribly.

how come nobody surprised it trades blows in IPC with current top of the line consumer the only one for gaming 4.5Ghz intel CPU being 9-13% cheaper(it is cheaper here, you mileage may vary)?

>>59178087
reviews tomorrow, it's speculations now

>>59175184
This

Jesus, AMD has completely destroyed Intel with Zen.

>>59178088
I hear that but I don't understand how a company makes hundreds of millions of dollars but can't afford a good programmer.

Okay, I mean I know the answer.
They don't need good programmers for games. It's all about marketing. Doesn't matter if it's poorly optimized and runs like shit. People will buy it because marketing told them to.

>>59178087
Yes, for that kind of workload the more coarz the better.

>>59178104
It's less about marketing and more about the lack of Civlike games besides Civ and Endless Legend. They can do whatever they want, people will still buy Civ.

>>59178104
because believe it or not, developing games is hard and comes with a ton of deadlines

>>59178115
>endless series

>>59178072
>TDP
>The thermal design power (TDP), sometimes called thermal design point, is the maximum amount of heat generated by a computer chip or component (often the CPU or GPU) that the cooling system in a computer is designed to dissipate in typical operation

A huge part of overclocking is cooling the chip, so yes, as you increase the voltage you increase the heat output. AMD chips use WAY less power than Intel's current line, so there is a HUGE potential to overclock, much more so than Intel's chips.

The 1800X just beat the world record for overclocking in Cinebench. Ryzen can get to higher clock speeds than Intel chips with comparable IPC.

do you find it amusing that Intel spent billions upon billions on foundries and TDP meme over last 7 years to be beaten by samsung tech process and AMD in one year?

>>59175141

I bought my 8350 because it was a tremendous value and AMD is no-bullshit. Intel, on the other hand, has a million different product variations due to binning and intentional hamstringing of their products in order to "create markets".

I'm still bitter that I bought a Core 2 Duo that didn't support VT-x.

>>59178165
You can only fuck dead body of Pentium 3 for so long.

>>59178072
This

So far Ryzen looks like a garbage overclocker

>>59178206
All octacores are shit overclockers you fuckwit.

>>59178173
>AMD is no-bullshit

Holy fuck that is the funniest thing I have heard

AMD's marketing is the most deceptive bullshit I've ever seen

>>59178173
>Intel, on the other hand, has a million different product variations due to binning and intentional hamstringing of their products in order to "create markets".

This exactly. The 7600k, 7500, 7400, and so on just exist to make the 7700k look like a decent value when it's a terrible one and the 7700k should have been a <$250 chip.

Then the swath of Pentiums when the G4560 is the only actual good one.

>>59178165
It is funny.
I expected Zen to be like $650 and to only be like 85-95% as good in gaming as the 7700k. But still worth it compared to the 6850k and 6900k.

Instead it's $329 and like 97% as good in gaming on average and will probably become better with newer games as you can see with Doom Vulkan. There's literally no draw back to it besides
>muh igpu my netflix 4k drm
What a fucking ass rape. Really shows you how horribly overpriced Intel was for so long.

>>59178231
>There's literally no draw back

Ryzen is worse in gaming
Ryzen is worse in 99% of applications
Ryzen has shitty motherboards with only 8 PCIe lanes
No mini-ITX boards whatsoever

There are a lot of drawbacks

>>59178241
Nice damage control you got there.

>>59178206
>>59178215
>http://www.pcgamer.com/amd-ryzen-7-1800x-hits-52-ghz-breaks-benchmarking-record/

Intel shills BTFO.

Ryzen uses significantly less power than Intel's chips, hence they can blow them the fuck out in clock speed too. You can even disable cores with a tool they are supplying to reduce heat output, so you can clock them even higher.

You could take a 1700x, disable 2 cores and clock it to 4.8GHz with 6 cores etc,

>>59178231
there is no such thing as overpriced just the price people were willing to pay

this isnt fucking bread and water we're talking about this is just a fucking expensive toy and/or tool most people dont even know exists

the fucking entitlement of some /g/ units is nothing short of marxist

how dare fucking home depot sell me reciprocating saws for X dollars!!! thank god harbor freight came out with a 6amp 1500RPM unit for 65 bucks! just shows the JEW TAX fucking home depot does. fucking kike motherfucks

>>59178217
No, AMD's marketing is the third most deceptive in the major semiconductor processor industry.
Leading the pack is intel.
If you need an explanation for that you are beyond saving.
Second is Nvidia. 4GB ring any bells?
Third is AMD. That 2x480 vs 1080 was technically true but horribly misleading. 2x290s would also beat a 1080 in DX12 Ashes

It was just a horribly misrepresentative sample, the epitome of cherry picking.

AMD rarely flat out lie though. Intel does a lot of that to best buy employees, OEMs, and anyone else that takes bribes.
Nvidia lies a lot too, wood screws?

>>59178241
I'll bite.

>Ryzen is worse in gaming
3% amrgin of error
>Ryzen is worse in 99% of applications
any proof of that? how can it be 3% in games and suddenly WAY worse in other applications?
>Ryzen has shitty motherboards with only 8 PCIe lanes
there are boards with 8 SATA and 2 m.2 slots what are you on about?
>No mini-ITX boards whatsoever
I'm sure that 0.5% is very sad about it

>>59178260
>Third is AMD. That 2x480 vs 1080 was technically true but horribly misleading
The thing is that even when they said that, they said "if the software makes use of it".

They never said "2 RX480 is the same as 1080 always"

So it's nothing like the lies of Intel and Nvidia.

>>59178281
Well, I guess nothing beats 3.5 and woodscrews.

>>59178241
Three Intel Inside (TM) Pennies have been deposited into your account.

>>59178281
It was clearly intended to be misleading.

>>59178311
it clearly was intended for niche CF market
I don't think many people left that still believe in SLI/CF or ready for drawbacks of it.

Waiting for the cheapest 6 core ryzen getting gaymen benchmarks at 4-4.5ghz oc

>>59178326
It's a very niche market indeed, consisting of people who only play Ashes of the Singularity.
AMD's decision to make DX12 cards is a good one for the industry, but that benchmark, being displayed as representative of that card was shameful.

where's the 1800x benches?

i dont give a shit about the rest

>>59178295
Nvidia has done a lot more.

They were heavily gimping Fermi GPU compute power after the 400 series because Nvidia spent a lot of money getting companies like Adobe to give first-class support to their GPUs and not AMDs, but people were buying GTX 480s instead of much more expensive Quadros

In some cases, their compute power of their top of the line cards was the same as previous gen Tesla cards, and like 1/3rd that of the 7970. This is a large part of why Fermi cards are so dogshit in modern games, as their compute was never great, then got gimped along the road, and modern games use a good deal of GPGPU compute when they hardly did when the 7970 came out.

>>59178394
7970 was still a killer card.

>>59178339
I have dual 480s. I have them because some games XFfire well, others get disabled. But mostly I have it for autocad and compute uses.

>>59178435
Yep I still have one. Waiting for Vega/Volta.

But I'm amazed that people were duped into buying 680s when it was so clear that the 7970 was a killer buy even when it came out before the 680.

Not to mention the 680 still being sold with only 2GB of memory on some models. Like, holy shit.

>>59178311
This. AMD just flat out lies with crap like this.

>>59175141
show us some 1080p charts surely its more stressfull to the cpu than 2k..

>>59177934
why does the oc get a worse average then the stock, why does ryzen get damn near the same average with worse lows and highs?

the bench is weird no matter how you look at it.

>>59177911
yes it uses less power too.

>>59178490
8cores actually work better at 1080p, it would look even worse for 7700k

>>59178104
the games they make are playable as board games, so instead of focussing in making the graphics perfect, they focus more on mechanics.

I sure do want to buy second best hardware said no one ever

>>59178534
its cpu intensive resolution this is what we wanna see benching a cpu on 2k and 4k while the gpu is the bottleneck doesnt shows us nothing

How is the majority of the games benched the 6800k is beating out the 7700k??

>>59178477
really? did ashes of the singularity NOT use 2 amd gpus?

>>59175596
......What

Where are you seeing this price difference, the gap between the 6800k and the 1700x is $25.......

>>59178258
>there is no such thing as overpriced just the price people were willing to pay

no, not willing, put up with is a better way to put it, don't make it sound like people WANT to pay that much, they grudgingly did it.

kind of the same way I use chrome.

There is no realistic other options so I grudgingly use it hoping fucking SOMETHING will come along.

>>59178550
honestly, 2k on a gpu that pushes gta5 to nearly 200fps, 1080p isn't going to much

>>59177444
You've got to be a troll if you can't ascertain the fact that highs mean literally nothing if you can't keep a high average.

>>59176324
amd seems to be going for vulcan rather then 12

>>59178543
I didn't comment about the graphics.

It's the fact of how terribly it performs and how single threaded it is. It's the most single thread intensive game on that last.

>>59178604
>kind of the same way I use chrome.
use Vivaldi.

>>59178589
Benchmark is the 1700 not 1700x

>>59178644
What i'm saying is if the gpu is pushing gta to that extreme, you are already hitting cpu bottlenecks most likely, hell, I don't even know how they got fucking gta5 to go what, 180 fps...

File: p4doom3.gif (19KB, 433x462px) Image search: [Google]

19KB, 433x462px

Intel will survive Ryzen, they always find a way with their illegal monopolistic practices.

The situation was much worse for them in the past anyway. Pic related shows a $1000 Intel CPU getting beaten by a $350 AMD CPU.

>>59175141
If you compare the 6800k and the 7700k from the charts the lower ipc 6800k does 10fps more. Wtf?

>>59178665
there is only one way actually

no aa
no af
no msaa
and im quite sure (10000000%) those numbers arent from 2k

>>59178695
215 FPS!!!!

>>59178089
because almost all of us who are on side amd thought it would be haswell ipc, then when amd said they were getting better ipc and that they were competitive, we remember the shit wrecker and new they meant actually competitive, not well, single core isn't as good but we got more threads competitive.

We aren't shocked that its this close, hell those of us who realize amd put out a tool that can disable 2-6 of the cores along with xfr realize that amd every well could outperform intel with a good enough cooler, and that anything shown by amd is using what it's giving people as stock.

on the second we will either say well that's a shame, xfr wasn't as good as we hoped, or we will all be pleasantly pleased we held off on buying a new cpu for ryzen.

>>59178165
believe it or not, amd has likely more experience with tdp reductions than intel does thanks to bulldozer and working on arm.

>>59178713
Yeah XFR has been the only disappointment for me. But it's a small disappointment. It's not like Intel offers something better.

I would have liked XFR to go higher if you have the cooling, and to have per-core voltage regulation for overclocking. That would have made it beyond my expectations and perfect.

>>59175141
>3Ghz vs 4.2Ghz
Spend $80 more on a 1700x and it will rape the 7700k in games and everything else

>>59178727
honestly, amd according to linus said it went to 4.1, however, and keep in mind, many pc people and reviewers are claiming that manual oc is dead...

They got ryzen
They got the core disabler tool
and they know something we don't.

there is still interest to be had there, just wont know for a fact till reviews.

File: INTHELOO.png (316KB, 882x758px) Image search: [Google]

316KB, 882x758px

IT WILL BE GOOD THIS TIME I KNOW IT. THEY WON'T LIE AGAIN TO MEEEE. HAHAHAH

File: Hmmmm.jpg (16KB, 480x360px) Image search: [Google]

16KB, 480x360px

How do AMD fags feel that most people want AMD to do good to buy cheaper intel CPUs

>>59178741
they dont realise that amd is sandbagging till the nda

its like those idiots that compares 7700k with the 1700 because amazon and micro made some cuts

i wonder what will happen when the 7700k goes back to the normal 350 and they will have the x to think about

>>59178732
1700x is definitely better in games but in multi threads 7700k gets brutally raped

>>59178760
Eh.
I just wish people were smart enough to understand how bad Intel business practices are.

There was no reason why they couldn't sell a 4c/8t CPU around $200-$250. It didn't have to be so high clocked or unlocked. They just don't that to artificially create a market where the i7-7700k seems to be a reasonable value when it's not.

The $60 2c G4560 has HT. The $160 i3-7350k has HT and is unlocked. HT is not expensive, it's just arbitrarily disabled on the 4cores except for the 7700k.

The 7400 and 7500 really should have had HT. They don't no reason other than so have to buy an overpriced 6700k and 7700k CPUs since they made that a high entry point for 8 threads.
They know people buying a 7400, 7500, or 7600k will have to upgrade soon because 4 threads just isn't going to be enough soon, and they were hoping they'd control the market to the point that they'd have to buy Intel again. It's planned obsolescence.

Supporting a company that does that is terrible, because they will keep doing it.

>>59177911
>why is Civ VI so shittily optimized?

You mean, why can't we run a Civ VI on a first generation Pentium?

I don't know. I don't know why it's a 3d game in the first place. Completely idiotic. But post-Civ III civs are superfluous anyway so might as well stick to III.

>>59175141
Why is this happening? 4 cores and megahurtzs is all you need for gaming, I hope nobody ever overclocks the 1700 or it wouldn't be fair

>>59178766
they aren't really sandbagging it, they just aren't playing all their cards.

They aren't saying shit because literally every single thing they say is newsworthy across the board, its free pr. sandbagging it would be showing the thing underclocked doing ok, they haven't really been doing that, but they have been going after the far larger targets of the 6900 and 6800 kind of just ignoring the 7700k

I honestly wouldn't be shocked if amd was worse then it more often then not even when oc'ed on 4 cores or 2 cores, but I don't believe that it will be worse by enough to justify 4 cores over 8 cores.

That and if I can have a saved oc preset that underclocks/volts the absolute fuck out of it and its stable, I could designate that sleep time clock rate and leave the pc on overnight during the summer and wake up to not a sweltering room, that has value to me if its simple.

>>59178908
the move to what was it hex based grid was a very good upgrade.

>>59178043
Well a lot of "multi-threaded" games just toss one thing like audio, AI, etc. onto a dedicated core instead of making some of them truly multi-threaded.

>>59179148
Considering that rendering performance hinders framerate the most and that rendering is often restricted to a single thread, then offloading even the smallest tasks of it will help the game run smoother. It's hard to make a multi threaded game, you know.

>>59179169
But much easier than before, thanks to better tools and libraries.
Else you wouldn't see a 8 core with a 1.3GHz clock deficit be within spitting distance of a 4 core with a 4.3GHz

>>59175141
on the CPU market AMD definitely won with 1700x and the 1800x .

>>59175184
Only upside to Intel is they can go ~5Ghz +-.1

Meanwhile AMD stutters at around 4.1Ghz. If it can go higher by disabling some cores, then it will be gud. Intel has no saving grace.

>>59175184
This, why is OP so desperate?

Why do Intel shills lie so often?

>>59179102
I'd say 1700 won, even if the difference in GTA V is significant, 1700 is a bit cheaper, cooler and MUCH better in multithread.

>>59179169
True, but it seems like all the multi-threading optimizations are being done on consoles and staying on consoles since 1st party teams from Sony don't port to PC.

Now DX12 and Vulkan are multi-threaded APIs but, yeah, there is more to gfx than just the API overhead.

A Ryzen 5 6c/12t part that's cheaper and higher clocked than the 1700 would be a better value for gaming.

So for gaming, unless you want to risk a bad 1700 OC, get a 1700x, 1800x, or wait for the top-end Ryzen 5 in a few months.

>>59179209
Not even, we got a CPU-z bench of a 1700X at 4.5 and Donanimhaber got their 1800x over 4.6GHz on a Noctua, these might be cherrypicked chips so who knows though.

As for the 1700 I don't really expect it to reach over 4.0, since its base clock and voltage are way too low to start with.

>>59179209
I've heard that not that many 7700k's can reach 5.0 GHz. Anandtech's one surely didn't (due to high temperatures and AVX being a power hog)

>>59179209
Unless AMD is binning them close to the limit, they'll go much higher than 4.1 GHz.

With the FX line anyways, they all can be OC'd to their Turbo frequencies with ease.

>>59179241
>believing donanımhaber
Not even here, in Turkey.

>>59179249
22% of them reach 5Ghz, iirc.

>>59179249
But doesn't literally everyone have 7700k's at 5.3 GHz already?

>>59179241
>As for the 1700 I don't really expect it to reach over 4.0, since its base clock and voltage are way too low to start with.

That's the thing though, depending on the demand, AMD can take perfectly good 1800x chips and label them 1700.

Some lucky fucker is going to buy one for $330 and OC to 4.6 GHz on a Hyper 212.

>>59179280
That will only happen if 1800X demand is lower than 1700 demand, which according to preorder numbers isn't gonna happen, currently.

>>59179276
I know these are memes but according to overclockers.com one out of thirty can reach 5.3GHz, and that's after deliding, sanding and a custom water loop come into play

>>59179229

But can it play Netflix at 4K?

>>59179312
A 6700k can't, but a $50 Pentium can, 6700k is literally finished It can't use DRM

File: df5c790c1108.jpg (49KB, 806x253px) Image search: [Google]

49KB, 806x253px

imho once the 4c/8t 200 bucks hits the market, there will be no more reason to buy the i7 7700k

>>59179318
>4 cores
Lel, I'd rather just get a cheap 8 core and overclock it.

>>59175141
<company I didn't buy from> IS BANKRUPT AND FINISHED

>>59179318
There is currently no reason to buy a 7700k. Just impatient retards and general idiots are buying them.

Same sort of people that bought 1070s.

>>59179336
4k Netflix

>>59179318
fake and gay.

the 1600x will be a lot faster than that.

>>59179363
Didn't AMD's own slides put the 1600X at 3.6/4.0?

File: fpf04_01_w_300.jpg (21KB, 300x400px) Image search: [Google]

21KB, 300x400px

>>59175141
Jews HATE him.

>>59179336

Yes there is. Kaby lake is ACTUALLY released and I can buy one right now.

Also, the Ryzen 1700, IF it's actually released, will only be 3% cheaper than the 7700K in europe (1700: 359 eu, 7700K: 371 eu). I'm not puting a slower, unreleased CPU without netflix support in my PC to save a bit over 10 euros.

>>59179390
>2017
>caring about 4k Netflix support
>even owning a 4K monitor in the first place

ISHYGDDT

Yes.. yes Netflix very important yes mm blessing from Tel Aviv

>>59179390
>Kaby lake is ACTUALLY released and I can buy one right now

Ryzen will ACTUALLY be released in just over 13 hours.

>>59179398
2017 US media goyim agenda:
Make consumers think they need a 4k display. Limit 4k playback to select devices. Limit physical media. Charge exorbitant amounts for the bandwidth required to stream it.

File: 1442743373405.png (199KB, 467x456px) Image search: [Google]

199KB, 467x456px

>>59175141
>3GHz is trading blows with 4.2GHz

>>59179421

AMD still has 13 hours to f*ck shit up. I'm not a believer until ryzen is released, bought and shipped to regular customers for normal prices and performs a bit like the leaked benchmarks.

File: 14883653687520.jpg (304KB, 2521x1222px) Image search: [Google]

304KB, 2521x1222px

>>59175141

File: 234.jpg (13KB, 400x400px) Image search: [Google]

13KB, 400x400px

>>59175141

I currently have a 4690 with 12GB of RAM.

Should I upgrade to the 1700?

File: 14883531170321.jpg (110KB, 750x537px) Image search: [Google]

110KB, 750x537px

greetings from 2ch, neighbours

File: 1411560289704.cached.jpg (190KB, 2000x1333px) Image search: [Google]

190KB, 2000x1333px

>>59179408
B-but my reruns and degenerate originals!

Gotta get Intel if you're a normie or a girl, soon sports will require Kaby Lake too!

>>59179455
Pretty sure because AMD didn't jewed out the cooler, the CPU runs on full Turbo frequnce. (3.7 ghz)

>>59178695
Their own Pentium M beat their Pentium 4.

>>59179390
here in Poland 1700 is actually more expensive than 7700k

>>59177911
who needs more than 20 fps in a turn based game?

>>59179518
My polak bratanki, Weiger here.

Thats just launch price shit. The RX 480 costed 300-350 euro here at launch. 1-2 months later it was down to 230-280

File: hqdefault-2.jpg (34KB, 480x360px) Image search: [Google]

34KB, 480x360px

>>59179515
daily reminder that a pentium m is just a pentium pro with shit strapped on

and that today's intel cpus are just pentium 3's

intel got tinier and faster but not really better in what, two decades? the most revolutionary development came from AMD toplelhat.jpg

>>59179483
cyka blyat

>>59178908
even in the old versions higher difficulties would have so many units stacked turn timers were in excess of 5 minutes for top line machines ~2003-2008

real question is why is civ on a benchmark to begin with

>>59179518
No, it's not. I've checked x-kom, Komputronik and Morele and 1700 is around 100 PLN cheaper than the 7700k.

>>59179518
>>59179390

Good to see Europe keeps poo out of PC.

>>59179490
Max turbo is only single core, just like 4.5 turbo for 7700k is only for single core.
If it was all turbo under 65W Intel would be annihilated as much as I want that to happen it's a no go.

>>59179527
I don't mind a Pentium Pro with a giant strapon.

>>59179544
you can buy brand new 7700k by using allegro or ceneo, they are for 1540-1590zl each so cheaper or about the same price.

http://allegro.pl/procesor-intel-core-i7-7700k-4-2ghz-8mb-box-i6722330114.html

also on ceneo there was offer 2 days ago to buy i7 for 1550zl

>>59179520
Dragging the map around at 20 FPS is a miserable experience.
I have around 350 hours in Civ V, about half of those hours were played at 30 FPS due to having a toaster.

>>59179552
Do you think if I upgrade my 4690 (non-k) to a Ryzen 1700 and overclock (hopefully all cores) to the same Intel Turbo frequince (3.8Ghz) will I gain any performance in gaming... Or at least... Would it worth it?

>>59179597
You'll gain massive performance everywhere, not just gaming.

Too bad the cost of DDR memory is sky high right now. So damn expensive to get new cpu+motherboard+memory

>>59179597

Unless you live in a 3rd world country like the US you can get the faster 7700k for practically the same price AND have 4K Netflix.

>>59179602
>Yeah, the i5 4690 is already a couple years old technology, its safe to assume Ryzen will beat it with Single thread performance

Its just pretty shitty they didn't released a single R5 6c12t processor. 8 core is still feels like an overkill for me

It is time to remember
> 2133
> quad channel is not supported
> 4 modules works unstable
> single thread is lower than intel
> fps in aaa games is lower than intel
BUT amd ryzen have kykyryze cores
LOL
Pwned by /hw/

>>59179597
1700 is the lowest binning. It being able to overclock 3.8GHz on all eight core would be massively impressive. Don't count on it.

>>59179657
What does 2133 mean? They support 3200mhz?

Whats the benefit of Quad Channel Memory?

>>59179636
>AND have 4K Netflix
That's not a feature or anything really desirable. It's DRM that only kabylake supports.

Pretty much any mid-high end CPU made during the past 5 years can play 4k video.

>>59179664
Don't OEMs already sell pre-overclocked 1700's at 3.7?

>>59179657

You forgot:

> no 4K Netflix

>>59179669
>Whats the benefit of Quad Channel Memory?
Essentially doubling the memory bandwidth (if you do have at least 4 sticks)

>>59179677
No, that's the boost frequency. OEM don't sell pre-overclocked CPU at all, this only happens with GPUs.

>>59179672

>4K Netflix
>That's not a feature or anything really desirable.

Maybe not for you but I watch netflix on a fairly large screen and I like 4K

>>59179552
>4.5 turbo for 7700k is only for single core.
that's not true though, as well as 3.7 turbo for ryzen only on one core, at least not confirmed anywhere
but considering that 6900k turboes to 3.7 on all cores i'd say ryzen does the same

>>59177444
>i7 oc'd gets 1fps avg lower than stock
>1700 get 4fps lower average than a stock i7 with 0.6 less ghz clock speed

AMD just better start liquifying their assets now.

>>59179697
Tell me more about this DRM feature. How is it good for anyone.

>>59177461

remember the price cut? :^)

Officially amd doesnt supported > 2133 MHz memory, please, see the cpu specifications from amd.

>>59177570
So which AMD cpu are you supposed to buy for gaming then?

>>59175457
It doesn't. None of these games are utilising all cores, but the i7 is clocked MUCH higher.

>>59179750
>Intel jewed every single cent out of your pocket (CPU manufacturing cost less still raised market prices)
>Finally competive AMD again
>Intel reveals they selling $50 and $100 CPU's for $200 $300

And you are okay with this

>>59179767
>Officially intel doesn't supported > 2133 MHz memory, please, see the cpu specifications from intel.

File: ok.png (254KB, 624x525px) Image search: [Google]

254KB, 624x525px

>>59179767

>>59175141
What's up with GTA V then?

It does amaze me that Intel shills thought Ryzen could never compete with i7-7700K at gaming, when we already know that Intel's lower-clocked lower-IPC hex-cores beat the i7-7700K in a decent number of new games, plus the fact that in most scenarios you'll be GPU limited anyway.

>>59179767
>lying on the internet
All the motherboard specs show Ryzen supports DDR4-2666
Intel only supports DDR4-2400

File: cNcfwxrhMqVbIJODZiaZPqt0cWLSDka1fle8Bo9jsZk.png (303KB, 1024x452px) Image search: [Google]

303KB, 1024x452px

Ayyyyy

Makes you really think

File: lol.png (55KB, 658x429px) Image search: [Google]

55KB, 658x429px

>>59179750
yeah microcenter is smart, getting rid of its kaby lake stock before ryzen hits!

File: 1346497415942.jpg (7KB, 176x200px) Image search: [Google]

7KB, 176x200px

>>59175362
>Why would anyone buy it?

>within margin of error for 5/6 games tests
>lower TDP
>cheaper
>better productivity from more cores and better SMT
>4 and 6 core versions will be even cheaper if productivity isn't required

>>59179798
You can also get a 1800x + 1080 Ti for a very close price

>>59178252
Holy fuck you are new. Read >>59178072

Ryzen is a poor overclocker and it has nothing to do with TDP

>>59179796
that just depends on the memory cache of the cpu, you can get a chip that supports only 2133 or a chip that supports 4666

i swear for a technology board you faggots know nothing about cpus

>>59179816

>>59178623
...look at the minimum

Ryzen looks like a stutterer

>>59179795
GTA5 is also pretty unreliable with performance. You can get two very different results just from alt tabbing out and back in to the game.

Seen so many people with performance issues that have to toggle fullscreen or disable+reenable vsync to get decent performance.

>>59179792
>$250 mobo

fuck off

File: ryzen-doa.png (177KB, 1089x677px) Image search: [Google]

177KB, 1089x677px

>>59178732
Really? Looks like the 1700x still can't beat a 7700k

>>59179832
We'll see.

>>59179347

>he feel for the 4k streaming meme

File: Mysterious Merchant.gif (4KB, 452x523px) Image search: [Google]

4KB, 452x523px

Oy! ryzen is DOA. BUY INTEL 4 CORES. NO NEED FOR 8 CORES.

>>59179229
>lower fps
>1700 won

AMDtard logic everyone

>>59179853
What games does userbench test and why is the 1700X without turbo?

>>59177461
>less than 5% difference in all cases.
>Twice as many threads
I'd buy the 1700

>>59179688
>essentially doubling the memory bandwidth

and totally forgets to say that quad setups offers absolutely nothing more
http://www.pcworld.com/article/2982965/components/quad-channel-ram-vs-dual-channel-ram-the-shocking-truth-about-their-performance.html

File: 100euro.png (264KB, 1459x767px) Image search: [Google]

264KB, 1459x767px

>>59179851
>MSI B350 TOMAHAWK
>Cost $100-30

Well ain't that a bitch

>>59179878
Is AMD's new slogan, moar threads!

They are making the same mistakes over and over again.

More cores and shitty real world performance.

>>59179853
5.15 vs 3.5ghz, if rizens can't overclock well (or xfr for that matter) than there's no doubt it will be another failure on amd's part, but if it can it will destroy intel easily

>>59179853
why you photoshopped out the non turbo on the 1700?

kek people are so idiotic sometimes

>>59179874
Userbench is CPUboss tier, it runs some DX11 APIs by the CPU and that constitutes as 'gaming performance'

>>59179900
They reached 5.2Ghz Stable in LN2 with the 1800x (on all cores)

Pretty sure 1800x reaches 4.4-4.6Ghz on air while the 1700x is slightly below (4.2-4.5Ghz)

>>59179899
So why is Intel introducing 6c/12t with Coffee Lake with their next i7s? Stock freqs will go down lower than 7700k if they want to keep the same TDP.

>>59179826
Motherboards clearly state which memory speeds are stock speeds of the CPU and which are OC.
Intel boards say 2400 is stock. Anything above is OC.
Ryzen boards say 2666 is stock. Anything above is OC.

File: 7700k.jpg (312KB, 1208x869px) Image search: [Google]

312KB, 1208x869px

hi, 1700x & 1800x customers

can your CPU do this?

>>59179953
No, but neither can 95% of 7700ks.

>>59179953
>throwing a 4c cpu into an 8c discussion
>trying to shill
>not realising that the 2449 cinebench score was actually near that speed

being this idiot

>>59179953
Can you actually run a benchmark with that?

>>59179953
run something with that, either will crash at 30% use or fry your mb

>>59180015
no

>>59180008
>that's not fair! he can't use a cheaper and better CPU to beat Ryzen! he's not playing by the rules!

>>59179869
>101 is a smaller number than 98
You're clinically retarded, Intelshill.

>>59179953
Those voltages and clocks don't match, run something that will 100% the CPU

>>59179911
Uhh there's no photoshop at all, that's literally directly from the userbenchmark website

>>59180008
>>not realising that the 2449 cinebench score was actually near that speed

it was also at like 1.85v, not 1.29. lol

>>59180027
>>59180015
unigine has been running all day, hence the load temperature

i have 7700ks which are better than this one by the way

>>59179672
If enough people buy AMD then Netflix will make some solution for showing their shitty reruns on Zen. Or not, and Hulu or another shitty rerun service gets their business.

We've seen this kind of shit before.

>>59180044
Holy fuck you are blind, you might want to actually look at the benchmark

Why do AMDtards always lie?

>>59180051
more cores needs more voltage? WOW

people skipped physics day i guess

>>59180051
>7700k at 6.0 under 1.3v

I don't believe even the CEO's son has those.

>>59180055
Sadly very few people are going to buy a $350+ processor that can't even compete.

>>59180066
>l-look at this cooked intel benchmark
>we're not nervous goy
>gulp
intel shills need to tuck in for night-night

>>59179657
Also:
> amd ryzen cannot be overclocked

>>59180084
Go ahead, keep ignoring the evidence AMDtard

AMDtards are allergic to numbers, don't you dare show them benchmarks.

>>59180051
yes 5.8ghz on all cores and you are able to run ALL DAY LONG unigine...

do me a favor since you have such a beast of a cpu run y cruncher for 30 mins on that speed record a video for us of it blowing up

>>59180066
you're just assblasted baby lake is a failed nuclear reactor

>>59179953
We got a world record here on /g/, reddit will be so jealous.

>>59179953
BIG frequencies

How can little amd white bois compete?

File: getchart.png (13KB, 500x342px) Image search: [Google]

13KB, 500x342px

>>59180091
>benchmarks
>as Intel's stock continues to crater precisely because of benchmarks showing zen ripping the shit out of intel's much more expensive parts
tippity top fucking kek, intel

>>59179953
>same of 30 7700k owners on overclockers
>2 of them managed to get 5.3 with 1.5V
>rest are under 5.1

And I'm supposed to believe you got a stable 5900 at 1.3v? Even with SMT disabled?

>>59179953
If this is legit then I congratulate you on some really amazing silicon there.
If I bought a hundred of those I probably wouldn't even get close.

File: 1488345199971.jpg (184KB, 1440x1251px) Image search: [Google]

184KB, 1440x1251px

What's with the homosexual overclockers? Most people never overclock, this includes probably 99.999% of gaymers. Is this supposed to be a point in Intel's favor?

It's just strange that it's brought up so often even thought it's completely irrelevant. Another sign of desperation from the kikes at Intel.

>>59180118
he is cooling it on space dude how dare you judge him he has proof and everything its not like you cant shop a photo nowdays

>>59180141
>Most people never overclock, this includes probably 99.999% of gaymers. Is this supposed to be a point in Intel's favor?
It's hard for overclocking to be mainstream, when intel cucks all their customers by locking it away into K models.

>>59180168
actually space is a very good insulator
theres a reason spacecraft that do engine burns in space need giant radiators

>>59179748

I don't care about the technicalities behind 4K. I just want a PC that supports 4K in the current year.

If AMD doesn't fix this than I'll pass. . .

>>59180200
This... AMD doesn't support 4k monitors in 2017... W...T....F..?

>>59178695

>tfw owned a 3200 back then.

Golden days of AMD.

>>59180223
lol you must have awoke from a cryostatis pod or something

>>59179851
What are you, poor?

>>59175161
>>59175181
This

>>59175564
>pass-through virtualization

With SRIOV coming with newer GPUs, no iGPU won't be a problem

When can I see a test amd vs intel, where intel and amd has the same frequency, the same count of cores (physical and logical) and the same ram?
Never.
Only stupid tests.

>>59180459
That's a dumb test

You should test the processors that are the same price, i.e. the 1700 vs the 7700k

>>59180141
>better overclocking is bad

Holy fuck AMDtard logic is like living in opposite land

>>59180480
And you should never test the 6900 and 6950 that are double and triple the price, because that would make Intel look bad.

>>59180099
Wow that's really insightful AMDtard. What an argument. Please do go on.

>everyone keeps spamming 7700K prices
>nobody mentions that Z270 30-40 bucks more expensive than same X370 boards

>>59180499
What makes AMD look bad is testing the 1700 vs the 7700k, because they are the same price and the 1700 fails badly

>>59180480
the well are you on about? what is the logic of having a cpu that will certenly go up on its normal price in a few days since its just a stock cut and not a real price cup ?

shilling much?

>>59180516
Holy fuck was that even English?

>>59180513
B-but more PCI-E lanes, g-goyim. Buy Intel.

>>59179798
Fufck you poorfag.

>>59180523
i think what he is trying to say is

dont compare a cpu that just got a price cut on micro and amazon with a cpu that has an msrp 40 bucks lower than its normal 7700k price

>>59179624
>This
Too bad the price of memory is putting a dampener on things.

>>59180480
In this case I will be know a clear performance difference betwen intel and amd.

>>59180556
That doesn't make any sense though

The 7700k is $339 and that's not a sale price

The 1700 is $329

>>59180514
>5fps slower in cherrypicked videogames
>50% faster in actual applications
>failing badly
Just listen to yourself

>>59180545
7700k has 16 PCIe 3.0 lanes on the CPU.

File: 2017-03-01-130407_597x411_scrot.png (40KB, 597x411px) Image search: [Google]

40KB, 597x411px

>>59175141
End of windows.

>>59180545
Only on the chipset. Ryzen and Kaby Lake both have 16x PCIe 3.0 on-chip.

>>59180575
It's actually slower in applications, hell even a 1700x can't beat a 7700k, see >>59179853

Literally the only thing Ryzen is good at is very specific apps heavily optimized for multi-threading, like video encoding.

>>59180545
but 7700K has less lanes, do not lie to me mister!

>>59180586
no, look
ryzen has on CPU 24 lanes
kaby has 16 lanes on cpu and everything else from the chip

>>59180590
Why is the turbo disabled on the 1700X? It's a 1700 in that case.
What games does userbench test?

I'm gonna ask this every time you post that nonsense until you give me an answer.

>>59180590
>3.0(turbo disabled) chip losing to 4.5

fantastic! how can it be!

>>59180590
>userbenchmark
Great application. I use it every day

>>59180615
Try googling userbench

>>59180626
Give me an answer, I want it straight from you, I already know what games it tests but I want you to tell everyone here.

>>59180631
Give me an answer, I want it straight from you, I already know what games it tests but I want you to tell everyone here.

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

245KB, 800x612px

>>59180651

>>59180590
i dont know seems to be pretty fishy considering this photo is photoshopped hard
http://cpu.userbenchmark.com/Compare/AMD-Ryzen-7-1700X-vs-Intel-Core-i7-7700K/3915vs3647

i guess giving the actual link doesnt help lying about it

>>59180604
So X370 has 32 lanes total? That's plenty.

Tomorrow is going to be a glorious day watching all the Intel shills get butthurt.

>>59180670
How is that photoshopped? It's literally directly from the page.

Looks like there are now 3 benchmarks, still showing the same thing.

If a 1700x can't even beat a 7700k, Ryzen is in major trouble

File: Screenshot from 01-03-2017 14:15:15.png (49KB, 1051x378px) Image search: [Google]

49KB, 1051x378px

>>59180621
>>59180615

See, when this was posted, some days ago, the userbench leak it was tested against a similarly clocked Kabylake to test IPC, hence the disabled turbo.

Now it's just become shill material since it's easy to test the 1700x w/o turbo against anything now, even a 2500k or any kind of clocked chip.

I'll try to find it...
Yup here is the original post.

http://cpu.userbenchmark.com/Compare/AMD-Ryzen-7-1700X-Turbo-Disabled-vs-Intel-Core-i5-7500/m243187vs3648

<-

tl;dr pure desperation

>>59180685

We'll see. Until now all AMD had to do is make promises. Tomorrow they will actually have to keep to them.

>>59178217
>AMD's marketing is the most deceptive bullshit I've ever seen
Hehe, Intel VLC and Nvidia "tegra scales to infinity" would like to have a word with you.

>>59180685
It's going to be a bloodbath to be honest.

Ryzen is the same crap AMD has been doing for ages, adding more cores and fucking up in real world performance.

Ryzen - the best CPU for WinRar extraction!

>>59180689
AHAHAHAHAHAH

>AM4 Motherboard you were going to buy suddenly out of stock
>decide to wait instead of pre-ordering just a cpu
Holy shit, looks like a dodged a bullet

>>59180689
I tested the two 1700x and 1800x samples against a stock 7700k and it loses some 20% in SC tests, since it's without turbo it's clocked 25-28% lower.

Doesn't look good for Kaby I'm afraid.

File: Screenshot from 01-03-2017 14:36:30.png (156KB, 1163x840px) Image search: [Google]

156KB, 1163x840px

As you can see Kabylake is literally noncompetitive even when it's clocked slightly higher, the only light in the tunnel for the 7700k is that it's overclocked 10% higher.

>>59180688
i dont think you know how to read this site at all..

we are talking about an 8c vs an 4c 90% of their benches are based on speed alone when the guys that will test it on a single and quad core setup come we will see a lot of different things..

till then its pretty stupid to compare them in that regard (to say the least)

>>59180792

No you didn't. Those CPUs aren't released yet. Stop lying, faggot.

>>59180927
>battle cruiser, UFO, Nuclear Submarine

This fucking site

>>59180969
He could be a reviewer but can't say anything out in the open yet so posting here anon is the only thing he can do right now. But outright calling him a liar is stupid.

>>59180969
>>59181015
Man you people are unable to read at all, are you? The score is already on the fucking site and comparable to anything you want.

>>59180927
>$500 vs $320 cpu comparison
ok then

>>59180688
tries to shill
doesnt realise that you can actually see the speed of the chips
http://cpu.userbenchmark.com/AMD-Ryzen-7-1700X/Rating/3915#Benchmarks

no one was using neither xfr of turbo AT ALL

and is still -3% slower than mhu 7700k

fucking idiots

>>59181057
You can test $2000 CPUs against $50 and you can still get a fair IPC comparison as you'd get against a $300 vs $300 one.

>>59180927
Yeah this isn't like Bulldozer where a 4-module CPU would have significantly worse single-core IPC than Phenom II and Core i7s but significantly faster multithreaded performance. This time, Ryzen has marginally slower single-core IPC but massively faster multithreaded performance.

File: dx11_cpu_3dmark_api_overhead_feature_test_dx11-100647720-orig.png (15KB, 1097x697px) Image search: [Google]

15KB, 1097x697px

8 cores was a meme for Bulldozer and it's no less of a meme now.

AMD needed to make a 4-core chip that defeats the 7700k, and they didn't.

>>59181097
>>59181085
IPC is not relevant though, total performance is

File: dx12_cpu_3dmark_api_overhead_feature_test_dx12-100647719-orig.png (14KB, 1097x697px) Image search: [Google]

14KB, 1097x697px

>>59181105

>>59181105
Why? Plenty of games get higher FPS using lower-clocked Broadwell-E CPUs than i7-7700K, now we basically have a much cheaper Broadwell-E platform that uses less power.

The future is clearly more than 4 cores.

File: Screenshot from 27-02-2017 14:25:31.jpg (88KB, 1238x940px) Image search: [Google]

88KB, 1238x940px

>>59181109
Eh, you sure you wanna go there?

8 cores clocked 12% lower will steamroll a 4 core with the same IPC any day of the week.

>>59181144
in synthetic benchmarks, lol

File: 1488190288070.jpg (115KB, 895x852px) Image search: [Google]

115KB, 895x852px

>>59181105
>>59181118
How about actual games instead of running 2 APIs without engine glue?

File: 51135.png (32KB, 550x450px) Image search: [Google]

32KB, 550x450px

>>59181174
Any more bright bulbs you got?

>synthetics are fine when AMD doesn't win though

>>59181183
holy shit, 4 fps!

>>59181207
So you're admitting 8 lower clocked cores run faster in games than 4 highly clocked ones?

Glad we got that clear.

>>59181215
that's what the dx12 api test shows too you fucking mouthbreather

the point is that it doesn't matter, not first because pretty much zero games use dx12, and not second because it's 4 fps

>>59181232
But 8 cores are still better even when they're clocked 1GHz lower, right?

File: image003.png (9KB, 450x228px) Image search: [Google]

9KB, 450x228px

>>59181243
hmm and with an l3 cache more than 2x larger hmmmmm im sure this has nothing to do with it

>>59181258
Oh yes the design of the chip is the problem now, lets just ignore that a larger L3 increases cache latency and directly lowers IPC.


Chips are different, especially from different vendors.

File: hqdefault.jpg (13KB, 480x360px) Image search: [Google]

13KB, 480x360px

>>59181207
its a 7700k with 500mhz at most to play vs some 8cs with 1.2ghz at most to play (800mhz to be more sane )

can you guess what will happen if they actually did an oc test on them?to lets say bring them close to the 7700k speed?

8c do to their nature they have double internal bandwidth than any 4c out there they dont need to push it hard

>>59181232
no you mongoloid its an average between many GAMES on both apis
https://www.computerbase.de/2017-02/cpu-skalierung-kerne-spiele-test/

stop trying to defy the laws of physics

>>59180505
You seem upset.

a wild 1800x appears on passmark http://cpubenchmark.net/high_end_cpus.html

http://cpubenchmark.net/compare.php?cmp%5B%5D=2966&cmp%5B%5D=2794

http://cpubenchmark.net/compare.php?cmp%5B%5D=2966&cmp%5B%5D=2874

uh oh muh 7700k