Ryzen

>>59132958

>I'll be mostly gaymin'


then wait for the R5 1600X

Wait until next week then read reviews.

Alternatively buy a 1700 now and gain lots of upboats on /g/

>>59132958
Haha oh wow very good use of that money senpai. It's not like these higher end processors were meant to run vms or graphics design or anything actually efficient for the power haha. Yep just get it for your gayming rig my dude :]

>>59132983
can't wait
I already bought DDR4 RAM and I don't want to wait™ for a few more months or however long it's going to take for the 1600x to launch

If you can wait until the 28th for reliable reviews then you might as well hold off for now.

>>59132996
So more than 4 cores is not worth for gaming?

>>59133044
I want to there is a very limited stock of Ryzen where I live. If I wait until the 28th to pre order I'll be wait until april when they restock

>>59133045

Even with Intel's 6-8 core CPU's, you see negligible changes in gaming fps. Maybe 5%. Give me a bit and I'll find some benchmarks.

A solid i7-k is all one needs to get the most out of a GPU.

>>59133057
You'll be better off being patient.

File: BFxT_LaCQAA9xzV.jpg large.jpg (71KB, 1024x1024px) Image search: [Google]

71KB, 1024x1024px

>>59132958
The 1700 has a LOT more OC headroom than the 7700K and will see massive benefits in anything new.

>>59133157
hmm, i don't know about that
all the rumors and leaks suggest that the 1700 can only OC to 3.8 and that is with the more expensive sli motherboards

>>59132958
It's going to be faster in games that use 8 threads because it has 8 real cores

You're literally basically getting 2 i7s for the same price, don't be retarded

>>59133195
>You're literally basically getting 2 i7s for the same price
7700k is clocked at 4.2 compared to the 1700 3.0
So I'm not getting 2 i7s
I'm getting twice as many cores

>>59132958
>R7 1700 instead of the 7700K
>I'll be mostly gaming

7600k, if you do in fact want the extra cores/threads for games that use them then just get the R7 1700.

File: 1468974476103.jpg (40KB, 600x600px) Image search: [Google]

40KB, 600x600px

>>59133175
The 1700 does 4ghz on a board with decent VRMs. That is a 30% increase over it's 4ghz base.

The 7700k can do 5ghz which isn't even a 20% increase over it's base clocks.

The AM4 platform will also have much longer legs than the 1151 platform. AMD plans on supporting AM4 until DDR5.

>>59133219
It's obviously going to have a lower base clock yes, but we'll have to wait and see how well it overclocks

>>59133250
>>59133219
That's not to mention the temps on the 7700k with an OC because Jewtel put shitty TiM for the heat spreader instead of soldering it

>>59133224
>4c/4t
>2017

File: Avuk3kgCEAESDuj.jpg large.jpg (47KB, 849x656px) Image search: [Google]

47KB, 849x656px

>>59133264
4ghz all cores and AMD allows for per core and program profile overclocking so you can disable the extra cores on things that don't benefit from it and use the TDP headroom to OC the remaining cores further.

>>59133250
I'm not buying an expensive motherboard. If I do get the ryzen I'm getting this http://www.gigabyte.us/Motherboard/GA-AB350-GAMING-3-rev-10
And I can't OC to 4.0GHz on this, I'll be really lucky if I get 3.8

>>59133264
see my previous post about the leaks on OC so far
it's not looking good

>>59133224
Come on now
http://www.kitguru.net/components/cpu/luke-hill/intel-core-i7-7700k-i5-7600k-kaby-lake-cpu-review/7/

>>59133045
it depends on the game. old games are almost all single-threaded so in order to get the maximum out of them you'll need a cpu that can reach the highest possible clock speed. modern games are increasingly becoming more geared towards utilizing multi-core CPU's, especially RTS games which see huge gains because of the thousands of effects and units being rendered and interacting with eachother all at once. certain types of games are unlikely to ever support multi-core in a meaningful way because of the way they're designed (minecraft for example has to be single-threaded to avoid race conditions).

for content creation more cores pretty much always beats any lower core system even if the clock speeds are lower. things such as 3d rendering, 3d modeling, 2d editors (photo editing), video editing, rendering, compiling or any type of software that require large numbers of raw calculations.

depending on how well ryzen chips overclock, if you are a pure gamer the sweetspot is probably going to be the R5 CPU's because they're likely going to achieve higher clock speeds than the R7's while still having a larger than average physical core count (6 vs intels 4). the gains from R5 vs R7 in games are likely going to be very small however, only noticable once you get to really high framerates (90-144+). if you do any content creation at all, you should definitely go for an R7.

again, we'll have to wait and see how well the R7 chips overclock, both compared to intel's CPU's and the R5's. if they get to 4GHz+ easily then there's little reason not to buy them, but we'll see once people benchmark them compared to 7700K's running at 5GHz+.

>>59133303
>use the TDP headroom to OC the remaining cores further
That's not how it works, retard.

>>59133175
> can only OC to 3.8 and that is with the more expensive sli motherboards
4.0 with good mobos, 3.8 with budget mobos.

>>59133312
The closer to content creating I'll be doing is reencoding anime to watch on my tv

>>59133311
>>59133303
I'm waiting for the actual reviews to see overclocking performance, one early leak said Ryzen was hitting 5ghz on Air which was obviously false unless half the cores were disabled and it was a golden chip

>>59133336
That's what I said.
3.8 on the B chipset
4.0 on the X one
I'll be getting the B

File: aeec59356686cd10e6b00c6ff8a07786.png (119KB, 765x589px) Image search: [Google]

119KB, 765x589px

>>59133325
That's exactly how it works retard. How the fuck do you think single core turbos work?

>>59133175
>all the rumors and leaks
There's only one. Stop saying "all". Some guy said he could only reach 4.1GHz on his 1700. Big fucking deal. It's one chip and it's 65w.

>>59133361
>>59133336
Why would the chipset matter? all that really matters is how the power delivery is set up

>>59133372
He said 4.0. And there's retailers offering pre OC Ryzen chips with 1700 at 3.8

>>59133385
ok let me make this as simple as possible to you
If I get the Ryzen I'll be getting this
http://www.gigabyte.us/Motherboard/GA-AB350-GAMING-3-rev-10#kf
This can't OC past 3.8 that's IF it can get to 3.8 in the first place.

File: 1458225668634.gif (2MB, 409x398px) Image search: [Google]

2MB, 409x398px

>open any ryzen thread
>crtl+f "gaming" or "games"
>100+ results
I want /v/edditors to leave.

File: c9be2e9c4443aca461fb525911215905.png (273KB, 765x589px) Image search: [Google]

273KB, 765x589px

>>59133361
Like i said before that is mostly due to inferior VRMs not being able to support all 8 cores @ 4ghz. Ryzen supports per core overclocking and application profiles so if your program can't use the extra cores just disable them and use the headroom to push the remaining cores further.

when talking about games, the big question is whether or not overclocked R7's can match a 6700K / 7700K running at around 5GHz. the R7 has a physical disadvantage regarding clockspeeds because of all the extra cores, but the question is if that really translates into less performance. with any kind of content creation the R7 is going to wipe the floor with the 7700K with ease, so it all depends on how modern games are going to handle 6+ core CPU's. judging by the way many mainstream game engines are going (getting ever more complex) it's obvious that multi-core is the way forward.

if you're not gonna upgrade for a while i'd pick the R7 any day even if just for all the extra performance with content creation, just like i would have picked an RX480 over a GTX 1060 simply because it has more VRAM. the RX480 initially performed equal or worse to a GTX 1060 but now nearly a year later we're already seeing that the RX480 beats the GTX 1060 easily.

>>59133389
>>59133408
>there's retailers
How many? Is it just one literally who or are there more? Can you stop shilling?

>>59133408
Nigga your ability to OC all depends on the chip you get, namely what voltage it takes to OC it, and the motherboard you get in how much power it can delivery and how stable the delivery is

>>59133385
You are right, but manufacturers usually don't do budget mobos with good VRMs and thick copper.

>>59133344
then just wait a few days, look out for overclocking potentional on the R7 and compare those benchmarks to a 7700K running at 5GHz. make sure you compare it on a wide variety of games (old and newer ones). the one genre which i can already tell the 7700K will lose hard is RTS games. other then that, you'll have to wait and find out.

>>59133476
>make a thread asking people to talk me into buying the 1700
>make counter arguments since we are so far discussing leaks and rumors
>FUCK OFF SHILL INTEL DRONE
How about you fuck off

>>59133500
yes, that motherboard can't do it because its VRMs are shit
why is this so hard for you to understand?

>>59133507
Depends, ASrock made a 970 chipset board for AM3+ that advertises 9xxx CPU support, I could go up to 1.5V on my FX 8300 to try for 5ghz, but I doubt my chip could handle it

>>59133539
Okay, so buy a better built B350 board, or just pony up the cash for a higher end X370 board that's around $150

>>59133507
thats not true with modern boards, the quality is the same its just that they use a lot less VRM's, capacitors and features. you can physically count the amount of capacitors and chokes surrounding the CPU socket. on budget boards it's always going to be a significantly lower amount, so overclocking won't be as good. the higher-end boards have a ton of chokes surrounding the CPU. it's very noticable.

the production quality is generally the same though (assuming you buy from a known brand like ASUS), it's not like in the old days where budget boards used garbage capacitors that would bulge or explode after 2 years of use.

the only difference between budget and high-end boards these days is the amount of (physical) features you get.

>>59133564
This is literally the only B350 board available to me.
The cheaper X370 one I can find costs around $100 more, that's extra money I'm not willing to spend.

File: 04b02ee2ef6b313d9dd849d644ee5ad2.png (174KB, 696x678px) Image search: [Google]

174KB, 696x678px

>>59133512
DESU older games will run fine on anything recent and are fairly irrelevant. What you want to look at is scaling in newer games in average FPS (not max FPS).

>>59133628
Why is max irrelevant in your opinion?
Both max and min matter a lot since those tell you if you can expect a lot of stuttering

File: 102cee2d7faf13b5744942deda834ad9.png (224KB, 799x540px) Image search: [Google]

224KB, 799x540px

>>59133619
As me and several people have already stated the 1700 will slaughter the 7700k in anything multithreaded even with a clock disadvantage (newer games are becoming more multithreaded not less). Aether of these CPUs are perfectly capable of running any current game at well over 60fps on stock clocks.

If you do need the higher clocks AMD supports per core overclocking on Ryzen and application profiles. This means you can lower the clocks on some cores and raise them on others with the extra headroom should you need it.

>>59132983
u dont have to get X, u can nonX and still oc same shit.

>>59133702
Can I in theory, which that shitty motherboard, disable 4 cores and clock the other 4 at 4.0+?

>>59133652
Because it skews results, lets say you drop 2 frames over the course of 10 minutes far below the average, the score looks like shit, same with max like scene changes or cinematic>content when max frames spike.

Frametime is more important, how long is spent on those min and max frames is what counts.

File: Aqh1ETYCIAI3GOl.jpg large.jpg (40KB, 909x585px) Image search: [Google]

40KB, 909x585px

>>59133652
Max FPS doesn't matter much. I can get a shitload of FPS in any game when staring at a wall and nothing is happening and i certainly don't need the extra FPS in those situations. What matters is your FPS when SHTF and average FPS reflect this far better than max's do.

Your best option is to look at side by side gameplay videos. There should be plenty when they come out.

File: 8dbca80acf6a4c24eceab58e2ae5716f.png (315KB, 796x563px) Image search: [Google]

315KB, 796x563px

>>59133737
yes
That's the point.
Shity motherboards fail to OC past 4ghz due to lack of power delivery.
Disabling cores lowers power draw significantly.
You can then use that extra headroom to clock the remaining cores higher.
You can set this up on a per program basis with AMD's "Ryzen master" utility.

>>59133702
>>59133826
Alright, pre-ordering the 1700. Thank you.
Is the Wraith Spire complete shit like intel stock coolers or good enough?

>>59133003
if you can't wait then it doesn't matter, just make whatever stupid decision you feel like making at the moment you dumb nigger

>>59133912
The wraith's aren't bad but hell man I don't think you'll get a 35%+ overclock out of them on a 8 core.

>>59133619
So move to america

File: 79ab0f4b2caf3d35c0135e4c66699f84.png (201KB, 799x574px) Image search: [Google]

201KB, 799x574px

>>59133912
It's better than Intel's stock and has some OC headroom (95 watt cooler on a 65 watt CPU) but if you want to really push your OCs get a nice air cooler or AIO.

>>59133912
Spire is shit, ye. Wraith max is god tier stock cooler tho.

>>59133971
Where the fuck did all these amada images come from?

>>59133969
No.

>>59133944
>>59133971
I'm getting a h110i soonish

File: amada_1024x768d.jpg (148KB, 1024x768px) Image search: [Google]

148KB, 1024x768px

>>59133997
Japan

>>59133652
>Why is max irrelevant in your opinion?
Not that guy but why would max matter attached all? There's no way of knowing if it consistenly hits it or if it's just a momentary spike. If it is a momentary spike, what good does that do anyone? If it's not a momentary spike, won't that be reflected in the average? Also, a higher max isn't an inherently good thing. It could be brief spikes of high frame rates driving the average up when the actual experience is awful.

As the avatarfagging anon said, frame times are far more important. A max, min and average figure are fairly useless.

>>59134003
>I'm getting a h110i soonish
but that isn't a master liquid 240

>>59134016
where can I download the folder?

also I thought japan was still on windows xp machines

>>59133312
>old games are almost all single-threaded
Not really.
Using several threads, one for engine loop, one for inputs, another for UI, for network etc, is an old practice in game development.
Only in AAA games though.

>>59132983
since you can disable cores on ryzen cpus, you can just make your 8 core into a 6/4/2 core cpu to allow for higher clocks.

>>59133003
>I don't want to wait
then there's no point in trying to convince you of anything, make an uneducated impulsive purchase like you would have anyway.

File: DSsBczp.jpg (159KB, 600x600px) Image search: [Google]

159KB, 600x600px

i got a Ryzen! you got a Ryzen! everoyne gets Ryzen!

File: ae5e4ecc7a501ed38ea325faaa1aa918.png (154KB, 771x696px) Image search: [Google]

154KB, 771x696px

>>59134026
http://www.mediafire.com/file/crgpy5ww8q21aa1/amada.rar

>>59134244
>>59133936
fuck off both of you
You're asking me to wait for something without a release date.

>>59134280
>not using a pomf clone
I expected better of you amandafag

>>59134307
>fuck off both of you
>You're asking me to wait for something without a release date.
Dumb nigger.

>>59134341
You have fun waiting™
I'll be enjoying my 1700 while you do that

File: e6db8c0e609d158883c268fec859bf61.png (199KB, 740x705px) Image search: [Google]

199KB, 740x705px

>>59134307
I normally use mixtape.moe but it seems to be down senpai.

>>59132958

The i7-7700k is now $ 299 and the 6700k is $260

>>>59134347

The 7700k is much better for gaming my friend.

https://youtu.be/96S9YXXexu8

>>59134448
It's been like that at micro center forever

>>59134471
if you think I'm gonna watch 1hr of that shit you're wrong

>>59134489
Okay enjoy your chinkshit cpu my friend.

>>59134471
Who listens to this little fuck.

>>59134502
I've seen that WAN report, they said fuck all about the 7700k being better for gaming

>>59134471
If you have literally nothing running alongside your games, you're right
Otherwise Ryzen is superior
And as this is PC, everyone has something running alongside their games.

File: BYwxjKNCUAAOZbK.jpg large.jpg (144KB, 1024x1024px) Image search: [Google]

144KB, 1024x1024px

>>59134502
Ryzen CPUs are made in New York senpai.

File: BZGkJMqCcAAfyaQ.jpg large.jpg (325KB, 1024x1273px) Image search: [Google]

325KB, 1024x1273px

>>59134563
>calling senpai someone that isn't your senpai

>>59134521
Yes they did actually my friend the 7700k had 40 lanes or PCIE 32 of which is dedicated for graphics. Twice that or Ryzen, for optimal 4K experience and being more future proof the 7700k is a better buy. Not only that it's now cheaper. Also Ryzen has a problem when going above 2400mhz ram speeds.

>>59133412
lets see your super computer networking cloud thingermadoo

>>59134599
Multi-GPU is a joke and Gigabyte already patched the DRAM issue.

>>59134599
>the 7700k had 40 lanes
>Max # of PCI Express Lanes 16
https://ark.intel.com/products/97129/Intel-Core-i7-7700K-Processor-8M-Cache-up-to-4_50-GHz

>Also Ryzen has a problem when going above 2400mhz ram speeds.
That has fuckall to do with Ryzen, it's ASUS motherboard problem and it's getting fixed in two months.
Not that it matter since they haven't actually said any of this during that wan show

>>59133157
>>59133175
>>59133250
>>59133336
>>59133361
>>59133737
I really don't think the 1700 should be considered at all for gaming. It's not a gaming chip, IMO. I think the 1700 is strictly for people who need cores. Depending on what the yields are like, the 1700 could be a chip that meets the bare minimum specifications to be a functional 8c16t chip at 3ghz. That means the cores might not OC AT ALL, even if you disable half of them, since there might be leaking issues, etc, on every core. The 1700 chips are the bottom of the barrel Ryzen 7. If you REALLY want an 8c chip for gaming, I'm predicting that you're better off getting the 1700X, which will be the middle ground between the bare minimum 1700 and the top shelf 1800X. At the very least, you should wait until launch to see how the 1700 works before buying one, whereas I think the 1700X is probably much safer in that regard.

All that said, I think the R5 series are going to be the real gaming monsters at great prices. 6c12t and you can have much more confidence in the quality of the remaining cores since the major defects are likely going to be limited to the two fused off cores, otherwise the chip would probably be a 1700. And that said, if the 7700k is enough for gaming right now, the R3 lineup is probably going to make budget builds much, much better. It's a good time to be a PC gamer.

I'm very interested to see what kind of monstrous APU AMD can build out of Ryzen and Polaris. I don't see any reason they couldn't build a 4c8t chip with the iGPU being close to a 470 in performance, and perhaps a full 480, since AMD seems to be developing MCM. AMD may be positioning themselves to start producing gaming chipsets that are sold as a single unit without the option of swapping parts out. That could reduce the price of a good midrange gaming PC to less than $400, or even lower. Again, it's a good time to be a PC gamer.

>>59134877
>Depending on what the yields are like, the 1700 could be a chip that meets the bare minimum specifications to be a functional 8c16t chip at 3ghz. That means the cores might not OC AT ALL
already proven false by retailers selling it with a mobo and cooler combo OC'd to 3.8Ghz
Nice FUD though

>>59134877
>I don't see any reason they couldn't build a 4c8t chip with the iGPU being close to a 470 in performance, and perhaps a full 480
Thank you, this was the most retarded I've read on /g/ all week.

>>59134877
It'll get over 60fps in every game with no issue, and gives you the extra cores should you need them or run into a game that fully utilizes 8 threads

>>59134900
>retailers picked the good 1700s for prebuilts
>this somehow changes the fact 1700 is the bottom of the barrel 8c and is not guaranteed to be able to OC at all

>>59134947
Maybe 1080p 60fps
No way in hell you'll be getting 60 at 1440p with the 1700 without some massive OC

>>59134958
>I got proven wrong
>I'll just make more FUD about the chips being as shit as intel shit binned chips
fuck off shill

>>59134980
Please, please stop

resolution has nothing to do with your CPU

>>59134900
>retailers selling it with a mobo and cooler combo OC'd to 3.8Ghz
I've seen that claim but I haven't actually seen who's selling that. It's been pointed out that while you may get a large overclock on a 1700, you need a high-end board to do it, a lot of cooling, and you might blow up your VRMs doing it. At that point, you might as well buy a 1700X with a slightly less expensive mobo and slightly less expensive cooler, get the same performance and not chance exploding your VRMs.

OR, as I said and you ignored, wait for the 1600X which, since it doesn't need voltage for two of the cores, will probably overclock higher than even the 1800X. This isn't FUD, you retard, it's realism.

>>59135037
>resolution has nothing to do with your CPU
Pirate Civ right now and test running at 1080 and 4k.
And then come back here and tell me it has nothing to do with CPU

>>59135005
Actually I agree with him. If you're not buying the XFR variants you are basically leaving it to chance that the chip may not OC much at all past advertised boost speeds. The 1700 chips are quite obviously the lowest acceptable chips but had no defects. Some might be great overclockers in the right situation but this is quite obviously a "you get what you pay for" situation. AMD is basically telegraphing it to enthusiasts with the lack of XFR branding.

File: ea74408d70eb8a05ea8cb2f6c7dbc389.png (153KB, 765x624px) Image search: [Google]

153KB, 765x624px

>>59134599
lol are you retarded? The 7700k only has 16x pcie lanes. Ryzen has 24x.

Ryzen supports DDR4 up to 3400mhz. The issues where with the bios on some early boards which the mobo manufactures themselves stated would be fixed in a bios update in a couple weeks.

>>59134877
Already proven false.
There are retailers selling 1700s OC'd to 3.8ghz and retailers do VERY conservative overclocks.

>>59134944
do you even know what an MCM is?

>>59135047
>At that point, you might as well buy a 1700X with a slightly less expensive mobo and slightly less expensive cooler, get the same performance and not chance exploding your VRMs.
If OCing a 1700 blows your VRMs it will blow them up if you try OCing the 1700 as well you mongoloid

>>59135068
Nice FUD

File: f6c.jpg (88KB, 1190x906px) Image search: [Google]

88KB, 1190x906px

>>59135062

>>59135047
>since it doesn't need voltage for two of the cores
if only you could disable up to 6 cores on the R7
oh wait YOU CAN

>>59135062
You can see how fast your CPU can run a game by running it at like 800x600

that's it's max potential for all resolutions

going from 1080p -> 2160p will cut your fps by 1/4th, or rather increase the workload on your GPU by 300%

It's why I can go to 4k in dota and still get over 100fps average with an RX 480

>>59135103
Not an argument. Higher res will require a better CPU.

>>59135100
How is that FUD? It's just common sense. It's a 65W chip with a lower base clock. Some of them are not going to be great overclockers unless you think AMD's yields were really good.

>>59135079
One of the reasons the 1700 is clocked at 3ghz and not the 3.4-3.8 of the 1700X is that the quality of the chip is such that more voltage leaks at higher frequencies. That's why some chips need more voltage than others to get the same overclock. On average, a 1700 is going to need more voltage to get to the same frequencies as a 1700X, if it even gets there in the first place.

Again, this depends on the yields. IF yields are good enough, the average 1700 could just be a 1700X that got sold as a 1700 because there weren't any chips with defects to bin to a 1700. That type of thing happened in the Phenom II days where you could get an X3 and unlock the 4th core without any problems. HOWEVER, we have no guarantee that this is going to happen, so preordering a 1700 and banking on it overclocking well is stupid when we don't have a large sample size AT ALL.

I am not saying the 1700 will not overclock. I am saying that it depends on the yields.

>>59135134
Wrong. While it is true that the increase in resolution is much more demanding of the GPU it does increase your use of CPU.
You can literally test this yourself by recording your cpu usage while testing multiple resolutions.

I have a 4670 non K.
It can't run Mankind Divided at 1080p at 60 but it can easily run on 720p.
And the bottleneck is entirely on my CPU. My GPU barely hits 60% usage while playing at 1080p

File: 1481588752272.jpg (48KB, 600x600px) Image search: [Google]

48KB, 600x600px

>>59135142
raising your res stresses your GPU not you CPU

>>59135195
What's your GPU

>>59135151
>>59135186
And all of this is 100% speculation.
Meanwhile you have retailers doing 3.8 Ghz OC on the 1700.

>>59135100
>MUH FUD
>MUH EVERY 8c RYZEN GUARANTEED OVERCLOCKERS DREAM

So what exactly will AMD do with the chips that just barely pass for 8c if not sell them as the lowest 8c model?
Are you just going to pretend they won't exist or something?

>>59135212
1070 GTX
>>59135211
wrong

>>59135122
If you refer to an earlier post here >>59134877 you'll see that it was pointed out that there's potential for every core on a 1700 to have defects that preclude any of the cores from going much higher than 3.4 or something like that. The 6c parts are less likely to have defects on the 6 functioning cores since two cores are fused off and those cores likely contained the majority of the defects. It's not JUST about the voltage, although that is a factor as well.

>>59135223
Of course it's speculation numbnuts, but it's grounded in reality.

So what, you're basing your opinion on one retailer OC'ing a 1700 to what, 100 MHz over its advertised boost? Holy shit, stop the presses. The 1700 can hit 3.8 GHz.

>>59135229
>So what exactly will AMD do with the chips that just barely pass for 8c
You'll be purchasing them in a few weeks when they launch the R5

File: 05f54730e4792fc68ccf83a0b66dbc6f.png (86KB, 799x647px) Image search: [Google]

86KB, 799x647px

>>59135151
Zeppelins yields are already confirmed to be EXCELLENT.

This is part of why they have no 4/6 coreparts on release. Not enough bad dies to cut down.

>>59135245
>that there's potential for every core on a 1700 to have defects that preclude any of the cores from going much higher than 3.4
I just ignored your baseless speculation

>>59135250
>you're basing your opinion on one retailer OC'ing a 1700 to what, 100 MHz over its advertised boost? Holy shit, stop the presses. The 1700 can hit 3.8 GHz.
3.8Ghz on ALL CORES, as opposed to single core boost

File: Mankind divided CPU benchmark.png (95KB, 1300x2173px) Image search: [Google]

95KB, 1300x2173px

>>59135236
You have no idea what you are talking about, no wonder you bought a 1070, did you also buy a G-sync display?

Here's Mankind divided CPU benchmarks, notice how an i3 and everything above it hits over 70fps with a 1080

File: 1070 Very High Quality.png (69KB, 1299x1739px) Image search: [Google]

69KB, 1299x1739px

>>59135236
And here's your 1070, hitting 50fps at Very High 1080p, because mankind divided is really poorly made

The issue is the game engine and your GPU holding you back

>>59135223
What retailers? Newegg, at least, isn't.

>>59135295
>70avg
The in-game benchmark is shit because it only tests one area that is not even open
In the actual game the area outside your apartments drops my 4670 fps to 45 while all 4 cores sit at 100% use.
You tell me what that is if not a CPU bottleneck

>>59135317
Those benchmarks are just not accurate at all. But you would need morons doing these to disable the MSAA

>>59135253
>chip runs perfectly fine at the advertised clocks all day long, i.e. it does everything we said it would do
>but it won't OC for shit
>OH WELL LET'S CUT TWO CORES OFF AND LOSE $100 BECAUSE WE'RE JUST NICE LIKE THAT

This is bordering on delusional.

File: proz_12_amd.png (79KB, 523x440px) Image search: [Google]

79KB, 523x440px

>>59135329
Your CPU bottleneck is because you don't have enough cores or threads.

This is how ALL new games will look and you are trying to recommend a CPU with fewer threads?

>>59135380
So why does it fucking run fine at 720p you colossal faggot

>>59135291
>baseless speculation
You haven't even said what "retailers" are selling overclocked systems, but you're calling bullshit on me because I'm describing to you how the binning process works? Jesus christ you're an idiot and you're the reason that everyone says "lol, AMD about to disappoint us all again" because you think I'm making baseless speculations when I make sound reasoning for ONE FUCKING SKU coming out of the Ryzen lineup probably not clocking that high. It speaks NOTHING of how good the architecture is, just that ONE SKU is probably not going to be the best choice for gaming. ALL I'm saying is to wait until launch day to see what the yields are ACTUALLY LIKE whereas you're saying "hey, my cousin Jimmy got his dick sucked for $3 and didn't even get herpes, who wants to go downtown?"

Are you braindead?

>>59135380
>4770K below the locked 6700

The benchmarks from that site are always retarded

>>59135394
Driver overhead and a lack of threads to split the load.

>>59135396
https://www.scan.co.uk/products/3xs-ryzen-7-overclocked-bundle-amd-1700-asus-b350-plus-8gb-corsair-ddr4-bequiet-dark-rock
>3.8
>on air
>on a shitty b350 motherboard
There, now fuck off shill

>>59135418
Haswell is slower than Kabylake. That 4770k is running at stock clocks. There is nothing wrong with the benchmark.

>>59135456
By like 2% at most

>>59133157
>>MASSIVE

>>59135433
>shill

What is it you think I'm shilling?

>>59132958
Buy neither, wait 6 month. Profit.

>>59135479
>What is it you think I'm shilling?
You're spreading FUD to discourage people from buying an excellent product from AMD.
I wonder what you could be shilling
Now please go bother someone else little shill

>>59135466
More like 5%
Clock for clock yes they are similar. Hower that kaby lake turbos higher and can maintain it's turbo longer.
If you OC'd the 4770k to the same clocks as the 6700's turbo the 4770k wouldnt be far behind.

>>59135433
...don't you think it's kind of a red flag that they don't say anything about the turbo or that it's an all core overclock?

That really could be that they set the turbo to 3.8

But sure, everyone else is wrong and this ambiguous web page made by a retailer that's try to sell shit to plebs is 100% legit. Unless you've got other examples?

>>59135509
Dude, I've got a 3770k.
But as much as I admire what AMD is doing with Ryzen, I'm not spending a ground in a new platform for marginal single thread performance.
Ring me when they reach 5Ghz stock.

>>59135509
...i've said twice now that the 1700X or 1600X are probably better buys than the 1700, but also that it's worth waiting to see what the 1700 is like with a larger sample size because the yields might be really good.

The issue is that if the yields are not good, people could get something they didn't really want in the 1700 and could have gotten for less money by buying the 1600X.

>>59135553
Single core reached the point of severely diminishing returns years ago. Everything for now on is going to be MOAR CORES and software is going to have to adapt to the new norm.

File: png.png (339KB, 863x808px) Image search: [Google]

339KB, 863x808px

>>59135543
>gets proven wrong once again
>"No, but surely it is not me who was wrong. Surely that company is scamming their customers on a country with much better consumer protections than US no less"
I mean just look at those low scores, I doubt this store even exists for real

>>59135573
>The issue is that if the yields are not good...
Except they are.

File: apple indians.png (299KB, 1378x1008px) Image search: [Google]

299KB, 1378x1008px

>Buy Ryzen!
but the 7700k is a better performer for gaming at a similar price point—it's simply a better value
>w-well it's good for multithreaded things!
but I can get a two (2) 8-core Xeon E5-2670's from eBay with a good motherboard and 128GB RAM for less than $700 total
>REEEEEEEEEEEEE

lol

>>59135573
The yields where already confirmed excellent. You are forgetting that a 8 core Zeppelin die is smaller than Intel's 7700k.

>>59135603
No. It's just that they can't seem to get higher clocks out of silicon.
I play a lot of flight sims, and for some reason, they all seem to be single threaded.
No use for moar cores as far as I'm concerned.

>>59135647
And they won't get higher clocks.
https://en.wikipedia.org/wiki/Dennard_scaling

>>59135620
>but the 7700k is a better performer for gaming at a similar price point—it's simply a better value
would be true if 1700 didn't exist

>>59134667

>super computer networking cloud thingermadoo
>not using 4 cores for your gaymen with windows
>not using the other 4 cores as a Linux VM for Database/Programming

i want /v/ to leave

>>59135509
>FUD FUD FUD FUD
>FAKE NEWS FAKE NEWS FAKE NEWS FAKE NEWS

Yeah, everyone who says something that isn't mindless AMD praise is a KIKE INTEL SHILL. You're very fucking clever.
Protip, faggot, just because someone says that the lowest grade of 8 core Ryzens that made the cut might not be the single fucking wisest choice for gaymen doesn't mean they shill for Intel.

Why not buy the R5 1300 unlock the extra cores and OC it to 4 GHz?

>>59135667
have you seen all of the threads posting the microcenter prices?
>b-but that doesn't count for some reason

>>59135678
But he wasn't saying that. He was saying that the yields were shit when its confirmed that they're not.

File: IMG_2541.png (2MB, 2208x1242px) Image search: [Google]

2MB, 2208x1242px

>>59135667
Except the 7700k mops the flood with ayymd cpoo

>>59135691
>Available for In-Store Pickup Only.
Yes it doesn't count for a very obvious reason
>>59135697
>4fps difference
>on a 3 old game
>mops the floor
sure thing

>>59135736
57fps vs 60fps is a game changer

>>59135697
You still posting this?

OH MY GOD THE MAX FRAMERATE IS 195. ITS LIKE THE MA WAS STILL LOADING IN SO ALL THE WORKLOAD WERE PUSHED ONTO THE 4.5-5GHZ CPU

You are actually retarded

>>59132958
Depends on types of games you play, if they take advantage of more cores you will get better performance, example battlefield

>>59135696
>He was saying that the yields were shit

He was saying you should WAIT AND SEE what the yields are like before buying a 1700 cat in the bag you retard.

>>59135696
>confirmed that they're not

Sorry if I'm not convinced by a single britcuck retailer that offers a bundle that specifies nowhere that the 3.8 isn't just a +100 MHz turbo.
Besides, didn't AMD say their turbo applies to all cores?

>>59135747
85 to 89 isn't
and since modern games actually fully utilize more than 4 cores I don't see a situation in the near future where we will actually have a 57 to 60 situation

>>59135736
>almost 20 FPS minimum difference
>intentionally ignore this
at this point I think you retards are just false flagging to make AMD supporters look worse

>>59135749
I can see how angry you are. A more expensive pajeet cpoo, that has not even been released, already lost to Intel. I actually knew 7700k would destroy anything ayymd would release

>>59135781
it would be relevant if there was any evidence of the game stuttering

>>59135736
>Yes it doesn't count for a very obvious reason
KEK thought so

>>59135816
>28 fps on amd cpu
>not stuttering

This is why everybody mocks indians

>>59135736
fuck I hate microcenter ever since they closed the Santa Clara store
yeah they were probably getting killed by the internet but goddamn it they were my go to for CPU/mobo

>>59135834
>Theres no official Drivers released to the AM4 platform.
>Also could be DDR4 memory bug

Why are you so retarded m8?

I'm on the same boat but I need to multitask ie have multiple programs running at the same time and still have the best performance

I want something that can handle 2 nox emulators botting games and watch my chinese cartoon and play overwatch without it struggling

currently have f6300, 16gb ram, gtx 1060 GC

>>59136060
7700k, no need to worry about your house catching fire and will beat all the neckbeards with their 1800x in any game

>>59136092
>heavy multitasking
>quad core

fuck off

>>59136150
1 intel core equals to 2 ayymd cores

File: 1471396486391.png (182KB, 578x691px) Image search: [Google]

182KB, 578x691px

>>59135834
Frametime is more important, how long is spent on those min and max frames is what counts. Not how low or high they are and considering the very close averages it seems the Ryzen CPU is spending less time at those two frames than the Intel one. This means that the Intel CPU is stuttering more often and for longer periods while compensating on the average with higher frame spikes. That is why the averages are so similar despite the sevier clock disadvantage.

>>59133045
battlefield 4 and frostbite engine games use 8 cores. witcher does as well

>>59136150

please educate me on the meaning of this reply. is a quad core not good for multitasking or...?

>>59135773
I think I read something about the 1700 having a single core turbo of 3.7 with an all core turbo that's lower. The information around the 1700 seems really sparse, which is why I'm trying to caution others from thinking it's a monster before we know what we can expect on average.

>single britcuck retailer
I added up the cost for the individual parts from that retailer, and it comes to £40 cheaper than the bundle does. Considering that bundles are usually cheaper due to the larger profit despite smaller margins on the individual items, I think this is highly indicative that there's a significant testing process that goes into getting those chips to 3.8, although I think that probably they ARE getting there. But we have no idea how long the B350 boards are going to last under those conditions, since there are others on the internet reporting that the top end boards have a hard time handling the voltage it takes to get a 1700 to 3.8+. Like I said earlier, if you want that kind of performance, you're probably better off getting a 1700X with a lower end board so that you don't have to worry about stressing your VRMs so much just to save $70.

File: b2ed7092b11485f88e8744283090713d.png (155KB, 765x624px) Image search: [Google]

155KB, 765x624px

>>59136194
^this
Almost everything new (the last 2 years) is using more cores. Unless you plan on keeping your new CPU for a few months at best it would be retarded to buy anything but Ryzen.

All these motherboards are all gaming oriented. Anyone know when there's going to be more motherboard choices? I heard Lisa saying that there's going to be 80 Mobos at launch.

>>59136224
against a 8 core with similar single thread performance?

no

>>59136165
Your illiterate technobabble fools no one.

just get a ps4 pro

File: 0281f23161deae675eb56f77de18cbf3.png (261KB, 696x669px) Image search: [Google]

261KB, 696x669px

>>59136252
Im personally waiting for the Asus workstation boards.

File: 1471397846624.png (165KB, 578x691px) Image search: [Google]

165KB, 578x691px

>>59136304
>middle school math is technobabble

Intel shills everyone!

File: BF4 6950x.jpg (76KB, 1089x802px) Image search: [Google]

76KB, 1089x802px

>>59136194
Honestly, the money spent getting more cores would be better spent on getting a better GPU instead.

>>59136252
>>59136326
I want to see the ITX boards and the R3s. If I can voltage control an R3 and get it to sub 25W TDP, even if it means slightly underclocking, I will be very happy.

>>59136379
Why not go with both

>>59136326
Yea man, hopefully we be seeing workstation boards for AM4 since it's geared to Gamers AND Creative customers.

>>59136299

i'm starting to think that amd is more for multitasking and intel is more for single task

looks like i will be getting amd again in the near future since I need those multitask performance

1700x is the sweetspot and will last even longer than the 2500k meme.
1700 is $80 saved but for gaymen it's only half as long lived.

>>59136753
>1600x is the sweet spot
ftfy
its much cheap plus has more than 4 cores

>>59136765
depends on actual overclocking results. If the 1600x overclocks better than 1700x/1800x it is indeed the bargain of the century.

>>59132958
>mostly gaming
buy 7350K

>>59134358
So you've decided then. Great. Delete thread pls

>>59132983
but why are they holding it back? i want a 1600x like right now. what's the business tactic here?

>>59136854
>buy 7350K
that thing sucks even when it comes to games. at least at the current price even the slowest, non-oc quad is better in 80-90% of stuff...

File: 428.jpg (56KB, 878x814px) Image search: [Google]

56KB, 878x814px

>>59136854
>2c, 4t
>$180
pathetic bait

>>59136854
fuck off intel shill

>>59133003
buy the 7700k, it's gud. you don't need more than 4 cores anyway

>>59137367
see >>59137111

>>59133372
wait, raising the clock rate doesn't change the power consumption?

Would the R7 1700 be a substantial upgrade from my i7 2700K running at 4400 giblets?

Also, is AMD's motherboard chipset that handles storage any worse than Intel's?

>>59136977
you cant get 1800x on amazon anymore and the 1700x is the best seller. so if they released the other ones now they risk bumping themselves out of the ranking

>>59137451
better get the 1600x 6 core if you were to upgrade

File: down.jpg (11KB, 260x194px) Image search: [Google]

11KB, 260x194px

>>59132983
This

>>59137406
It does. I think that anon was saying that the base power consumption of the 1700 is low, so it has a lot of overclocking headroom, which isn't really accurate. You can expect Ryzen to gobble power when you start overclocking it. The 1800X is cream of the crop; the chips are as close to perfect as they're going to get, so the voltage leak is very limited compared to the 1700 and there's only a 30W TDP difference between the two. Don't get me wrong, it's still impressive progress over what Intel's got with their octacores, I'm just trying to illustrate how voltage leak relates to binning.

File: getaloadofthisguy.jpg (10KB, 200x219px) Image search: [Google]

10KB, 200x219px

>>59135620
>>59135607

Actual shilling or just lazy copy-pasting?

>>59136977
>what's the business tactic here
It's not a business tactic; it's a logistical problem. The R5 and R3 lineups are going to sell more units than the R7 lineup just because they're cheaper and will offer more value to most consumers. AMD doesn't want to launch those lines and be unable to satisfy demand, so they have to wait until they have a large enough inventory built up to launch those product lines rather than deal with thousands of pissed off people that wait weeks, if not months for their processor.

I'm hoping AMD launches R5 and R3 later in March and I think they will, since back in December I heard rumblings of R7 launching in January. I'm thinking AMD may have pushed back the R7 launch so the launches would be tighter together so as to maintain the buzz.

>>59137575
don't understand. the 1700 uses 31% less power with 16% less clockspeed while wasting more power?

>>59133470
on that note well likely see older amd 6 and 8 cores stay somewhat relevant since they were never utilised properly

>>59137648
This
Also price gouging. The price gouging for the RX 460 (and really the whole RX 400 series lineup) was really bad because they didn't have stock built up. Supposedly GloFo isn't producing seperate 4-6 core chips, meaning that the R3 and R5 line are binned R7 chips. This way, they can have sufficient stock for all of these processor series and gauge interest/learn from mistakes while making a ton of money from enthusiasts or people wanting to upgrade. It's a pretty savvy business move actually

>>59137826
Power consumption increases exponentially with clock speeds. This is true of ALL cpus. The 7700k is consuming 220 watts + when OC'd to 5ghz

>>59137993
well thanks for that

File: 1481915065259.jpg (52KB, 932x576px) Image search: [Google]

52KB, 932x576px

>>59138057
a bit more reading if you want
https://en.wikipedia.org/wiki/Dennard_scaling

>>59137993
>220 watts
CPU-Z measuring about that much for my 7700K at 4.8GHz. Looks like I lucked out, probably karma for my 4.8GHz 2700K at 1.39V.

>>59134219
Source on that? AMD never mentioned that kind of feature. That's something really impressive, it would've been used for ad purposes.

File: b24574689cc104c17ec036c358e0d01b.jpg (30KB, 600x526px) Image search: [Google]

30KB, 600x526px

>>59138088
People don't realize that there 5ghz 7700ks are basically on a long term suicide run. 14nm silicon is going to burn out on those voltages in a couple years at best.

File: 02b8a0ed69a4e29c8cecee5d5b11776c.png (263KB, 799x549px) Image search: [Google]

263KB, 799x549px

>>59138131
http://wccftech.com/amd-ryzen-master-overclocking-utility-detailed/

>>59132958
if you want to use a Mini ITX motherboard you're fukt for now. That's what I'm waiting on.

>>59138155
jim, what have you done? also checked

>>59138155
I'm sorry, I'm kind of an illiterate, but... How do we know that disabling single cores will allow for better clocks for the ones still active?

>>59137464

From the look of it, theres a whole mess of people wanting a mid range upgrade from their 5 year old mid range PC.

Damn guys AMD really dropped the ball there for like 20 years
Who the fuck was leading it anyways?

>>59138155
>pootech

>>59138242
I'm coming from a mid-high range built 9 years ago, and that 1600x sure looks sweet.

>>59138207
more thermal/power headroom

>>59138207
We don't. It's all supposition at this point.

>>59138283
Can you be a little more specific? Are you telling me that 6 cores produce more heat than 4 cores?

But as AMD proved, you can 6/8 cores that consume in total only 95W.

>>59138304
Ok, I get that. But again, that would be a terrific feature. Why hasn't AMD talked about it?

>>59138307
>Are you telling me that 6 cores produce more heat than 4 cores?
Yes, they do.
>>59138307
>you can 6/8 cores that consume in total only 95W
Sure, but you'll hit a clock ceiling due to heat/power sooner. There's a reason why the world records held for clockspeed are on one core.

>>59138207
The idea is that if you disable two of the cores on, say, an 1800X, you've cut 25% of the voltage requirement, which will reduce voltage leak and VRM load. The VRM load is less important, but leaking voltage will increase heat AND reduce stability, so in theory you should be able to clock higher. CanardPC reported 5ghz on an engineering sample with seven cores disabled, but I'm going to reserve judgement until after the launch. I'm probably going to buy a Ryzen part at some point this year either way, though.

>>59138350
Ok, I understand, let's hope it happens.
>>59138376
>leaking voltage will increase heat AND reduce stability

So it's like any other overlclock. Why didn't AMD show this, though? I mean, you can compensate for your lower clocks and optimize your CPU for older games that use less cores, and yet nobody talked about it. Maybe it's a marketing strategy? Wait for it to be out to have the press do their job?

>>59138268

Three ass holes, then Rory Read came in to stop the bleeding, then gave control to Dr. Su who is leading a resurgance.

>>59138418
>Maybe it's a marketing strategy? Wait for it to be out to have the press do their job?
Pretty much. It's hard to talk about this kind of "strength" of the architecture without also highlighting another weakness because it involves some tweaking at the BIOS level that only a certain percentage of people are going to want to tinker around with anyway. The real enthusiasts that know how to ask these questions and wonder about this kind of stuff will get the answers they need.

>Have preorder for Ryzen
>But also have two machines with delidded 6700/K's
>Don't know if it's worth the time, money or effort to with do a new build or switch machines

Honestly, it's the more cores that gets me--video processing is a hobby of mine, and having more cores without having to pay through the nose gets me hard.

>>59136304
jew exposes itself

>>59133045
at this point no, it isn't. fast 4 cores and 8 threads (aka Ryzen 3) is all you need for gaming. anything over that is diminishing returns and you'd be better off spending the extra bucks on GPU

>>59136379
buy AMD and get both

AMADA

IS

C U T E
U
T
E

>>59132958
the 7700k runs current games well
the 1700 will run every game from now on better, and won't really do worse than the 7700k after you OC it a bit

>>59139131
I think waiting for the 6-core might be much better value, though. It's cheaper, It's still two more real cores than the i7-7700k, and would surely be able to achieve higher overclocks.

>AMD Ryzen 7 1700 Wraith Max cooler RGB

https://www.youtube.com/watch?v=425XavsUdXQ

Buying a 1700 with the MSI Mobo And 8gb ram. It won't be terrible for gaming with 8gb I assume which is okay. Super hype, it should be great with this 290 I have, since I only game at 1080p.

File: amada_99_1024x768c.jpg (66KB, 1024x768px) Image search: [Google]

66KB, 1024x768px

>>59135697
https://www.youtube.com/watch?v=2Fy87gUVisQ

Nice meme

The vid is real. THe guy is confirmed to have one.

https://www.youtube.com/watch?v=425XavsUdXQ&t=1s

>>59139512
only a street shitter would believe this looks good. god fucking dammit.

File: 1456681584195.png (223KB, 644x580px) Image search: [Google]

223KB, 644x580px

>>59139712
I personally dislike the RGB meme but you can't deny that it sells.

File: young-man-rape-two-female-sexual-predator.jpg (35KB, 644x480px) Image search: [Google]

35KB, 644x480px

>>59139512
When this RGB meme will end?

>>59136379
old game check bf1 it uses more than 4c/8t

>>59132958
Just get AMD,
intel is doing they Jewish tricks again trying to undermine AMD, just google it. Why would you support someone who obviously want to have monopoly and ass fuck you.

>>59136165
dumb 9 years old anime shitposter

>>59137993
Eh... I believe power scales linearly with clock speed, but to the square of a voltage increase.

So with the same voltage, a 30% increase in clocks, roughly increases the CPU power usage by 30%.

However, if you not stable at that speed, and you need to increase the voltage by 7%; that's a 49% increase in power in addition to the 30% increase in power from the clock speed bump.

That's why 140 Watt AMD FX chips use over 220 Watts when overclocked near 5 GHz.

>>59139891
>ntel is doing they Jewish tricks again trying to undermine AMD
No they're not. Those are unfounded rumored caused by that loser Charlie from SA. He's still salty that AMD is the only electronics manufacturer that hasn't blacklisted him.

File: not my possession fake benchmark ryzen.jpg (46KB, 920x220px) Image search: [Google]

46KB, 920x220px

>>59139688
>nice meme

>The vid is real. THe guy is confirmed to have one.

except it's not true look at what he said when i asked him if he can play with it more

>>59135691

http://www.microcenter.com/site/content/AMDRyzen.aspx

Do preorders ship on the second or arrive then? I assume ships on right?

>>59134559
That makes absolutely no sense, who does other stuff while gaming? At most what listen to a YouTube video, play a music file, but if you're thinking that someone is rendering 4k videos while gaming you're retarded. You'll need to give an example of why it would be better for to get the Ryzen instead because your example was down right stupid.

>>59139933
he has another vid of the system

>>59139859
what is this?, source

>>59134900
Isn't 3.8 ghz just 100 mhz above the boost clock, though?

>>59140140
Lots of people start processes that will take a long time to complete, and a lot of them might want to play a game while they wait.

Could be anyone that works with videos, software development, maths, or just someone that wants to record and/or stream.

Because the platform isn't for someone like you doesn't mean it isn't right for someone else.

File: Untitled.png (119KB, 800x600px) Image search: [Google]

119KB, 800x600px

>>59133157
>>59133250
>>59133303
>>59133365
>>59133416
>>59133628
>>59133702
>>59133773
>>59133826
>>59133971
>>59134016
>>59134254
>>59134280
>>59134373
>>59134563
>>59134592
>>59135069
>>59135211
>>59136165
>>59136248
>>59136326
>>59136364
>>59138086
>>59138133
>>59139688
>GLOBAL RULES NO.13: THOU SHALT NO AVATAR LIKE A FAG

>>59140289
gee, you sure fixed the shitposting by shitposting
go back to fucking reddit

>>59140249
You still didn't give me an example of a program that would utilize the cores to the point of it negatively effecting your gaming performance though.

>>59132958
I made another thread about it, ill post the op with a bit cut for space

2) we have one benchmark that was rushed as fuck and seemingly made before the press conference and they thought nda lifted that day so they posted it, it gave an interesting look at gta5
3) we got a second benchmark that had gta5 ingame with ryzen beating it
4) everyone who has the chip in hands are talking about how this is the end for manual overclocking
5) ryzen master, apparently an oc tool lets you oc per core, or disable up to 6 cores entirely.
6) there may be an issue with ram at launch, but due to my knowledge the way the info was presented, and the hyperbole of tech sites, /g/ included doesnt allow me to confirm how bit an issue or how far reaching it is
Basically so far, we know nothing but interesting shit keeps getting posted.
Looking at xfr, here is what we know, it self overclocks and has something along the lines of 20-30 sensors per cpu core to help this
Gta5 benchmark showed lower min and max frame rate but near equal average, which again is interesting, makes it seem that the 7700k at 5ghz has an average higher bounce from max to minimum and stays lower longer then amd
The ingame bench mark that claimed to be a 1700x showed 100+ fps as an average for in game.
Now here is some conjecture, that first review did not oc or really even test at all, knowing there is an oc tool along with xfr, we can assume that was base, the second one is showing us what most people who have it and are under nda are alluding to which likely had xfr or the oc tool.
So going off that, what if xfr sees the load, and clocks up the cores you use, and down clocks the ones you aren't as needed?
Possibly, you set up a profile for games/programs where you tell it how many cores this thing uses, and it shuts the unused off when that is in the foreground, again, with xfr, overclocking the 2-6 cores still in use higher then 8 core?
This honestly seems like how it works best of both worlds

>>59140355
The mainstream i7 struggles more at game streaming compared to the r7s. It was one of the demos at the ryzen event.

>>59140355
I still didn't because that's my first post in the thread.

And I did give you examples. You need me to be so specific like being a software developer at mozilla and having to recompile firefox a lot? Or you're a uploader at a torrent site and you encode videos with x264 for their internals section.

In the past these people got the Q6600 when a dual core would've been enough. Now these people get stuff like the i7 6800k/6900k, and now possibly ryzen.

They can either dedicate all cores to the task, or do other stuff while they wait, like watch a movie or play a game.

File: 004_20130125085452.jpg (82KB, 640x480px) Image search: [Google]

82KB, 640x480px

>>59140289
posting Amada pic in a AMD thread is not avatar faging fagot.

>>59136379
But anon, the AMD chips are literally cheaper than their Intel counterparts, perform at the same level in games, and are all 8c/16t. You'de have to fucking retarded to think Intel is a better deal. Are you fucking retarded?

You can buy AMD and get the same performance for less, with more cores, and a more future proof motherboard. And then you can spend more on a GPU than you could buying Intel. It's a no-brainer.

16 threads...
Imagine the amount of tabs you could have open in Firefox
Doesn't this mean we'll DEFINITELY want 8 or 16gb of ram to harness all these cores?

are emulators going to be able to use all those cores?

>>59141844
Depends on the emulator. PS2X or whatever can do (has to for some games) software rendering in the CPU and highly benefits from more cores.

File: 1481149033202.jpg (705KB, 1920x1080px) Image search: [Google]

705KB, 1920x1080px

>>59140289
>shitposting instead of just reporting and getting over it will surely make things better!

>>59141844
If you're running 16 emulators

>>59132958
Do you want to stream while gaming?

One of AMD's briefing demos showed a 7700k dropping frames when streaming + playing Dota 2. Obviously AMD set up the demo so huge grain of salt there.

That said, if you want to stream + game, more cores will be beneficial.

>>59133250
>AMD plans on supporting AM4 until DDR5.

That's an easy 20, 30 minutes of support.

1 and 1/2 more days til nda is lifted, exciting times are coming.

>>59135603
This desu

Software has to catch up, can't just keep upping the clock

>>59139688
>all those dislikes

Lol, why

>>59133045
It is. It's just that people who got i7s are massively butthurt that people can now get 6 core 12 thread processors for cheaper than their shitty 4 core i7 because they want to maintain their false sense of superiority.

Get ryzen and overclock it. You're good for the next five years, unlike i7 users.

>>59133003
>Can't wait
Then why ask? You know the correct answer is to wait, and good goy answer is to impulse buy the 7700k.

File: 1481913906167.jpg (49KB, 400x495px) Image search: [Google]

49KB, 400x495px

>>59143120
Yeah, unlike LGA1151 which definitely won't be replaced when cannonlake/coffeelake arrives

and don't forget
(Due to issues with 10 nm yields intended for Cannonlake, Intel has opted for a 2nd refinement of their 14 nm process for higher-end families)