>AMD Ryzen >i5 price >with i7 preformance ITS OVER

>>58993531
i don't care. 2600k master race

>>58993771
This

>>58993531
DAMAGE CONTROL
A
M
A
G
E

C
O
N
T
R
O
L

>>58993771
2500k at 4.8GHz here, you're full of shit. I bet you can't stream while gaming, take a while to compile and encode, and not have many resources left when running multiple VMs. I want to upgrade.

>>58993771
Who the fuck cares about your obsolete trash?>>58993778

File: 1485884683922.jpg (17KB, 385x387px) Image search: [Google]

17KB, 385x387px

>mfw i bought a skylake system 3 months ago without hearing about kabylake or ryzen

>>58993825
Nigga just follow someone, anyone from the tech world.

>>58993825
Reading is your friend.

>old ass AM3+ system is on its last leg
Can't come soon enough
Intel can fuck off with their super expensive dual core CPUs

>>58993874
same. I've overclocked my 8120 within an inch of it's life, less than two weeks left

File: IMG-20170213-WA0004.jpg (270KB, 750x728px) Image search: [Google]

270KB, 750x728px

>>58993531
Intel assmad

File: 1476838684028.jpg (127KB, 789x792px) Image search: [Google]

127KB, 789x792px

>8 cores 16 threads for $320
>4 core babby i7 costs $350
>tfw got jewed

File: IMG_20170218_015854.jpg (14KB, 768x144px) Image search: [Google]

14KB, 768x144px

>>58993825
KabyLake ia just Skylake with hardware DRM. Why would you cuck yourslf like that? You've done good. I'm waiting for Skylake-E to come out to finally upgrade.

>>58993825
Sell it. Retards will pay nearly full price for Skylake CPUs on eBay. I ditched my 6700K this week for only £26 ($32) less than I paid for it brand new last May. People are dumb.

>bought 6600k at release
>krabbylake comes out with similar performance
>ryzen coming out to maybe start competing
>mfw

>>58993914
Kaby Lake also delivers more clocks/watt, which enabled the higher speeds

>>58993887
>audio died years ago
>doesn't power on if it has been disconnecter from mains for over a minute (it's not the BIOS battery)
>SATA controller only works in IDE mode, AHCI causes freezes
>Can't even OC any more

At this point it's not worth getting a new motherboard, and I can't sell the old one with a clear conscience.
Can't wait to get a brand new platform.

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

102KB, 345x497px

>Tfw bought a 6700k 4 months ago to replace my good old 8350
>have money ready for Vega, and a great CPU
>by the time Zen+ hits, I'll be ready to upgrade again and be all AMD, once again

>>58993916
This, same goes for 4790Ks. I bought mine for £250 brand new and sold it for £270 last week.

File: 1484355115105.jpg (15KB, 210x222px) Image search: [Google]

15KB, 210x222px

>tfw 3570k

>>58993941
I feel you, my system is on a shitty assrock board (990FX extreme3). had to buy tiny heatsinks to cool the VRMs

>upgrading your system because NEWTHING arrived
>system still works fine and won't need upgrades in the next 5 years
>still can play all games on max

WEW look at all u cucks

File: 1470577223510.gif (2MB, 270x279px) Image search: [Google]

2MB, 270x279px

>>58993950
>tfw live in eastern Europe
>intel and nvidia are extremely popular and maintain their resale value
>sold my 2.5 year old 4670k for 160 euros

File: Screenshot from 2017-02-18 02-09-09.png (149KB, 1125x488px) Image search: [Google]

149KB, 1125x488px

>>58993967
What a coincidence i'm using a shitty Asrock motherboard myself.

>>58993952
>tfw 3770K

I wish ivy bridge weren't housefires.

>>58994016
>AMD chipset
>Intel audio
>Qualcomm Ethernet
fucking parts bin motherboards

>>58994021
>ivy bridge
>housefire
wat? that's haslel 47xx

File: Untitled.png (340KB, 1366x768px) Image search: [Google]

340KB, 1366x768px

it just works

>>58994094
They don't call it inferno bridge for nothing famalam

>>58994103
>Sandybridge
>6GB
once programs adjust their processing demands for Kaby/Ryzen, you'll begin to notice slowdown.

>>58994052
Yep. As you can imagine the DPC latency isn't great in Windows.
I'm never buying Asrock again that's for sure.

i7-4790k here...

have an A10-5800K housefire edition acting as space-heater/HTPC...

can't wait for Ryzen

>>58993941
>>58993967
I know the pain as well. When I built my first Bulldozer I got an AsCock 970 Extreme3 and half the SATA ports wouldn't work after a while, It would refuse to boot if you went anywhere near the OC settings, coldn't use the RAM's XMP profile. Luckily I managed to sell it on eBay and got a Sabertooth 990FX

>>58993771
it's ok if u missed the 2500k mothership

>>58993771
I just replaced my 2600K. Can't say I have really noticed.

>>58994236
Happy to know I'm not the only one disappointed with Asrock.
At one point everyone was sucking their dick and saying they're basically Asus but for less money.

>>58994278
>At one point everyone was sucking their dick and saying they're basically Asus but for less money.

Can confirm that to be bullshit, cheap yes but Asus? hell no. I've never had a single Asus board fail on me in 10 years. Also had an A88M-ITX in another rig for someone else and the SATA ports died in the sense that I could only use 1 at a time

>>58994145

that's just shitty coding

File: 1469403143186.jpg (277KB, 1616x2048px) Image search: [Google]

277KB, 1616x2048px

>>58993531
>mfw

>>58994439
it won't stop them from doing it, unfortunately.

>>58994368
>>58994278
>>58994236
Fuck I don't want to hear this, the asrock taichi looks like the perfect board for me.

>>58993808
too right ol' cobber

Will AMD compete with the Atom series CPU? I need to know.

>>58994805
you mean the CPUs that are currently having clock generators burn out? No I don't Ryzen does that yet

>>58994817
Dayum. Looking to build a low power VPN server.

File: ryzen-prices.jpg (60KB, 960x474px) Image search: [Google]

60KB, 960x474px

O H N O
H

N
O

>>58994893
Looks like Intel is about to get Ryzen'd.

>>58994893
the R5 1300 could be a killer price/performance chip if you like overclocking. the 4 core Ryzen SKUs have the highest TDP per core, so they might be trying to beat up some i5s

which ryzen model would give me performance somewhat close to an i7 7700 and at what price?

not interested in gaymen or overclocking, btw

>>58994893
It's actually SIXTEEN MB ON THE 4/8 CHIPS

MFW
MFW
GONNA VROOOM ZOOOM BADABOOM

>>58994923
1400x. Just let xfr do the overclocking for you.

>>58994923
people will point to the R7 1700 by default because of the price comparisons, but there might be something closer. if you buy a good CPU cooler and keep XFR on, the 1400X might be a good option

File: told.png (143KB, 437x282px) Image search: [Google]

143KB, 437x282px

>>58993531

>>58994943
Yeah, two cores disabled off each ccx while cache goes untouched.
4mb l3 on all cores is actually absurd.

I bought an amd 64 like 2 months before Conroe came out, finally this time I won't be on the wrong side of a shake up.

>>58994943
source? I thought it was 8MB per 4 core unit. Even if the 4 core chips are binned 8 cores, the extra 8MB cache won't be accessible to the 4 working cores.

>>58994983
You forgot sweet 512kb L2@core.

File: sM77zRs.jpg (36KB, 709x765px) Image search: [Google]

36KB, 709x765px

>mfw: south american
>mfw: every Ryzen CPU will be atleast 100 USD more expensive than USA

I fucking hate not being american

>>58994923
>not interested in free performance
Fucking kill yourself, normalshit. Overclocking is literally changing one or two numbers. Are you braindead? If you can install a CPU, you can change two values in BIOS. It's not like you have to push the chips to crazy frequencies like 4.8-5ghz. All ryzen chips will be able to run at at least 4GHz.

File: 1486618153386.jpg (60KB, 609x676px) Image search: [Google]

60KB, 609x676px

>year 2011
BULDOZER IS GON B THE END OF INTEL
>release
its fucking shit, please take it off our hands for reduced price

>year 2017
RYZEN IS GON B THE END OF INTEL
>release
???????

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

13KB, 381x286px

6 cores
12 threads
3.3ghz base clock
3.7ghz turbo
XFR
performance competing with the $600 140w i7 6850k
only 65w TDP
less than $300

>>58994990
We don't know that for sure, everything below l3 could be tied to the core itself.

>>58993771
I wasnt that like from 8 years ago?

>>58995018
we knew Bulldozer's weakness before launch, the hope was for multithreaded workloads to catch on.

>2500k still going strong
Best tech purchase ever.

>>58995051
Not for long, corelet :^)

>>58995018
*shrug* my 8320 has been awesome for the last 2 years. ryzen is probably gonna be awesome.

File: 1487381456220.jpg (110KB, 437x282px) Image search: [Google]

110KB, 437x282px

>>58994978
>It Begins
That was when they announced it

Right now were about here

Is the year of the AMD desktop finally here?

>>58995018
You missed the part in 2011 where AMD outright stated bulldozer would probably not be good for consumers but pretty nice for datacenters.

Every fanboy and their mothers hyped it to shit though because MUH 8 COARS

This time though AMD is shaking all the trees and throwing out as much hype for it as they can muster. Reminds me of the athlon days.

>>58995093
AMD will crumble if Ryzen fails, this is their hail mary. they called in the savant designer and dumped every dollar they had into his program.

>>58993825
you did good.

Ryzen is going to be Bulldozer 2.0, and Kabylake is the ultimate tier of Intel Jewery (can only work on W10, marginal performance boost, lying about die size, etc...)

Skylake is the last decent intel line (but still doesn't hold a candle to Sandy Bridge).

Relax friend, you did perfectly fine

>>58995019
absolute madman

>>58995112
this. If Ryzen fails AMD as a CPU manufacturer is finished. They literally cannot aford another bulldozer level fuck up

Third worlder here (Netherlands). Should I buy ryzen for gaming?

>Queen Su-bae actually a semiconductor researcher, not a Jewish bean counter
>the Certified Shit Wrecker himself

Intel is finally getting a wake-up call, papa bless. The future is bright.

>>58993531
LMAO

AMD IS GOING TO FAIL AGAIN LIKE THEY ALWAYS DO

BE PREPARED FOR MASSIVE DISAPPOINTMENT

IT HAPPENS EVERY FUCKING TIME AND YOU DUMB FAGS KEEP HYPING IT UP

>>58994922
that's exactly what their aiming for.

>>58995158
Because it would be terrible if AMD completely fails since Intel's jewery will not stop with no competition as the past few years has proven

File: 148216071178.jpg (44KB, 604x604px) Image search: [Google]

44KB, 604x604px

>>58995158

Kek intelfags can't hold they got massively jewed

>>58995158
you should be worried then. AMD will likely die if Ryzen is shit, leaving Intel to sit on their ass until Qualcomm makes ARM a competitive platform for desktop.

>>58995169
>replying to classical bait

File: orgasm-expand.jpg (6KB, 275x183px) Image search: [Google]

6KB, 275x183px

>The feels when I can upgrade my A10-7850K

>>58995183
I knew it was bait, I see similar in pretty much every AMD thread but Intel fanboys can really be so deluded to post it unironically

File: 2500k.png (52KB, 963x377px) Image search: [Google]

52KB, 963x377px

>>58994103
mu niggu

File: Bogdanoff.jpg (49KB, 500x387px) Image search: [Google]

49KB, 500x387px

The call that saved AMD

>>58995141
Third wielder here, I wouldn't know.

>>58995211
Quick rundown?

File: bogdanoffs.jpg (286KB, 1024x807px) Image search: [Google]

286KB, 1024x807px

>>58995218
quick rundown:

-Rothschilds bow to Bogdanoffs
-In contact with aliens
-Possess psychic-like abilities
-Control france with an iron but fair fist
-Own castles & banks globally
-Direct descendants of the ancient royal blood line
-Will bankroll the first cities on Mars (Bogdangrad will be be the first city)
-Own 99% of DNA editing research facilities on Earth
-First designer babies will in all likelihood be Bogdanoff babies
-both brothers said to have 215+ IQ, such intelligence on Earth has only existed deep in Tibetan monasteries & Area 51
-Ancient Indian scriptures tell of two angels who will descend upon Earth and will bring an era of enlightenment and unprecedented technological progress with them
-They own Nanobot R&D labs around the world
-You likely have Bogdabots inside you right now
-The Bogdanoffs are in regular communication with the Archangels Michael and Gabriel, forwarding the word of God to the Orthodox Church. Who do you think set up the meeting between the pope & the Orthodox high command (First meeting between the two organisations in over 1000 years) and arranged the Orthodox leader’s first trip to Antarctica in history literally a few days later to the Bogdanoff bunker in Wilkes land?
-They learned fluent French in under a week
-Nation states entrust their gold reserves with the twins. There’s no gold in Ft. Knox, only Ft. Bogdanoff
-The twins are about 7 decades old, from the space-time reference point of the base human currently accepted by our society
-In reality, they are timeless beings existing in all points of time and space from the big bang to the end of the universe. We don’t know their ultimate plans yet. We hope they’re benevolent beings.

>>58993531
>i5 price i7 performance*
*When and if you overclock it by 1ghz. Only applies to certain workloads. Some restrictions may apply.

>>58995112
>AMD will crumble if Ryzen fails,
Its not really consumer sales that will keep AMD afloat, its enterprise, and semi-custom. They could be a profitable company with meager consumer hardware sales if they had strong enterprise sales again. Though it just so happens that some characteristics of a truly good enterprise part also apply to consumer hardware, particularly mobile hardware. You need high IPC, and you need low power per core for both enterprise and mobile. When you have a 32c/64t part, or a sub 35w mobile chip, in each scenario you need to reach compelling performance with about 5w per core, sometimes even less than that. You don't get there without a combination of arch, process, and bleeding edge power saving IP.

In the top binned desktop Ryzen chip we see a 3.6ghz base clock inside of a 95w TDP. If the uncore for this 8c/16t chip takes 15w, it leaves 10w per core to hit that 3.6ghz base clock.
In some of the lower SKUs we see base clocks around 3.3-3.4ghz, and it looks like they're pulling around 7w~ per core.
We could predict the clock scaling curve for Zen based on these data points, and it looks like Zen hits 3ghz at 5w per core.
If we assumed 5w per core, and still 15w for uncore, we'd get a 55w 3ghz 8c/16t die. 4 of which in an MCM would yield a 220w package.
Choice dies with better clock scaling could easily get it to 200w~. The top Naples SKU that we know of is 180w, and that is only a stones throw away from the napkin math here.
It'll be clocked north of 2.5ghz for sure, and thats a decent edge over the 2.1ghz 32c Skylake Xeons.

Intel has spent so long barely upgrading their processors while still having a much bigger R&D budget than AMD. I'm 100% sure that if Ryzen really is this great, Intel will implement all their improvements they've been hoarding as soon as they can to compete with what comes after Ryzen.

>>58995339
How would they improve an arch as old as theirs? They need to start from scratch the same way AMD did with Zen.

K
I
L
L

>>58995339
intel spent their R&D attempting to get into new markets
RAM, SSDs, ARM, etc

>>58995169
>>58995182

YOU DON'T GET IT

INTEL WANTS AMD TO STICK AROUND AS TOKEN COMPETITION

THAT WAY INTEL WON'T HAVE TO FACE MONOPOLY LAWS

IF AMD FAILS, INTEL WILL BE FORCED TO BREAK UP

AMD HAS NO CHANCE AND INTEL LOVES IT

>>58995367
>RAM
No.
>SSDs
Still behind fucking Samsung.
>ARM
No.

>>58995371
note I said "attempt"
their NVMe drives are the closest thing of those three to a success

>>58995339
The semiconductor industry doesn't work like that.
Intel hasn't been hording designs and keeping them from market. Intel's biggest competitor has been themselves, they're competing against their own prior gen parts for sales. They haven't been stifling R&D. From their position they just have to work much harder for each percentage performance uplift because they've been refining the same base arch for so long.
We already know what they have planned out for the next 4 years or so. Coffee Lake, Cannonlake, Ice Lake, Tiger Lake. Nothing new on the table, and these designs extend out to 2020 at least. They've got nothing radically new coming down the pipe. Nothing until they launch a new ISA or a totally new core arch on their 7nm process in 2021 or 2022.

>>58995368
Intel can and will argue that ARM is a competitor to avoid monopoly charges.

>>58995384
That's are still lagging behind samshit ones. The only decent success is Intel MIC which is okay for HPC segment.

>>58995305
>at 2.6ghz AMD would have a 25% clock advantage
>Zen isn't 25% behind Skylake in per core performance

AMD really hit a home run here, and this is only their first lot of 14nm LPP CPUs. Imagine the later revisions where they increase clocks even higher as they tune the process. Holy shit.

Why do Ryzen motherboards come with HDMi/DP/VGA if the damn cpu doesn't even come with an iGPU?

>>58995625
Socket AM4 also serves their APUs.

>>58995625
some boards might have old school onboard GPUs, or the board also works with the future APU lineup

>>58995625
Because APUs will use the same AM4 socket.

>>58995625
Because they may use the same socket for a later APU

>>58994052
>AMD chipset
>Intel audio

>What is the azalia standard?

File: file.png (172KB, 820x709px) Image search: [Google]

172KB, 820x709px

Upgrading is a meme

>>58994893
>1700
>65W TDP
Nigga how.

Might buy it or the 1700X. Maybe the 1800X if they have something else aside from better binning.

>>58995879
3ghz base clock.
Thats a big reduction in power from 3.6ghz.

>>58994989
look at
>>58994983
also
http://digiworthy.com/2017/02/17/amd-ryzen-5-1600x-benchmarks/

>>58995949
looks like an error in their chart, the source they have for that chart even says 8MB

The CCX has 4 separate L3 slices, each slice is tied to a given core.
Other cores access the L3 via a ring buffer that provides them all with an equal average latency. Similar to what AMD has implemented in the PS4 APU.
A CCX with 3 active cores would not have access to 8MB of L3 unless that slice was being written to through another slice, which is a complete waste of time, because the L3 is a victim cache.
Things only end up in L3 if they've been written to in the L2.
L2 is private, so each slice of L3 defacto belongs to a core.

There are too many discrepancies shown in these leaks, and some of them outright could be fake. It wouldn't be the first time someone altered the ID string of a processor to present it as something else to create a sensational headline and generate traffic for ad revenue.
It is also possible that software is simply identifying the hardware incorrectly.

Take everything with a grain of salt.

>>58993531
more like i3 price with i7 performance.

How much longer do I have to wait? I finally saved up enough for my first PC (about 1k) but I'm tired of waiting.

>>58996067
~2 weeks.

>>58993941
try undervolting the north bridge.
mine was crashing, then i noticed the northbridge heatsink was really hot.

>>58996067
Recent news pointing to March 2. I doubt they'd risk missing the goal they set for Q1 2017.

are there any gaming benchmarks yet?
still have a fucking Phenom II 955 OCd to 4GHz

>>58996067
Pretty sure it's before or after GDC.

File: 1448321267007.png (86KB, 630x622px) Image search: [Google]

86KB, 630x622px

>>58993531
> no iGPU
> cheaper than a processor with integrated graphics
WOW, IT'S FUCKING NOTHING
> Intel releases the same processors without iGPU
> Massively increases die size for pure CPU purposes
> sell it for around the same price as Ryzen
AMD PLACEBO

>>58994943
intel is kill.

>>58993531
>i7 performance
I'm a gaymur, if it can't get me more fps I don't care. Single core performance won't match my 7700k at 5.2 ghz.

>>58996160
>implying anyone wants an iGPU
kys intel shill

>>58995158
D A M A G E C O N T R O L
A
M
A
G
E

C
O
N
T
R
O
L

>>58996167
>5.2
How to spot a shill in three steps.

>>58995244
are they aliens?

>>58996160
>Intel
>releasing a CPU without their integrated gphx bullshit
pick none

>>58996148
Everything is before or after GDC

>>58996160
why do you want an i7 or i5 with an iGPU on a desktop?

>>58996160
>b-but if intel sold literally the same thing!
good point anon but what matters is that they don't

File: 1460833797348.png (164KB, 500x500px) Image search: [Google]

164KB, 500x500px

>>58996177
> nobody wants iGPU's
iGPU's are the most popular GPU's out there, literally, so there's a HUGE market for it.

>>58996206
> What is Xeon?
Also, if AMD proves that there's an actual good market for CPU's without integrated graphics, then Intel will just jump in and kill AMD again with superior IPC and similar prices.

>>58996230
i use quicksync to decode the bdmv i'm encoding.
about 3% faster than software decode.

>>58996273
>iGPU's are the most popular GPU's out there
because shit tons of people buy laptops and most of the remainder buy what is currently the most useful CPUs. Use your brain.

File: igpu intelcuck.png (43KB, 1386x257px) Image search: [Google]

43KB, 1386x257px

>>58996273

>>58996276
you know a dedicated GPU can do this better right?

>>58995368
>AUTISTIC SCREECHING
The post

File: beautiful_ass.jpg (262KB, 960x1280px) Image search: [Google]

262KB, 960x1280px

>>58996292
iGPU's are pretty useful on laptops anon, even if the laptop has a dedicated GPU, the integrated GPU kicks in to save battery when necessary. I agree that for desktops it's fucking useless, and the reason Intel puts it inside their desktop i7 are purely $$$$$, so hopefully Ryzen changes that.

However, remember the desktop market is not that big, it is actually slowly dying, so will AMD really change anything? Isn't the demand for this product to small? Also, don't you think that if Intel remove their iGPU's and drop the prices, AMD will be REKT again?

AMD did absolutely nothing amazing, they basically released Haswell without integrated graphics and dropped the prices.

>>58996373
>AMD did absolutely nothing amazing, they basically released Haswell without integrated graphics and dropped the prices.
We don't know anything yet. All these leaks about performance and prices are noise to me. They have their demos, but nothing too big aside from competing with Intel on multi threaded stuff. What we do know is that they want a piece of the enterprise market. Releasing 32core monsters will surely be tempting to those above.

>>58996373
Performance SKU desktops have actually been on the rise for a few years now, and the lowend desktop market is skewed from intel's perspective, since small desktops/AIOs are using laptop CPUs.
that said, it's not like AMD is ignoring the iGPU market. the APUs are launching in Q2, and intel is even rumored to be in talked with AMD over using Radeon iGPUs.
as for intel dropping prices, they've never done it before. even when AMD chips were outperforming them left and right years ago, they just dumped more money into marketing to maintain marketshare.

Anyone know when Vega or gtx refresh might come out?
Im building my first PC and I don't want to spend money on a new GPU if the next wave is just over the horizon?

>>58996448
we might get a launch window at GDC, given AMD is going

>>58996409
>the APUs are launching in Q2
Yeah, but here's the thing, do you really think their Raven Ridge APU's will be much cheaper than the current Intel CPU's? Based on the current Ryzen prices, I don't think so, the prices will probably be very similar.

Hopefully their graphics performance will be much better then Intel.

>>58996373
Its competing with 8c/16t Haswell-E*, while being clearly better in some metrics.
Ryzen has been shown head to head outperforming the Broadwell-E i7 6900K while clocked lower. Ryzen was faster in Blender, and Handbrake. CanardPC even confirmed the performance figures.
The arch has power saving performance enhancing features that intel has no equivalent to. Intel has literally nothing even remotely similar to XFR. Ryzen is flat out a HEDT competitor to intel's i7-E line being offered at $500 and lower, with a $250~ CPU taking on the $600 i7 6850K.
All on a process that trolls like you claimed would never clock over 2.8ghz
And they have mainstream Raven Ridge APUs coming out later in the year so intel will lose any hope of having an IGP advantage either.

Your concern trolling is fucking pathetic, kid.

>>58995244

Woo... That's... That's some shit...

>>58996459
APU graphics is already competitive and superior for the price to intel's current offerings, it's the CPU design that held them back. there's a good chance we see 2c/2t and 4c/4t APUs in Q2 that are set for the i3/i5 igpu market. I don't expect AMD to bother with anything more than 4 core units, but given how they've been doing APUs they'll top their SKU list with something to go after Iris Pro.

File: mbh.jpg (9KB, 200x150px) Image search: [Google]

9KB, 200x150px

>>58995244
>-You likely have Bogdabots inside you right now
BLACK HELICOPTERS

>>58996524
>Iris Pro
Vega-based iGPU and some 2gb of HBM2 will eat that alive.

>>58996195

No, they're the people who aliens go to for problem solving advice.

>>58996554
> 2GB of HBM2 in a CPU die
totally happening anon, totally happening.

>>58996195
They're beings beyond what our puny human brains can comprehend

>>58996578
Yes, it's inevitable.

>>58996524
Why wouldn't AMD have SMT on the APUs?

File: XA3NPwx.png (208KB, 1002x346px) Image search: [Google]

208KB, 1002x346px

>>58996650
Its already confirmed they do.
Raven Ridge is a 4c/8t die with 12 GCN CU.

>>58996448
If you're doing 1080p, then the RX480 will feed your needs. A 1060 was fine at launch, but it has aged like milk.

>>58996578
It's only a matter of time. No, really. I don't think it's happening this or the next year, but it will probably happen.

>>58996671
Sweet, had me worried for a second

>>58996650
the flagship MIGHT be a 4c8t, but I don't see much of a point.

>>58996698
Shut the fuck up you FUD spreading cretin

>>58996698
Why would an 8 thread APU be pointless? I've been wanting exactly that for years

File: 1344689036393.jpg (84KB, 544x600px) Image search: [Google]

84KB, 544x600px

>bought an i3 6100 a year ago with a Z170 motherboard
>able to upgrade to Cannon Lake if Intel panics and releases a good lineup for once
>not too far in the hole to feel bad about upgrading to Zen or Zen+ down the line
>Might grab a cheap i5 6400 off ebay and overclock it

>>58996721
I suppose you're right, the workstation crowd needs a highish end APU. I'm not a fan of igpus for CPUS in general, I prefer to separate the thermals to a Motherboard or PCIe based GPU. Separate devices fail separately as well. I can't replace an iGPU on a CPU as easily as a motherboard or discreet card.

>>58996792
But the GPU chip itself almost never fails, when a GPU dies it's because the power supply failed in some way. Using one set of VRM MOSFETs for the CPU and iGPU cuts down on points of failure.

>>58996810
it's my design autism
I prefer modularity over elegance

>>58996762
>i5 6400
>overclock it
Not so fast, goy.

>>58996833
BLCK OCing. :^)

>>58996833
Do you live in a hole?

File: mia.jpg (127KB, 606x613px) Image search: [Google]

127KB, 606x613px

> tfw i7 4770HQ in my laptop
> tfw 4c/8t since 2014
> tfw there's nothing to beat Iris Pro yet
> Great fucking battery life and godly performance
> Even handles GTA V just fine
> AMD can't even compete
Will Raven Ridge change that /g/? My CPU is 45W, base 2.2GHz, 3.4 Turbo.

>>58996838
Have fun with your 200mhz oc and constant crashes :^)

>>58993968

you're poor.

>>58996855
Ye. They can even slap some HBM2 on die for funzies.

>>58996855
>laptop
>4770HQ
>45W
>great battery life
dude how big is your battery?

>>58996855
The Iris Pro IGP in desktop Broadwell is only on par with Kaveri. Broadwell has a huge CPU perf advantage, but if you raise resolutions in games to 1080p the disparity between them evaporates. You see the reason why if you look at compute performance.
Carrizo and Bristol Ridge improve on Kaveri by a pretty big degree, if they weren't memory bottlenecked they'd massively out perform Broadwell in game benches. Thats a generic 28nm HPP part vs intel's 14nm Trigate FinFET.

Intel has nothing that can come even remotely close to Raven Ridge, their IGP arch was always god fucking awful and power hungry.

>>58996230
mostly so i can put off getting my actual gpu for later

>>58996947
an unused igpu will impact thermal performance, since it will soak up heat as well as forcing the CPU off to the side, where the thermal dissipation is less efficient.

File: 1343982373938.jpg (34KB, 340x321px) Image search: [Google]

34KB, 340x321px

>>58996855

I'd suck a spicy curry turd from her street shitting orifice

>>58996833
>>58996868
I'm currently running that i3 with a 1GHz overclock, Rabbi :^)

>>58997048
Sure thing, shill.

File: Smug+jones_17c30b_6089763.jpg (45KB, 712x447px) Image search: [Google]

45KB, 712x447px

>mfw oc'd 5820k

>>58997131
That's the one I was looking at a few years ago. It was a nice buy, but technology leaves you behind fast.

>>58993845
Can you recommend someone?

>>58996891
The biggest allowed by law.

File: 65465.png (375KB, 572x464px) Image search: [Google]

375KB, 572x464px

>>58993531
>16 motherfukcing threads
I don't give fuck nigga for 300 bucks we get bulletproof cpu for work and games.

>>58993771
2500k was objectively the better overall cpu

>>58993914
>waiting for skylake-e when ryzen exists
cuckold

File: 14231423.png (279KB, 898x790px) Image search: [Google]

279KB, 898x790px

>tfw first generation i7 920 at 4ghz
>havent given money to the Intel jew in 10 years
>runs literally anything

Finally something to upgrade from my 965 to. I seriously thought about getting a 5820k back when they were going for like $300.

>>58997205
Linus Tech Tips

>>58996160
Not a problem again poorfag :^)

>>58994686
It is the only line that is okay for asrock high-middle line.

>>58993771
>>58993808
This.

I love my 2500k that I bought 6 years ago and (((intel))) has given me no reason to upgrade it.
But AMD* is giving me a reason to upgrade, finally.
6c/12t, and less than half the power usage over the i5-500k. Now I'll be able to stream without downsampling and dropping quality and shit, if I want to.
I won't have to close down background applications because I don't have the spare resources.

It's a good upgrade. Well worth it. No reason to hang onto it now that Ryzen is coming.

>>58993825
Your own fault. I knew about Zen coming more than 2 years ago.

File: z31r7m.png (34KB, 403x402px) Image search: [Google]

34KB, 403x402px

>>58996855
Laptop guy here. I currently have a 4Ghz i7-4810MQ and I do expect Raven Ridge to compete with older Haswell laptop processors. Allowing us to have our beasts of overclockers while eating our cake on the latest mxm3.0 gfx cards.

>>58993531
>>58993825
Ha. I am staying Core II Extreme Penryn masterrace.

Why are you all chasing numbers? Is it gaymen?

>>58994094
>>ivy bridge
>>housefire
Please explain. I've got 3 Ivy bridge Thinkpads and they're all quiet as a mouse.

>>58997575
Ivy Bridge switch to TIM under a heat spreader, and temps were pretty awful for overclockers. People were seeing a 20c-30c lower temp delta from delidding

What is the performance delta between zen cores and zen cores with SMT? does SMT add an appreciable amount of performance to command the price difference?

>>58997657
All of them have SMT I believe. Maybe the rejects of the rejects, the quad cores, have it disabled.

What kind of performance upgrade can I expect going from an FX-6300 to the base Ryzen model, given it has broadwell-e ipc?

>>58995388
Most knowledgeable post in the thread except for one thing:

Intel could change some designs a bit.

For Coffee Lake, they could make a 7700k without an iGPU that they can price around $250 or less instead of $330 to compete with Ryzen. Won't come until late 2017, though, if it comes, but that would be their likely response.

I would imagine they'll cut down on their number of manufacturing processes as well. They have so many different SKUs. Dual core, dual core with HT, quad, quad with HT, and all those at so many different speeds.
Meanwhile AMD is making one single chip. They're making an 8 core, 4 per CCX, that bins down to 4 without SMT at the worst, and their 6 and 8 all have SMT.

Intel could, without too much engineering effort, at least cut the iGPU die size in half on Kaby Lake, and manufacture 6 core CPUs for roughly the same cost and a basic enough iGPU to satisfy system builders. Then they could bin those down to 4 core at the worst instead of making 4 cores binning down to 2 cores because 1 or 2 don't work or don't work well.

They don't need to design a whole new architecture (though a new Socket would likely be needed). They just need a more consumer friendly design like what AMD has done.

Intel could have been selling a 6core KL for $340 instead of a 4core all along. They are just fucking themselves with hubris.

>>58997559
>Penryn
My nigga.

>>58997683
What most people don't realize is that AMD has been killing CPU development since their beginning.

>>58996373
You know the i7-6950X doesn't have an iGPU, right?

Of course you don't know that, because you're a fucking moron.

But it's a $1700 CPU that is losing to the $389 1700X in real world applications like Blender.
And there's the $500 1800X that's even better.

Go be retarded somewhere else.

>>58995158
Apart from bulldozer all the recent AMD chips performed exceptionally well for the price point they released at

>>58996578
It was in their slides late last year. It's coming and Intel won't have anything that can compete. They might also work together with an AMD GPU similar to Crossfire but just for compute and such to lower frame draw latency.

Ryzen+Vega APUs are coming in the second half of this year. They might not have HBM initially, but it is inevitable that they eventually will as it's what Jim Keller designed.

>>58996901
Intel is apparently licensing from AMD for their integrated GPUs in the future. Will be a while until they come, though.

>>58997738
>It was in their slides late last year.
There hasn't been a single mention of Raven Ridge featuring HBM anywhere. AMD has been absolutely silent on that matter.
The only reference of an APU having HBM came from a technical document stating they had planned such a design, which is obvious, but it was from 2014 and not referencing anything specific.
The only other mentions of HBM came from the 2015 FAD, and they mentioned nothing about Raven Ridge featuring it.

http://www.anandtech.com/show/9239/amd-financial-analyst-day-2015-roundup
https://asc.llnl.gov/fastforward/AMD-FF.pdf

>>58997781
HBM will probably be Zen+

File: 18 degrees outside.jpg (106KB, 677x532px) Image search: [Google]

106KB, 677x532px

>>58997338
>920
My man.
8 years still going strong.

>>58997676
You'll be able to emulate Nintendo games on Dolphin.

>>58997789
I have my ram slots full though. Need 24gb for my vms.

>>58997726
>real world applications like Blender
and, uhh, what percentage of the world's computer buying population uses Blender?

tick, tock, tick, tock...

>>58997811
I use it for vidya if you can believe that

>>58997838
>I use it for vidya
facepalm.jpg

>>58997676
Pretty significant.

Ryzen is a about a 60% IPC increase over AM3+, and 80% better performance per watt.
But in the real world, when you're bottlenecked, it's going to feel like way more than a 60% increase when you're suddenly no longer bottlenecked.

But you'd get a much bigger increase going to like the $175 1300 over the $129 1100. More threads which are actually used now, and Ryzen's SMT is better than Intel's HT.

>>58997704
Right. they were the original pioneers. Without them (and without them being screwed over by Intel's illegal business practices enough to cripple them), Core 2 duo and sandy bridge and shit never would have existed.

So they've had a 7 year slump. Whatever. That's the same as the 7 years from 1999-2006 that they completely dominated Intel.

Now it's looking like Intel will never catch up.
Ryzen will be on 7nm in early 2019. Intel might be skipping 10nm that would have come out in 2018, and probably won't have their 7nm in mass production until 2021 or 2022. And Ryzen APUs are going to shit on Intel ones and might work together with discrete AMD GPUs to shit on everything else.
This is just the first step in Intel being the one dominated for 7 years again. And the same illegal business practices they used in 1999-2005 aren't going to work this time

File: AMD-x86-Zen-HBM-Greenland-GPU-APU-635x531.jpg (37KB, 635x531px) Image search: [Google]

37KB, 635x531px

>>58997781
>There hasn't been a single mention of Raven Ridge featuring HBM anywhere. AMD has been absolutely silent on that matter.
This sure looks like one to me. It's the font and shit. I guess it could be fake

>>58995018
3 weeks before launch we knew bulldozer was a piece of shit.

I had upgraded my motherboard and was planning on getting an 8150 but then the leaks occured and i bought a 2nd 955 (I had been planning on giving my wife my old 955).

I ended up getting an 8350 a year or so later, was a minor gaming upgrade, but a decent general computing upgrade.

File: AMD-Fast-Forward-Project-1.jpg (113KB, 840x627px) Image search: [Google]

113KB, 840x627px

>>58997781
>>58997917
But this one definitely looks real.

There's no reason why HBM on the die wouldn't work with AM4. It doesn't need to interface with the rest of the system through the socket, just directly through the CPU and integrated GPU.

>>58997874
Some intrepid individuals figured out that Zen is actually around a 50% uplift over Excavator.
If it was still just a 40% uplift like they originally targeted then it'd be around 64% higher IPC than Piledriver. So Ryzen is somewhere north of 70% higher perf/clock than Piledriver when it comes to single thread.
Passmark scores showed a 3.4ghz Ryzen chip out performing a 5.15ghz overclocked Vishera chip.

>>58997917
That very clearly isn't Raven Ridge, genius.
Thats a planned enterprise chip, one that AMD teased at an event in Japan a long time ago. They never shied away from the fact that HBM was a target for enterprise in the near future, but they have not mentioned a single word about Raven Ridge featuring it. Exactly as I stated.

>>58997925
Yet again, that isn't Raven Ridge.
That graphic is a Trinity die first of all, and exactly as I stated it isn't about any specific product.

>>58997940
sry sir

File: 1436641056588.jpg (14KB, 249x228px) Image search: [Google]

14KB, 249x228px

>>58993531
>intel been marking up prices for years since AMD can't compete
>AMD finally makes something half decent by doubling the core count with lower prices
>intel simply lowers prices and released more 8c16t chips at nearly same prices with better performances
>AMD loses again

>>58997993
You don't get it do you?
Zen has equivalent IPC- Intel has to BEAT AMD in core count to win overall.

And THEN they have to also sell it cheaper

>>58998004
Intel can already beat AMD in core count, they've been doing it for years but charging retarded prices for them since the average person doesn't need them and they can rip off retarded enthusiasts and companies.

If you think Intel has any issue beating AMD then you are delusional as fuck.

>>58997919
Now we have 200 weeks before ryzen release

>>58998021
..Are you talking about server-tier processors?
Because kek are you retarded.
Anyone can make a $4000 with 64 cores.

File: based-jim.png (105KB, 255x345px) Image search: [Google]

105KB, 255x345px

where were you when Intel got BTFO by this man?

i have i7 6700k i got lik a year 1/2 ago.. tell me what ryzen is is it good im not smart

Would it be worth upgrading from my i7-2600 (non-k version)? I wanna boost my PC's responsiveness and get some improved video/photo editing and rendering performance. Not concerned with a 1-3FPS difference for gaming.

Some people have said I'd see a marginal change but I dunno.

>>58994368
well shit now I'm worried. I have a z77 extreme4.

>>58998254
It depends what you use it for. For gayming Ryzen wouldnt really be an upgrade but if you need moar cores than Ryzen is way more competitive than Intel's current 6 and 8 core consumer processors.

>>58998305
im a GAMER whats the best cpu for me

>>58998305
Skylake-E will rekt Ryzen

>>58998313
count the number of cores on the AMD-backed PS4, this will tell you what model Ryzen to get

>>58998351
7700K is the current king of gaymes, Watch Dogs 2 is really the only game where 6 core or 8 core helps. Most games benefit slightly from more cores now but the ridiculously high clockspeed still pushes quad cores to the top.

why do cpus hav such dumb names why cant they just do like the AMD 1, Intel 2, etc. like Xbox One, PS4, etc.. like wtf for real all this shit is confusing

>>58998326
I hope it does.
Don't want AMD to get complacent either.

Why do dumb cunts actually want AMD to fail and not at least get 30-40% marketshare?

t. my last 4 CPU's have been Intel

>>58998385
i want my shekels to not be in vaine young goy

>>58998377
They'll have numbers that make sense. Take Bulldozers for example. FX-6100 is a 6 core, FX-6300 is a better 6 core. FX-8320 is an 8 core, FX-8350 is a better 8 core.

>>58998299
Bump

>>58998377
It's intentionally confusing so the average consumer can't compare them easily.

This is like tech marketing 101.

>>58993771
i handed down my 2500k to my brother and all he wants to do is to upgrade now. i don't know why you guys are so proud of owning outdated hardware that barely squeaks by.

Asrock is great you fucking nigars!

File: 1487118585256.jpg (14KB, 262x263px) Image search: [Google]

14KB, 262x263px

>>58995699
>2GB ram
have fun with that

>>58997811
What makes you think that Blender doesn't do things that other programs do?

>>58998773
Not him. 2Gb of ram works well with disk cacheing dumbass. Basically it's a web browser but still a manageable one.

>>58998021
It costs amd less to produce zen chips because they take less die space. Also the yields are better then expected. The shortest move for intel to take is to move to 10nm but it will cost lots of money. Intel will probably use it to make their chips more efficient and smaller, my guess is that performance per core will not change. At least not in the coming 2 years.

>>58998369
single core performance was expected to be worse, but Windows didn't know how to schedule for the architecture for months after launch. They had to hotfix the OS several times to get it to even use the 8 cores individually.

what the bet these fucking Ryzens costs $600+ in Australia

File: LbBIflv.gif (390KB, 500x362px) Image search: [Google]

390KB, 500x362px

> mfw
K R A K A T O A

stupid question, how hard is it to start CPU company these days? is jsut money enough?
GPUs are no go due to drivers being main osbtacle, too much legacy stuff
but CPU just execute instructions as they go

Third player in both markets would be great.

>>58997575
Laptops are fine. It's the desktop ones that are housefires.

File: ....jpg (14KB, 338x210px) Image search: [Google]

14KB, 338x210px

>>58993771
My nigga, this CPU is doing so well, I fucking love it, 350€ for for 6 years of perfect use, this means 50€ per year or 4,16€ per month.

You people need to do the math, I'm hyped for Zen but if Intel can offer another "wonder" CPU I'm in for it, fuck all you fanboys, the better CPU wins.

>>58999073
Niggah you just make it process 0s and 1s that's all there is to it. 0s and 1s.

File: Ouch.gif (410KB, 500x281px) Image search: [Google]

410KB, 500x281px

>>58999205
Lmao I fucked up the math I'm a fucking retard 350/6/12 = x.

Calculated with 300 bucks lmao

>>58999205
tbqh i will probably get the 8/16 ryzen just because it has more cores and that will make it a beast in video encoding
just look at x264 encoding, the old crappy fx-9590 beats the i5-7600k and this new ryzen would absolutely smash any intel cpu
i don't care about games that much because i would need a gtx1080 to see any improvement over an older cpu

>>58996195
Just two french science fiction producers with too much plastic surgery.

I T S O V E R
TT
S S
O O
V V
E E
R R

>>58999217
guess that's why russians, chinese and everyone else failed they've been using 2s and 3s

>>58998055
Again.

>>58998326
Lets see the prices.

>>58999073
If you don't intend to fabricate chips yourself, you need a few million dollars and a bunch of experienced engineers: http://www.adapteva.com/andreas-blog/a-lean-fabless-semiconductor-startup-model/

But that is only if you don't intend to make x86 processors, because x86 is shielded by a shitton of patents and licensing agreements that kill any new competition.

>>58993531

>still no release date

>>58993771
Screwtape proposes a Toast by CS Lewis

File: 1473740367868.jpg (41KB, 450x185px) Image search: [Google]

41KB, 450x185px

>>58999322
software tools are mighty expensive
that's what I wanted to know, thanks

>>58999289
Chinese and Russians made their own MIPS, SPARC and Alpha CPUs.

>>58999336
march the zero two
reveal GDC february the twenty eighth

>>58999370
Russia makes their own 8 core Cortex A57 ARM chips now too.

>>58999370
no where near consumer
they wanted to make it wide spread at first

>>58999378
it has 1.8v voltage regulator, on 40mm2 chip
how does it even work?

>>58993968
this.
I don't get why people buy new stuff every time if they don't really need it.

my FX8350 works fine for what I do on my computer.. I can run all my games in high on linux with a 280X.

>>58999378
There is also Elbrus, a VLIW architecture:

https://www.youtube.com/watch?v=rkh7Bs1uPg4

https://www.youtube.com/watch?v=gCqb65_Sz_I

https://www.youtube.com/watch?v=E1yyXweo2w8

>>58999403
Its Russia. The wire interconnects are all probably iron and nickle instead of copper.

File: angry driver cat.jpg (74KB, 526x567px) Image search: [Google]

74KB, 526x567px

haslel here, still see no reason to upgrade

Worth upgrading from 4790k i7?
I really just want reasons to make a mATX computer.

>>58995368
theres, *cough*... VIA...

>>58999554
VIA gave up a long time ago.

>>58999563
I remember having their chipset with my Pentium 4 cpu, it was the worst days of my life.

>>58999583
at least you didn't have nForce2

>>58999563
Nah, VIA is still trying, they're just very far behind. Their newest arch barely competes with AMD's Jaguar core, and Intel's Baytrail Atoms.

What's Ryzens single core performance vs Sky/Kaby lake?

Is it likely to be lower or equal?

>>58999661
Depending on the workload its around Haswell-Broadwell at equal clocks.

Waiting for >>58999999

Make AMD great again.

>>58995625
Raven Ridge will work on AM4 too. You won't have two different sockets for HEDT and mainstream APU's.

File: 1468905993920.jpg (43KB, 600x451px) Image search: [Google]

43KB, 600x451px

Intel on fucking suicide watch. How can they recover?

>>58999476
>24 per clock at 750mhz
>but proprietary compiler
>universal core VLIW but only with said compiler
>65nm

as far as I can guess it costs as much as xeon

>>58995084
Unless Intel actually has IPC gains up their sleeves that they were just holding until AMD started competing again. I still think Intel will be (slightly) better for gaming. A lot of games are limited to 4cores.
Unless the 1200x really proves to be insane. Like beating i5s in games that don't parallelize well. Then RIP intel.

amd wins

>>58999798
I don't think they have anything up their sleeves. The silicon semiconductor technology is finally hitting the ceiling, and if anything x86-64 is going to be dumped soon to make way for something more efficient and exotic, maybe some kind of asynchronous dataflow architecture since Intel bought Altera.

>>58999926
Daamn, but please don't let it be in 5 years, I need to hold on to x86-64.

>>58999774
It's pretty expensive, yes. The production volume is pretty low compared to anything made by Intel though.

>>58999926
I hope this doesn't mean that intel will lock everyone out of newer arch by filing patents and then deals with OEM's that only allow the use of the new incompatible arch.

>>59000450
this would only happen with laptop and mobile chips cos desktop OEMs are dying desu

File: 1486845941922.jpg (36KB, 480x662px) Image search: [Google]

36KB, 480x662px

>>58993531
>Ryzen's low price processers to compete with i3 are all quad core

>>58993921
Where's the face, retard!

File: 1487323338434.jpg (187KB, 1920x1080px) Image search: [Google]

187KB, 1920x1080px

>>58993825
Me too.
I had to do it because my entire computer broke down though (thanks to my PSU). TFW no 8 core 16 threads goodness

>>58997375
this.

also Anandtech

>>58993825
You do not need the best of everything.

>>58993531
>i5 price
>with i7 preformance
don't they have literally the same performance per core? and the only difference is HT

>>58995121
>Ryzen is going to be Bulldozer 2.0

intelshills are still delusional, I see...

>>59001018
Same performance per core, but if the rumors are true AMD is selling 6 cores for $229 and $259.
That's i5 pricing

>>59000615
Yep. If the rumors are to be believed, Ryzen has
>4c/4t unlocked at same pricepoint as 2c/4t Intel
>4c/8t or 6c/12t unlocked at same pricepoint as 4c/4t Intel
>8c/16t unlocked at same pricepoint as 4c/8t unlocked intel

2 y/o 4690k reporting in

>>58993531
Except it's gonna draw 3x as much power and burn down your house.
I remember when people were all like
>Bulldozer will rekt Intel

>>59002367
Dual core i3s are pretty much dead, a quad core system is the bare minimum for 2017.
Hexa core systems will be the new mainstream.

>>58994021
Feels good having a Xeon and not being cucked by intel's shit consumer-grade manufacturing process.

>>58993531
Hopefully it lives up to all the hype. I gotta get rid of this fucking FX 6300/970 chipset.

>>59002458
They are 65w, 95w for the highly binned parts.

>>58996273
>Intel will cut their prices by 2/3 to compete

This has never, and will never happen.They'll resort to less direct, more nefarious attempts at squashing competition just like they did in the early 2000's. Because they simply can't compete on price point alone.

You do realize that what you're suggesting is they actually use MORE silicon and charge less money for it, right?

>>58995210

>5.9

HDD pleb detected.

>>59002471
Except this is bullshit. A dual core i3 is plenty for 90% of desktop users. Basically any desktop not intended for gayming or science stuff will work perfectly well without quad core.

>>59002446
Same here. Great CPU. It just werks and it didn't bottleneck anything so far.

>>59002646
Get over it. Dual core is dead.

>>59002653
Sure.
>Posting from my 2c/4t i5 from 2010

>>59002646
>Except this is bullshit. A dual core i3 is plenty for 90% of desktop users. Basically any desktop not intended for gayming or science stuff will work perfectly well without quad core.
t. i3-7350k babby
Get over it. Multicore has been and always will be the future and now 4c8t is well within the buyable position for most people, and for the people that it isn't, there's a 4c4t for even cheaper than your garbage 2c4t chip.

>>59002796
>i3-7350k
What the fuck were they thinking? It's literally pure trash with 180$ MSRP while hexacore Zen costs fucking 229$ and top binned quad-core is 200$.

File: ▄█▀ █▬█ █ █ █ █ █ █ ▀█▀.jpg (38KB, 450x495px) Image search: [Google]

38KB, 450x495px

>>58998055
Which time?

When I was 13, I made my first PC. I had the choice of a Thunderbird 1.33Ghz for $130 or a Pentium 4 1.5Ghz which I think was something like $190.

Well I got the Thunderbird. And with my ATI graphics card, I think it was a 9600, I was getting like 110-120FPS in Unreal Tournament. If I had that more expensive P4, it probably would have been about 25% lower framerates.

>>59003278
I'm sitting on my FX6300 for 2 years now, i'm saying the dude claiming dual cores are still in vogue is a babby that bought a 7350k and is now buttmad that there's something dramatically better coming out in a couple weeks.

>>58998254
Yes, you're not smart getting Skylake 6 months ago when Zen has been talked about for the past 2-3 years. Enjoy ur overpriced trash

>>58998299
Absolutely...?
It's a non-k, and something like the 1500 that is less than the 2600 was new should be about twice as fast in most applications now days.
Not to mention, you won't have to worry about memory bottlenecks even if you could overclock that 2600, by moving to DDR4.

Especially when it comes to video editing, the 1500 could be well more than twice as fast.
If you wanted the equivolent (not even equivolent, it's worse actually), the i7-6850k 6 core, it's $600 compared to the Ryzen that's $229.

>>58998392

File: intel btfo.jpg (39KB, 1019x358px) Image search: [Google]

39KB, 1019x358px

Pretty well known overclocker posts a picture of him and some buddies looking at a Ryzen review system with shocked expressions on their face. People have been speculating that it was meaningless sensationalism, or that the chip showed huge OC frequencies under LN2.
He later clarifies in pic related.

Ryzen has ridiculous performance at its price level.

>>58999322
>because x86 is shielded by a shitton of patents and licensing agreements that kill any new competition.

This is why Qualcomm should merge back with AMD.
They should essentially let themselves be acquired for free, in exchange for access to AMD's patents and autonomy.

AMD sold Qualcomm, their mobile CPU division, for only 65 million a few years ago. Now they're worth billions, but don't have those x86/x86-x64 patents and licensing.
Qualcomm wants to compete in the desktop space. Laptop manufacturers want to use them for windows.
Imagine ARM+VEGA SoC running Windows/Linux. Intel would just die.

>>59003443
So INTEL IS BETTER STILL AHAHAHAH AMDFAGSKEKS BTFO RIP RIP RIP RRIP

>>58999553
For gaming? No. That was a really good chip for gaming except maybe for like Total Warhammer, Cities Skyline or something else like that modded for massive cities, and so on.

Not unless you live in the south and could use the lower heat that a 1400X or 1600X would put out.

>>58995015
I know that feel, mi hermano.

>>59003469
>I might not have IOMMU and full virt support, ECC, a buttload of threads, cache, long-lived socket, cheap platform, and low power consumption, but at least I got 3% higher IPC for double to triple the price!
GHz GHz GHz GHz I CAN REACH 0.148GHz higher!!!!

t. cucks

>>58999926
>>58999940
Na I think x86-64 will be the end of the line for desktops. We'll be using it for decades.

But in the compute space? Super computers and so on, yeah there will be fancy new stuff to make specific software for.

>>59003520
Kids, not cucks. Kids that've never seen P4 housefires.

>>59003520
it's literally the second coming of p4

>>59002646
90% of people are normies, anon.

Go wander around Walmart if you want to be one of the 90%.

>>59003467
AMD sold Qualcomm some GPU IP, they did not sell an entire division of the company. When they originally made the deal they were doing almost nothing with it. It was back when they were producing their Geode chips. AMD really didn't lose anything from selling off their low power GPU IP. Qualcomm's present day GPUs are drastically far behind GCN in perf/watt, even in low TDP parts.
Regardless, Qualcomm posts revenue over $20 billion a year, far out of the range of AMD. The notion of AMD acquiring Qualcomm is so retarded its practically something out of an autistic kid's fanfiction.

>>59003520
Not to mention the shit TIM on Kabylake that makes them run 10% hotter.

>>59003587
>10% hotter.
>deliding drops temps from 90C to 65C
>"10%"
Sugoi Math Onii-chan

>>59003577
Again, I said more of a merger. Not anyone buying anyone.

File: AMD-Ryzen-5-1600X-Benchmarks-Leak-Cinebench_1.jpg (31KB, 600x450px) Image search: [Google]

31KB, 600x450px

3.63GHz Turbo(3.3 stock) Zen 6 core Cinnebench single score.

File: Untitled.png (18KB, 502x380px) Image search: [Google]

18KB, 502x380px

Haslel here, still see no reason to upgrade :^)

>>59003678
>i3
Lmao

I'd rather buy a non OCable 4 core Intel than a i3

>>59003694
Not even games use more than 2 cores nowadays.

File: Screenshot from 2015-01-08 23:03:17.png (365KB, 639x479px) Image search: [Google]

365KB, 639x479px

>>59003718
Is that how you convince yourself?

@59003743
No (You) for u

File: 1266849750319.jpg (230KB, 645x486px) Image search: [Google]

230KB, 645x486px

>>59003750
There are some games that don't even LAUNCH on twocores.

But enjoy your delusion I guess.

>>59003621
Agree with your sentiment, but I'd just note that mergers of equals are basically fictions. There's always a company that is acquired, and an acquirer.

File: 1482269324836.png (731KB, 1518x2018px) Image search: [Google]

731KB, 1518x2018px

>>59003781
what the fuck, did they censor his hands or something?

@59003781
Don't quote me or my posts ever again.