Ryzen & Intel Discussion

>>59211541
op here, also with that said though - I think Intel will still be the main choice for the absolute top end of the gaming market

>>59211576
Yea well no shit, AMD has nothing that can beat the i7-6950X.

>>59211595
I should have specified high end rather than top end.
I'm thinking the $300-500 crowd which is still a LOT of $ to drop on a CPU just for gaming.
In that range I think Intel will maintain that dominant market and AMD will fall behind. However the 4c and 6c chips might be able to secure them a VERY good portion of the average gaming crowd assuming their price points are where people are expecting.

>>59211541
BIOS updates are coming out nigga, benches are being redone. There was an issue where it'd go into power saving mode and clock down to 1ghz or some absurdly low shit.

>>59211541
Most gamers will still be playing on AMD hardware because of the console situation. There is only a 10-15% difference in single threaded performance at this point, and it's pretty obvious AMD will attempt to leverage this situation into better performance with their MOAR COARZ through connections with developers. With the advent of DX12 and other optimized API's we could possibly see the single thread advantage turn to nothing, simply because developers for consoles will need to play nice with AMD.

>TL;DR
The "just wait" meme.

>>59211541
It wasn't that bad, I mean, it's clearly an improvement over Bulldozer bullshit, but the problem with this one is that it clocks like shit, if it was able to reach over 4.0 GHz base it would be perfect.

>>59211541
What I don't understand is why you would get the amd cpu for potential games that benefit from the amd cpu strategy right now. Those games may appear in a few years, but atm they're not available. I know it's kind of a catch 22, but it's still kind of dumb as a consumer to shop for the 'future proof' computer. Normally that concept gets laughed at in builds threads and stuff.

In my specific use case it doesn't make sense to get the new Ryzen stuff, but I don't think I have that common of a use case.

I can see it's pretty good compared to buildozer stuff. They don't really have it for gaming atm though. I would wait maybe a month or so though and then try to see where everything is once maybe the launch bugs are figured out or something new is discovered. Looks like maybe the lower core stuff might be more generally desirable.

>>59211824
Or, and the more likely scenario I believe, is I get to do shit in the background AND game at acceptable framerates.

>>59212025
because people in build threads are retards who upgrade every other year weather there is an upgrade or not, my computer is going on 8 years, while I doubt that quad cores will be obsolete by then, my 8 core will still be able to get shit done better then the quad core.

as for waiting, I intend to do that because of an rma on my gpu and I wanted to have 1 fully working computer while i'm building ryzen incase something goes tits up and i need to look for a fix. the memory shit will likely bring ryzen to around an i7, hell youtube already shows it as far closer to the 5.0ghz then may reviews do.

>>59211541
By saying those words you've tripped at least 70% of the board's filters amidst this shitstorm. Next time, consider censoring them in a way that would avoid most Regex filters.

File: average-gaming.png (72KB, 601x830px) Image search: [Google]

72KB, 601x830px

>>59211541
Yeah I agree with everything you said

If you want a high-end gaming chip, by far your best choice is the 7700k

PCGamer did a nice average of all the games they tested, the 7700k comes out on top by far especially for its price

>>59211704
This is why you should never preorder this kind of stuff.

>>59212225
But how long will that 10% difference in performance last when AMD has a majority of the gaming industry running their hardware because of consoles? By the time the next gen of consoles launch it might not amount to much. Would it be worth not having to replace your platform next upgrade cycle if games actually do use MOAR COARZ in the near future?

>>59212225
https://www.youtube.com/watch?v=BXVIPo_qbc4
gap isn't even that large lol

>>59212229
as long as you update your BIOS it's all fine

>>59212272
It's huge

A $340 intel chip is beating a $500 Ryzen chip

>>59212261
That's all speculation, but I wouldn't count on anything changing.

The major factor you have to take into account is the 7700k is a whopping $160 cheaper than the closest Ryzen and still beats it.

>>59212261
Game devs aren't exactly known for pushing the envelope.

>>59212295
But isn't it also speculation that the 7700k will stay ahead in gaming performance? I mean, I don't think there's much more to be squeezed out of a 5GHz OC'd 7700k...
Meanwhile, a number of games are only using a little over half the AMD chip right now. With AMD's connections to game developers I think it's reasonable to say there will be at minmal, SOME performance increase. 10% isn't unrealistic.

Also, 110 average FPS seems pretty acceptable to me for now.

File: IMG_4651.jpg (111KB, 404x462px) Image search: [Google]

111KB, 404x462px

>>59212280
>>59212295
And that video has a $320 AMD chip at 3.9ghz being at worst 4 frames behind a $340 Intel chip AT 5 FUCKIN GHZ

I told you, BIOS update, the guy making the video did the BIOS update. Performance improved massively. All launch reviews are pretty much null and void until the BIOS patch is installed nigga. This isn't even a thread scheduling fix either, this fixes the core parking and L3 cache issues.

christ how many times do I have to beat this into shills heads

>>59212323
Speculation is assuming anything will change at all

Keep in mind that the consoles have had 8 cores for years, and look at how PC ports are performing right now

Even if you assume a 10% performance increase for Ryzen, that means you are still paying an extra $160 for Ryzen to get the same performance as Intel.

>>59212336
That one guy is an amateur reviewer called Joker who literally sells AMD shirts.

See >>59212225 if you want to see reliable performance charts

File: file.png (123KB, 835x407px) Image search: [Google]

123KB, 835x407px

>>59212280
>>59212295
>>59212336
incase you fucks don't bother watching the video, in the first one he does these are the results he gets

>>59212363
What in the fuck are you shilling about, a 1700 is $320, a 7700K is $340. The 1700 is unlocked and will overclock.

>>59212375
The 1700 doesn't even get close to the 7700k in performance, see >>59212225

>>59212280


For gaming, which is one thing out of thousands that a processor can do.

Personally, if you have a large budget, only game, and intend on staying at 1080p, you should get a 7700k.

If you intend to do other tasks like video editing, encoding, hosting a server, streaming while gaming, the 7700k is okay if you're on a budget, but the 1800x or (personally) 1700 with an OC will be overall the better buy.

As far as I'm concerned, buying a socket 2011v3 mobo with the intention of getting anything less than a 6950x is foolish at the moment(unless you absolutely NEED quad channel memory).

Of course, once skylake/kabylake-E come out, ryzen will become far less of a concern for intel.


AMD was actually able to make a compelling product, and It's all thanks to intels attempt to segment the market for maximum profit.

Intel is quite afraid of giving the mainstream more than 4 cores/8 threads, and now that ryzen IPC has proven itself to be "close enough" to intel's, coffee lake will be coming relatively soon (Q4 2017, on 14nm) to counter AMD.

>>59212388
Sure it does once OC'd, see >>59212375

I'm gonna wait 1-2 months for bios updates and revisions and probably get 1800x for productivity reasons. 16 threadz is really good for compiling large projects and encoding stuff.

>>59211541

Very reasonable thinking. I'm surprised somebody on /g/ is capable of it.

I'd agree with everything you said. I don't understand why so many on /g/ seem overjoyed by the fact Ryzen is a bit slower than comparable intel chips in games.

We want AMD to be competitive. We want them to be in a price and performance war. At the very least you get a nice discount on the intel chip you were already going to buy.

At this point the 5 series is interesting. If it clocks higher than 7 series it should get a nice performance bump.

>>59212363
8 cores with shared FPU's and a vastly inferior arch. What's speculation right now is whether or not Microsoft's new console is using Zen.

>>59212372
>>59212388
You stupid fucking bastards, how many times do I have to explain it. The default BIOS that shipped with all Ryzen motherboards had a flaw that would put the processor into power saving mode (and thus clock down) at random times. The L3 victim cache was not functioning either.

PC Gamer used the BIOS that shipped with the motherboard. The one with the flaw. These two videos have the patched BIOS:
https://www.youtube.com/watch?v=V5RP1CPpFVE
https://www.youtube.com/watch?v=BXVIPo_qbc4

>>59212372
http://jokerproductions.com/products?page=1
I don't see any fuckin AMD shirts here, do you?

File: ryzen-photoshop.png (112KB, 722x554px) Image search: [Google]

112KB, 722x554px

>>59212390
It's not just gaming, Ryzen is bad at basically anything that isn't rendering/encoding.

>>59212425
TL;DR
The "just wait" meme is real this time.

>>59211541
Alright /g/. How much am i really losing out on by going with AMDick instead of Shilltel? (Software, relaibility, features, etc)
AMD: https://pcpartpicker.com/list/XgbRRG
Intel: https://pcpartpicker.com/list/gpHq9W

>>59212425
Watch his video with Gamers Nexus, Joker is literally selling AMD shirts.

>>59212420
>shared FPUs
No it fucking doesn't. Each core can do two 128 bit floating point SIMD operations, or one 256 bit floating point SIMD operation. Shared FPUs haven't been a thing since Steamroller.

>>59212432
So...same as that fucking 6900k, are you fucking retarded or what ?

>>59211909
First iteration of Zen is literally Nehalem of AMD. So don't worry, the actual arch is stellar, it'll certainly be gud sooner or later.

>>59212465
It's the Ryzen $500 CPU getting beat by the $340 Intel CPU yet again

>>59212465
Yeah, m8. Broadwell-E is complete garbage too.

File: file.png (112KB, 984x384px) Image search: [Google]

112KB, 984x384px

>>59212450
looks like a pretty good joke desu

>>59212449
also my usage would be gaymen while streaming media (spotify, dramatube, etc.) sometimes even video chat (oovoo, skype can suck my ass)

>>59212480
$340 Intel CPU also beats $1100 Intel CPU so it's even.

>>59212494
Intel beats Intel...I guess the choice is Intel

>>59212492
Both are gud but I'd swap 1060-6 for 480-8 (XFX one).

>>59212498
Or AMD. You can't compare quad to octacores, quads have higher clocks so they are superior in serial workloads.

File: file.png (123KB, 601x814px) Image search: [Google]

123KB, 601x814px

>>59211541
Currently games aren't optimised for multi-core CPUs. Therefore CPU's with with high single core/thread performance will run games better.

>>59212225

File: file.png (132KB, 601x830px) Image search: [Google]

132KB, 601x830px

>>59212446
This is actually true as well as:
>lol amd has no drivers
AMD still hasn't optimised their CPU's for gayming.

>>59212498
>$500 AMD = $1050 Intel
Reasonably priced AMD beats overpriced Intel... It looks like the choice is AMD.

Have any of the reviewers tried disabling cores and overclocking the remaining yet?

>>59212561
AMD it is, unless you really need additional PCI-E lanes.

>>59212561
And then back in reality, the $340 Intel beats $500 AMD

The choice is obviously the $340 Intel

My hope is that this "blunder" prepared AMD for the r5/r3 release.

>>59212458
>DDR3 1600
>No CMT
>Farily minimal instruction set
>1.75 Ghz
>Only 4 cores in a module

I mean, I think if even ONE console uses Zen, it would be a game changer compared to last time around. Bit of a difference in 8 shit console cores and Zen there.

>>59212574
No. It's not even close in heavily threaded workloads.

>>59212522
>You can't compare quad to octacores

Except you can and we did, see >>59212225

>>59212561
So Ryzen killed off the X99 platform.

>>59212574
Only in ST workloads.
>>59212576
It's basically a countdown until everyone compiles their shit optimised for Zen.

>>59212449
https://pcpartpicker.com/list/WkDQgL
swapped the PSU and monitor, Superflower Leadex Gold is alright but Seasonic's is better and doesn't have the EVGA brand premium.

>>59212587
Yeah it's not even close in 99% of applications either.

The choice is obviously the $340 Intel. Not only are you saving $160, you are also getting better performance.

>>59212594
>gaymen
Yes, you can't compare quad to octacores.
>>59212608
Nice shilling m8.

>>59212608
>99% of applications
>literally (You)

>>59212574
>$330 AMD almost as good as $340 Intel, but with double the cores and threads.

The choice is easily the $330 AMD. The multi-threaded true reality.

>>59212603
>m.2 is sata
Nah, the regular ssd is cheaper and will probably be cooler than a gumstick

>>59212612
>you can't compare quad to octacores

I guess we broke the laws of the world

File: file.png (165KB, 1435x778px) Image search: [Google]

165KB, 1435x778px

>>59212582
oh my response was thinking you were talking about Ryzen

yeah Jaguar isn't CMT and can't do the 256 bit operation like I mentioned.

>>59212628
The $330 AMD (1700) is quite horrible compared to the $340 Intel (7700k)

See >>59212225

>>59211541
I disagree. The ability to keep shit open and launch a game and not worry about performance makes ryzen superior. Plus most of the current benchmarks underrepresent ryzens power which has pretty damn good IPC performance.

The 1700 will prove to be the de facto standard choice for enthusiast grade budget gaymen and productivity to boot.

>>59212599
for it to even score this way without optimization is something people should look into. People forget that the A64 vs Netburst started out with Intel with a "lead"

File: file.png (53KB, 600x550px) Image search: [Google]

53KB, 600x550px

>>59212633
faster

>>59212650
(you).

>>59212650
Except it isn't.

See >>59212375

No the R7s really aren't great for gamers at all. They're merely good.
Not anywhere as near as the badness of bulldozer though, jesus christ.
It's not bulldozer again, by any means.

It comes closer to Phenom II again

>>59212650
The benchmark you're linking is before the BIOS update.

See >>59212425

>>59212663
Nehalem also started clunky as all hells. It's just this board is full of newshits.

>>59212666
First of all, if its sata its not even worth it. Secondly where is 2.5" drive on there

File: 1480885625668.jpg (688KB, 756x4315px) Image search: [Google]

688KB, 756x4315px

>>59212674

moar 4K

full review here:

http://hothardware.com/reviews/amd-ryzen-7-1800x-1700x-1700-benchmarks-and-review

Seriously, after reading and watching a whole bunch of reviews my main take away is that motherboards BIOS are a fucking mess right now.

ASUS Crosshair board sounds particularly bad at the moment.

>>59212689
2.5 inch drive is SATA mate, the mSATA versions just go direct to the SATA bus without the cable. M.2 is PCI-E.

http://www.anandtech.com/bench/product/1700?vs=1459

>>59212667
(retard).

File: file.png (367KB, 753x583px) Image search: [Google]

367KB, 753x583px

>>59212705

Why did not one single reviewer try to shut cores down on Ryzen, and overclock? Then turn off SMT, then overclock the RAM.

Why wasn't the minimum frame rate being higher, resulting in smoother real world game play results even if it was slightly lower average frames made a discussion point?

None of these youtubers did anything with their Ryzen chips except ram benchmarks down our throats on buggy BIOS.

>>59212650
>The $330 AMD (1700) is quite horrible compared to the $340 Intel (7700k)
(In single threaded applications, eg >>>/v/)
In multi threaded ones it's wrecked.

File: 1480527044030.jpg (450KB, 848x1680px) Image search: [Google]

450KB, 848x1680px

>>59212689

moar storage

random benches on wccucks

https://wccftech.com/amd-ryzen-7-1800x-review-processor-leak/

>>59212492
for gaming the 7700k will be significantly better than the 1700 based on the current benchmarks

>>59212650
>Sees $330 AMD at 96 fps and $340 Intel at 112 fps.

Both of them go well over 60 fps, which is still the most popular frame rate. However, one of them is more future-resistant than the other one. I wonder which one it is?

Spoiler: It's the $330 AMD. No-brainer. The multi-threaded future is inevitable.

>>59212759

Fuck that site, any one linking it is an idiot.

>>59212717
Apparently you have done no research. The Samsung 850 Evo M.2 2280 is an m.2 that uses sata bandwith. No point. Its more expensive. My life doesnt depend on a file downloading

>>59212762
See >>59212425
>>59212375
shits like you keep trickling into this thread.

>>59212762
>significantly better
No, somewhat better, not significantly better.
The difference is about the same as between kaby lake and two intel generations prior- IE very little.

>>59212774
it's 8 bucks and it's faster, your pick

>>59211541
shill, jew, goyim, btfo, rekt, suicide watch, housefires, jewtel, manchild, cuck, amdrones,etc.

>>59212797
Oy-vey, filtered.

File: 1461178864080.png (16KB, 625x479px) Image search: [Google]

16KB, 625x479px

1080p gaming "Achilles heel"

https://www.extremetech.com/gaming/245204-amds-ryzen-7-1800x-reviewed-zen-amazing-workstation-chip-1080p-gaming-achilles-heel?source=Computing

>>59212825
That's, that's pretty low.

I went for the 7600k recently after much debate. I'm upgrading from an FX6300. Did I fuck up too badly? I hear a lot of shit talking but also praise for kaby lake and I just want good gaming performance to match with my r9 290. I would have gone ryzen but there waiting to release the full lineup and I fucking hate that.

>>59212852
>Did I fuck up too badly?
Nope not at all, the i5 is a really solid CPU and has great gaming performance.

>>59212550
The difference is literally 5fps in this case... how does this matter when taken alongside the plenty of other things the chip does better?

>>59212825

Godly efficient power consumption brought to you by AMD.

Intel is proud to have the new title of "Space Heater Provider."

No really though. That 8 core's beating the shit out of the 4 core in power efficiency. AMD's probably gonna wreck the shit out of Intel in the server market, where it really counts.

>>59212852
No unless you play BF1 24/7.

>>59212869
Awesome. I was worried I'd gotten super jewed but I was getting sick of waiting and I didn't feel like buying the top of the line since I'm not a content creator in any way. I got 3 day shipping too so hopefully it gets here Monday or so

>>59212796
>faster
>2.5" has a higher max sequential write by 20 mbps
Are you sure

>>59212910
I play bf1 a lot actually, but only 1080p

>>59212852
idiot

the 1700 would have been a better choice and is faster than the 7600k is literally everything except a 5fps difference in games, 100fps vs 105fps

nice job getting jewed

>>59211636
4 core is a joke.

>>59212938
That's okay but the game LOVES threads and you'll experience stutters if they make some Aromored Kill-type DLC with gigantic maps.

>>59212949
7600k is only ~230 dorras though.

File: ryzen gaming benchmarks are a meme.png (192KB, 1249x662px) Image search: [Google]

192KB, 1249x662px

>>59212954
this. then you realize, what about intel 8 cores? that's when you realize ryzen is better than all of those 8 cores, and that the gaymen performance was a meme, because the 6950k is on par w/ ryzen for gaming performance

kiddies are getting jewed out of future proofing, 8 cores, and literal faster processing because of retarded FPS differences like pic related

>>59212954
>4 core is a joke
Is that why 4, even 2 core intel cpus are still ahead in gayme performance both from 2011/2012 and 2017 amd 8 cores?

>>59212999
>i3
>beating ryzen in anything but the most cherrypicked of cherrypicked benchmarks

nope

>>59212999
if every ryzen chip is easily pulling 100FPS in games, why should I buy intel when ryzen is objectively faster in everything else

are you really going to buy the more expensive, , slower in everything else, 4 core intel because it's pulling 115fps in counter strike rather than 105 fps

that's pretty fucking retarded my man. let me guess you own a 60hz monitor as well and cant even take advantage of that ridiculous marginal frame difference

>>59212938
You'll be fine for a while. But expect multi-threaded workloads in games to potentially stress the fuck out of your processor in the near future.

We already got stuff like Witcher 3, Battlefield 1, and Watch Dogs 2, and the list keeps on getting bigger. Also keep in mind that APIs like Vulkan and DX12 are beginning to gain ground, and that Bethesda is teaming with AMD to make multi-threaded games more of a thing.

Heck, even Intel knows that shit's going down. i5's will become 6 core cpus very soon.

>>59213052
>Todd is teaming up with AMD
So they will fix Bethesda's shitty engine?

>>59213052
Eh fuck intel in this case. Not the guy you replied to but I was debating the 1700 vs an i7. Gonna get the 1700, the wraith cooler, and the fact that it's soldered rather than the IPC meme seals the deal.

4 and 6 cores can suck my dick. I'll take 8. This ain't bulldozer.

>>59213065
Yes, AMD will fix the Bethesda's shittastic engine of suck. That seems to be the case.

>>59212990

You know something is up when the lower clocked i5 (which has less cache to boot) is losing to its i7 bigger brother.

Then again with the 9590 barely behind all other results I really don't think it matters.

>>59213094
Oh yes now they need to give it to Obsidian and get Avellone back. I want another good Fallout game.

>>59212920
Latency nigra

>>59212977
http://www.microcenter.com/product/472532/Core_i5-7600K_Kaby_Lake_420_GHz_LGA_1151_Boxed_Processor

>>59213065
Word is theyre getting Vulkan, not sure what that means entirely, for sure FO4, SSE, and ESO, will have it patched in, as well as all forthcoming games getting it too. Otherwise we should see them all get big updates to become better aware of more cores, and since its Bethesda, thatll open the more cores argument up to alot of normies.

File: 3-57-212-choose-the-right-card-and-i-will-free-3563108.png (66KB, 500x397px) Image search: [Google]

66KB, 500x397px

>>59213121
>in store only

>>59213065
They day Bethesda gets rid of the hacked together shit engine is the day Bethesda can actually make an impressive fallout game

>>59213135
Fuck gamebryo or whatever it is they use. Shit sucks. It was impressive with Oblivion in 2005, but they're still using it in 2017

>>59213135
You mean they can give said engine to Obsidian.

>>59213139
Morrowind used it

The same engine for 15 years

File: businessman.jpg (479KB, 5760x2880px) Image search: [Google]

479KB, 5760x2880px

I thought it would be important to share my experiences with all here on /g/ as I assume this is a younger demographic. I just want you to know that as someone who works in the high business world, where time is money. The Rizen CPU represents something of a total paradigm conversion. The amount of raw power I can devote to software I need in my line of work to maximize my production is incredible. I spend a lot of time in 7zip and every second counts getting those datum processed. I know some of you are here for video games or are just scrapping by with builds you can afford thanks to Mom and Dad or through your Social Security Insurance. But I also know that there are many here like me, people in the real world. Taking big risks for big rewards using enterprise software people don't even realize exist.So take my word for it, if your time is your income, if your time is important you want Rize. You want to unleash the high core world upon org and you want to dominate.

Thank you AMD, through you I will realize financial goals above and beyond anything I dreamed possible.

File: file.png (223KB, 1280x720px) Image search: [Google]

223KB, 1280x720px

>>59212911
Good onya mate >>59212938

>>59213096
>I really don't think it matters.

That's the whole fucking point. Any of these processsors will push out 120+ FPS on 1080p.

Intel wants us to buy their inferior chips because they push out an extra 15FPS in gaming, when everything else is already pushing out 100FPS.

>>59211541

Guru 3D

https://www.guru3d.com/articles-pages/amd-ryzen-7-1800x-processor-review,1.html

>>59213152
Naples when
Also 16 core workstation chips when

>>59213144
Obsidian still can't get past the general shittyness of the engine, unless they fix it or just go to a different engine. I just want something with decent physics that can allow for playable vehicles, destructible enviroments, and not shit itself when there is too many NPCs around

>>59213154
pre BIOS fix
post BIOS fix here >>59212425
At 1080p, 7700K at 5ghz vs 1700 at 3.9ghz
>>59212852
>>59212911
>>59212938

RMA

File: HYPE.png (57KB, 688x575px) Image search: [Google]

57KB, 688x575px

>>59213052
>>59212958
well I'm definitely okay with this then. I'm super excited. Will probably have to use Winshit10 but I guess I can stomach that.

>>59213212
AMD are working on something really special when it comes to workstation chips. But like most things AMD, it'll come out 2 years too late

>>59211704
>There was an issue where it'd go into power saving mode and clock down to 1ghz or some absurdly low shit.
There's literally like four different BIOS-related issues.

>>59211541
AMD FAGS ON """""NOOSE LOOKOUT"""""

>>59213121
There are very few Micro Centers in the United States. Only 25 of them. '''Twenty-fucking-five of them.''' It's only a good deal for people who live near by, being in-store only.

For the everyone else, it's a potential fucking waste of gas money and time.

When bigger retailers like Amazon, Newegg, or Jet start giving big discounts, then it'll be something to give a shit about.

>>59213212
>16-core
>not quad socket 32-core with 2TB of ram

>>59213096
>You know something is up when the lower clocked i5 (which has less cache to boot) is losing to its i7 bigger brother.
Hyper-threading really fucks up performance on Windows since games end up trying to use the hyperthreaded core.

>>59213239
Zeppelin, I know

>>59213235
you're gonna kick yourself when AMD releases its quadcores in a few months

>all those cucks just spamming their shitty benchmarks which were taken with the BIOS Bug
>ignoring the post BIOS Fix benchmarks where the 1700 beats the 7700K on a quite reasonable amount of games
And now we will have to deal with this shitty benchmark shilling for weeks or even months because AMD and their Partners couldn't do a decent release without bugs, leaving a bad first impression on day 1.

>>59213324
Not only will there be quad-cores, but also quad-cores with eight threads for only $200. A fucking steal. Hell you could even get the lowest end one for $175.

The price rumors were true about the R7 processors, so I don't see how it would be different for the R5s and R3s.

>>59213324
>>59213234
Telling me to RMA my chip? Why the fuck would I do that?

>>59213324
Well I can't really take using this old system anymore. With quite a few games coming out that I'd like to play, I don't want to wait longer and from end of March until June or something for them.

>>59213338
Man, Ryzen's actually incredible for the price. This BIOS problem shit is bringing back memories of the RX 480 launch.

AMD, why do you keep doing this to yourself?

>>59213387
>yfw they intentionally did it so that the R5 and R3 can snicker away with better scores, blindsiding Intel.

>>59213387
It is, but thanks to this stupid mistake AMD will lose a lot of marketshare because the general consumer is retarded, they just take the amount of benchmarks with the bugs favoring Intel to facevalue because it's in the majority and don't care for reasons why AMD is weaker at the moment.

Want an example?
Go to /v/ and look for yourself.
Go to any kind of cancerous community like Reddit, 9gag or whatever else and enjoy the shilling.

It's mindblowing.

>>59213378
3.9ghz 1700 within margin of error of 5ghz 7700K in BF4. Shipping BIOS was fucked, patched BIOS improves performance massively. You've only had your system for a few days, RMA mangolango

>>59213402
>R5's and R3's are actually where the performance gains show up. The R7's release is just a ruse to have Intel lower its guard

>>59213387
I think one issue with AMD is that they have alot smaller engineer and R&D team and its hard to keep up with the giant that Intel is unless they rush out products.

>>59213417
Eh when the R5 launches it won't matter too much

desu smart decision staggering the launch like this. When R5 comes out a new round of reviews will come out.

>>59213418
Why does he have to RMA the chip though?
All he has to do is update the MOBO BIOS or am I understanding this whole bug fiasco wrong?

>>59213436
He bought a 7600K

>>59213418
I don't have it yet. Also it's cheaper overall than the ryzen7 build by 60 bucks or more

>>59213378
Hey, I'm the same guy telling you about the incoming future of multi-threaded games.

As I said before, your i5 will be fine. And you might be able to upgrade to an affordable 8 core i7 in the future if Intel decides to price their shit more aggressively.

I may be partial towards AMD. But I'm not going to get all retarded if someone bought an Intel processor already. Competition's a thing.

Let's be real though. The multi-threaded future awaits us.

>>59213447
Oh, it's that dude that posted the order screenshot, didn't notice that.
Yeah, he will kick himself quite hard in the balls after the R5 will get released with quad core for around 200$.

Would definitely RMA if all he wants is gaming performance, which the R5 will perfectly deliver with 4 cores, he can go with the R7 series with the Octa Cores for Multitasking while gaming.

My 4790K is for example an amazing chip IMO, but I do notice the bottleneck when I start playing vidya while watching tv on another monitor, browsing the web on a 3rd monitor, and use the 4th monitor for something else like having a document open or some other shit to help me at vidya like a guide.

>>59213455
60 bucks for twice the cores and four times the threads? Come on mang.

>>59213461
8 core i7s use another socket. Skylake-X is confirmed for LGA2066. His best upgrade path is a 7700K and that's just SMT

>>59213488

>playing vidya
>watching tv on another monitor
>browsing the web on a 3rd monitor
>4th monitor for something else like having a document open or some other shit to help me at vidya like a guide.

Five Eyes at work

>>59213461
Yeah I feel that, I'm also sure that I could upgrade if need be. I don't see that happening in the near future, so I'm okay with it too. I wanted ryzen bad but the price point was just out of reach. I wish they launched the r5 line.

>>59213507
And here I am with "only" 2 monitors on my PC and I rarely ever use the second one

>>59213507
Well, most shit I play does not require a huge attention span.
>Grand Strategy
>Spreadsheet vidya
>Comfy basebuilding
So I can do a bit of everything.

>>59213494
>>59213455

LOL! WELL, FUCK ME THEN!

Yo, dude with the 7600k. You might actually want to snag a R7 1700 or wait for the R5 1400x. I fucking forgot that 8 core i7s use a different socket. Switch while you still can!

>>59211541
As you've said, the 6 core variant should be much better at single threaded performance.

There's also a lot of talk about bios updates etc being issues, as gaming benchmarks are all over the place depending on the reviewer.

>>59213494
Yeah, with very minimal increase in gaming. It was actually closer to 150 difference since I'd have to go with 8gb ddr4 at 2133 instead of 16. The kaby lake is like 100 bucks less.

File: 1700_vs_7700k.png (441KB, 1670x1250px) Image search: [Google]

441KB, 1670x1250px

>>59212225

>>59213535
Dude, I actually feel really limited with two monitors. I could really use a third one surprisingly.

Probably 'cause I'm an art fag.

>>59212272
>>59212280
1700 is cheaper than the 7700k and is on par with it in gaming, plus it completely destroys it outside of gaming in multithreaded applications

>>59211541
>I do think the 1700/1800 cpu's are a bit of a flop for the predominantly gamer crowd.

Why do people believe it's a "flop" for gaming? A flop is not "less than #1", it's completely failing to compete (like bulldozer). In most of the gaming benchmarks it comes within 10% (or better) of the best currently available performance, while also offering much better parallel performance on par with a $1000+ chip (which is also out preformed by the best available).

It's more than competitive as is, better on some reviewers setups (currently unfolding motherboard / UEFI update situation). There's no world where zen / ryzen is flop in any aspect. Again, look back to bulldozer if you need a reminder of what a flop really is.

>>59212465
fucking this

how are people so dumb

>>59213587
don't forget his tests where at 1080p too bro

>>59213587

Remember guys. This is Joker's benchmark if I recall correctly. He used a Gigabyte mobo and it seems everyone else who used the same mobo didn't suffer from any shitty bios problems. This is closer to being the true performance of the R7 1700.

If you're going Ryzen, get Gigabyte boards.

The i7 7700k is getting it's ass kicked in overall performance.

File: msm.png (176KB, 681x749px) Image search: [Google]

176KB, 681x749px

>>59213601
It doesn't help when the tech illiterate msm also report it as a flop.

>>59211541

>Earlier today I cancelled my 1700x / x370 prime
>general consensus that the current release Ryzen chips are not bargains for the enthusiast gamer
>I'm going to make the plunge and pick up at 7700k.

OP is a fag

>>59213626

He used a 1700 non-X at a specific stable overclock but I wouldn't go jumping and shilling gigabyte boards off the bat as if that's a relevant factor

>>59213648
Yeah, you got a point.

I'd suggest getting a Gigabyte mobo immediately, if people don't want to suffer from shitty BIOS problems right when they get their R7 as of March.

Give it a month or two, and all the mobos in general will work like they should.

>>59213594
If I actually had 3 matching monitors I would most likely use all 3. I have 3 monitors but there mismatch only use two on my main PC and sometimes its hard to read shit on the other ones due to different DPI and position. A monitor arm would probably make it better though

Ryzen turned out to be what AMD promised. No one ever said they were going to beat intel in games. If gaming is all you do, go buy a 7600k. You're still v
going to get ripped off, but whatever. Once the high clocked R5s are rolling off the production line they're going to give the i5 a run for their money.

All in all, at any price point, ryzen is a much better offering in terms of features and performance which wasn't necessarily the case with the FX8350. This cpu cost around the same as an i5 whilst falling 30%+ in performance in games.
Ryzen closed this gap to around 10% whilst shitting all over intel in any other scope at a given price point.
Also, it will be interesting to see what happens once AMD bumps up theclock speeds.

Lastly this: AMD wants to get back in the server market. This is where the big $$ is made, your gaymur shit is prestige+marketing at the most. Ryzen fucks intel up the ass in the server market, once naples is here intel will face competition they haven't had for more than 10 years.


TL;DR imma get ryzen and there is nothing intel can do about it.

>>59211541
Additionally, isn't buying a new processor today based on the ideals of the industry as it was in 2011 a little short sighted? Games and game engines won't be optimized for 4 threads and 1080p forever. Personally, I moved to 1440p in 2011 and I could never go back (general use, productivity or gaming).

>>59213677
Yeah, I'm gonna try to make sure the third one has the same height. If I remember right, monitor stands aren't too expensive on Amazon. I might get one to prepare for a third monitor.

Another route I'm thinking of going is using a triple monitor stand with two ultrawide monitors. It has potential to be fucking awesome.

>>59213731
This.

I Googled some old-as-fuck threads from over 10 years ago and found people recommending 4 cores for upcoming games. The opposition responded to them by saying dual-cores were cheaper and having 4 cores was pointless except for workstations builds.

Well. Look where we fucking are now. It's the same shit all over again.

>>59213732
4 monitor fag here.
I have 3 monitors in front of me next to each other, the 4th one is above the middle one, so my setup is the following:
>middle is a 144hz TN for vidya, fuck viewing angles for vidya who needs that shit
>left and right are the same IPS monitor model
>top middle is another IPS which I use for watching TV, will probably replace it with an actual TV and place it further back...
The only thing those 4 monitors share is that they're all 1080p and they're all Phillips.

>>59213817

Sounds like a great setup. I've been thinking about getting a TN panel. But the IPS panels look so fucking delicious.

>>59213806
yes that made sense for the time

are people using ten year old 4 cores for gaymen now?

no they are not.

The cheapest Ryzen (1700) matches the i7-7700K in gaming.

https://www.youtube.com/watch?v=BXVIPo_qbc4

File: 1463292680512.jpg (502KB, 1149x659px) Image search: [Google]

502KB, 1149x659px

>>59213694
>No one ever said they were going to beat intel in games.
If your code is written properly, it will probably be faster on the CPU with more horsepower.
Too bad programming is almost a lost art, and nearly everything is written by idiots for ignorants.

>>59213633
7700k is not a bargain by any measure
it's also a dead platform with no upgrade path

>>59213914
That looks different from the other benchmarks. Nice shilling.

File: 14709975917861.jpg (100KB, 1000x1000px) Image search: [Google]

100KB, 1000x1000px

>>59213979
nice argument

>>59213997
AMDrone BTFO.

>>59213885
Get IPS if you don't play videogames.
if you do play videogames get at least one 144hz monitor, worth it even if you don't hit 144fps.

Are there any real benchmarks being done? All I see are synthetics and gayming where it doesn't matter anyway because most of the computation is done in the GPU.

I want to see something like compiling the kernel or ffmpeg HEVC encoding.

File: 1484797146710.png (187KB, 409x409px) Image search: [Google]

187KB, 409x409px

>>59214006
nice argument

AYYMD HOUSEFIRES

AYYMDPOORFAGS CONFIRMED ON SUICIDE WATCH

>THEY FELL FOR THE RYPOO MEME

>>59214008
hevc equal to 6900k
kernel faster than 6900k

>>59214011
Nice shilling, Poojeet.

>>59213694
It's actually even better than initially expected. A year ago many were speculating zen's performance to be similiar to Haswell's with the 40% increase in IPC they promised.

>>59214029
see>>59213997>>59214011

stay mad.

File: 1486081020339.png (266KB, 980x742px) Image search: [Google]

266KB, 980x742px

>>59211541
>https://pcpartpicker.com/list/xCgrLD
Rate my Ryzen build /g/

>>59213907

Yo dude, look at this article.

http://www.tomshardware.co.uk/opinion-quad-core-upgrade,news-22774.html

Quad-cores used to be 1200 fucking dollars!

Here's a fucking snippet from said article:

"One application area that cannot quite handle the extra cores yet is gaming. Valve, developer of the Half-Life series recently explained its multi-core gaming strategy in detail and promised to bring multicore capability to its games in the first half of 2007. If Valve can translate its vision into reality, quad-core machines will be able to do stunning things with game code and enable, for example, physics simulations. Tom Leonard, one of Valve’s multi-core gurus, told me that the first generation of dual-cores were somewhat of an disappointment for game developers, which turned very quickly into hope with the Core 2 Duo and sparked excitement with the quad-core."

The video game giant, Valve, sparked the transition to quad-cores. Now it's time for the other video game giant, Bethesda, to spark the transition to octo-cores.

http://www.pcgamer.com/bethesda-partners-with-amd-to-optimize-games-for-ryzen-and-vega/

Vulkan's on its fucking way.

THE MULTI-THREADED FUTURE IS INEVITABLE.

File: 1370599568274.gif (540KB, 232x196px) Image search: [Google]

540KB, 232x196px

>>59214062
>$130 case

Get 3000 MHz RAM instead and a $60 case.

>>59213979
>>59214006
Reviewer used the Gigabyte motherboard which does not have the BIOS bug which ASUS and Co has nerfing the L2 cache and going to power saving mode cutting out 1GHz.

Read the fucking thread you retard, it has been mentioned over and over again.

>>59214007

Thanks for the advice dude. Since I'm an art fag, I thought I should get all IPS panels. But a single 144 Hz panel probably wouldn't hurt for gaming.

I also got an RX 480, so a 144 Hz FreeSync monitor would sound pretty fuckin' swell.

File: 1486480615162.jpg (60KB, 350x510px) Image search: [Google]

60KB, 350x510px

>>59214090
I would, but I've not seen anywhere from the specs that that specific motherboard would benefit from RAM higher than 2666 and it was only 99.99 after a deal for that specific one.

Plus, a case is something you should generally only buy once. Might as well go big.

>>59214098
Really? It looks like shilling, Rajeesh.

>>59214131
Trust me, you'll cum buckets just with the desktop experience, moving windows and your mouse around.
Once you go at least 120, you can never go back.

File: 1486483283466.jpg (32KB, 408x632px) Image search: [Google]

32KB, 408x632px

>>59214150
for (You)

>>59214156
The stutter from scrolling just...DISAPPEARS!

>>59214067
>THE MULTI-THREADED FUTURE IS INEVITABLE.

>promised to bring multicore capability to its games in the first half of 2007
>2007
>10 years ago
>key word future.

youd be a retard to buy raisin today. in 10 years buy whatever 8 or 10 core chip is the value. i mean 95% of shit most people run is being done better on far cheaper parts. why the fuck would you shit out money for hardware that is more but not better

>>59214150
Whatever fits your agenda the best.

>>59214149
BIOSs are buggy as shit now, I'd bet in a month or so, Ryzen will show good scaling with memory speed like Intel CPUs do.

File: 1466002997661.jpg (109KB, 960x691px) Image search: [Google]

109KB, 960x691px

>>59214024
sauce?

reading a few more reviews I want to know a few things besides fps

frametime graph at different resolutions, 8 core leverages that apparently well enough compared to 4c. Basically I want full videocard review treatment.
second thing is test with AMD card on amd drivers compared to intel cpus, i'm not convinced that nvidia drivers just not ready for new arch more than likely and happened before

select few graphs show that ryzen has much less frame jumps, it a thing i'd consider future proofing more than max fps

also I want to see more worse running games like dishonored 2 which showed improvement in fps at 1440p against 7700K.

this shit is way too complicated, and nobody had time to make complete research.

I'll post it here I want someone to explain this to me or where i'm wrong because 1. I really don't want tor eaplce 2600 with another intel 4c 2. 1700x looks promising even at current graphs

>>59213461
>incoming future of multi-threaded games
That's the same shit we've been hearing for the last 5+ years. It's not happening in this generation, and won't happen for another 12+ years.

Game developers don't give a fuck about embracing new technology. All they care about is maximizing profits, and the easiest way to do that is to use existing technology. AMD might be a pioneer in computer tech, but it means nothing when that tech doesn't get fully utilised until 12-15 years later. Look at x86-64, it's only in the last 12 months that we're seeing great improvements, and the x86-64 instruction set was introduced over 15 years ago.

AYYMD POOJEETS BLOWN THE FUCK OUT.

>>59214164

I would have agreed with you if the R7 1700 wasn't a $330 octo-core that happily trades blows with the $340 i7 7700k (Micro Center discounts don't count). This is not another Bulldozer, believe it or not.

After the shitty bios crap is dealt with Ryzen will most likely do stellar and could make buying the 7700k almost pointless without a good sale. Buying a quad-core is retarded in this day and age.

>>59214180
1 core into 2 core into 4 cores happened quite fast though, in one year period or so games became stutterfest on 1 cores, then dual cores bcame obsolete in two years period with same suttering

>>59214183
I'm pretty sure that if you got banned right now, 3/4th of the shitposting in this thread would be instantly pruned.
What's the point of doing this shit?
Can't people have a healthy discussion without retards like you jumping on the /b/ train?

File: 85875.png (47KB, 650x350px) Image search: [Google]

47KB, 650x350px

>>59212432
Stop cherry picking.

>>59212949
>idiot
Nobody is going to listen to you when you talk like that. If you don't want people to listen to you, why bother posting?

>>59214180
>Game developers don't give a fuck about embracing new technology.
>AMD is a pioneer in computer tech.

Here's the thing. Bethesda is a massive game development studio. Almost everyone including a literal fucking grandma plays Skyrim. Bethesda's going to team up with AMD and make multi-threading more viable in future games. If the trend goes the way it seems to be going, octo-cores will become the new mainstream very soon.

I feel for you dude. I got an FX 6300 that suffers from having neglected cores, but it looks like things are happening for real. Vulkan and DX12 are replacing DX11.

And yet again I say this: Intel knows that shit's going down. They know where the trend is going. They're ACTUALLY going to make hexa-core i5s in the next year or two. Hell they're already acting now, with 2 core, 4 thread Pentiums.

>>59212480
>It's the Ryzen $500 CPU beating the $1000 Intel CPU yet again
FTFY

>>59212432
photoshop is really poorly multithreaded and often uses only one thread.

File: Ryzen5_1600X_Performance.jpg (92KB, 1095x821px) Image search: [Google]

92KB, 1095x821px

>>59213324
So do you think this projected performance is accurate? I'll probably get a 1600X but I'm a bit annoyed at this conjob AMD has pulled. That's what the poos will do though.

>>59214259
>Stop cherry picking.
you're doing the same thing though

http://www.pcworld.com/article/3176100/computers/amd-ryzen-7-1700-vs-a-5-year-old-gaming-pc-or-why-you-should-never-preorder.html

>People who preordered Ryzen 7 CPUs to replace an aging gaming PC—like the venerable Core i5-2600K and 3570K parts so many gamers still cling to, myself included—may have very well purchased a $300-plus processor that actually delivers equal or significantly worse performance in some common gaming scenarios.

>YFW YOU DIDN'T FALL FOR THE RYPOO MEME

AYYMDPOORFAGS CONFIRMED ON SUICIDE WATCH

>>59214318
>very soon

Yeah in about 5 years

>>59214344

This will most likely be the case for multi-threaded workloads. However, I would wait for the eventual benchmarks, since the R5 1600x having less cores may be an indication of said chip having more room for overclocking.

And the only reason the R7 processors appear disappointing is because of the shitty BIOS problems mobo makers are having. Except for Gigabyte, they're golden right now with their mobos at the moment, it seems (no issues, Joker got a Gigabyte mobo and had great results).

>>59214344
multithreaded yeah

will prob be a bit behind in single threaded

File: Clipboard-2.jpg (252KB, 1920x1084px) Image search: [Google]

252KB, 1920x1084px

>>59212375
There is an issue with games at present not utilizing all cores properly. It's either a driver or game issue. Don't forget that games and some applications have all been optimized towards a 1-4 core Intel structure for so many years now. We just don't know for sure what developers have done in their 'closed source' code. It is probably just a case of the scheduler needing a patch.

>>59214423
lmao what the fuck only 3 cores are above 40% utilization

>>59214423
That's a GPU bottleneck. You don't bench CPUs like that. Nice shilling, AMDrone.

>>59214423
>GPU: 99%

It's GPU bound, of course the CPU isn't going to be at 100% utilization in that scenario. You would see similar utilization if the Intel CPU also had 8c/16t.

>>59214423
Most likely a result of the developers being used to optimizing for Intel for several years. Give it a little time and we'll see the 1700's actual performance. Actually I think I just restated what you just wrote.

>>59214486
>>59214488
Same guy here. Actually that makes more sense. Seeing that the gpu usage is almost maxed out for both.

>>59214392
Coffeelake has confirmed hexcores on the mainstream platform faggot. People don't know how the branding will roll but I'd expect hex HT i7, quad HT i5, quad i3, dual HT Pentium, dual Celeron

There's also a rumor going on that the top Skylake-X SKU has twelve cores.

>>59214484
>>59214486
>>59214500
So you mean to say that the 1700 won't bottleneck an overclocked GTX 1080 at 1920x1080, and is effectively identical to a 7700K at that point.

>>59214616
Eh, no? 7700k is much more powerful than poozen.

>>59214635
Then why are they both pegged at virtually the same FPS in that video with the GTX 1080, with the 7700K overclocked to 5ghz while the 1700 is at 3.9ghz. You're saying it's GPU bottlenecked with an overclocked GTX 1080.

>>59214635
I which this board had mods.

>>59214594
Coffee Lake and Skylake-X are still on 14nm, which means the per core price is going to be the same as Broadwell/Skylake and are going to be just as limited by thermals as before.
6 cores are going to be $400 or more

>>59214658
Because GPU is bottlenecking. Not CPU. Lower the resolution to remove GPU bottleneck and test again.

>>59214690
But no one plays at 720, everyone plays at 1080 or higher. If an overclocked version of the fastest GPU on the market, bar that ridiculously overpriced Titan X, can be fed fine with either processor, then what's the point. At this point I'd just go with the 1700 because it has more threads.

>>59214718
Dumb AMDrone, that's how you bench the CPUs. You remove GPU bottleneck. 8core poozen is literally weaker than 7700k.

>>59214484

You do realize that since both of them are experiencing max gpu usage at 1080p, the R7 1700 is no different than the goddamn i7 7700k.

Just as strong, twice the cores, godly price.

- - -

Actually the way things are right now, it looks like Kaby Lake might be the new NetBurst.

i7 7700k Death Throes Checklist

>Shitty power consumption compared to R7 1700.
Check.

>5.0 Ghz overclock for 7700k similar to to 3.7-3.8 Ghz R7 1700
Check.

>Poorly applied thermal paste causing shitty default temps.
Check.

>Terrible bang for the buck.
Check

Congratulations! 7700k has evolved into [SPACE HEATER!

If I want to play games while running VMs do I go R7 1700X or i7 7700K?

who the fuck is buying $500 processor pair with $1000 gpu playing games under 2K resolution???

File: ？.png (3KB, 297x295px) Image search: [Google]

3KB, 297x295px

When will r5 and r3 be out my mans?

>>59214739

If you're going to be running VMs at the same time, the extra cores and threads of the R7 1700 will do wonders.

The 7700k might be okay, but I wouldn't count on it.

>>59214735
Nice shilling, m8. 7700k is better for gaymen than any poozen CPU.

>>59214739
Always go Intel. Better dead than red.

>>59214749
R5 is coming Q2 2017.

R3 is coming H2 2017 (coinciding with Raven Ridge and Project Scorpio).

>>59214757
>calling someone a shill while having posted yourself at least 3/4th of the replies in this thread bashing AMD
Yeah...about that...
Mods when?

why hasn't anyone tried SLI titan X or 1080 to see how the bottleneck pans out then?

>>59214423
>1700
CPU1: 95%
CPU3: 64%
CPU5: 55%
these are the only cores with >50% utilization
the rest are <40%, most hovering under 20% and CPU2, 6, 12, 14 and 16 are basically not used at all.
on 7700k all cores are >30%, max utilization is CPU5 with 59%
looks like some sort of scheduling issue on ryzen.
any tests with SMT disabled?

>>59214770
The hell. Why wait so long.

>>59214757
Fucker ain't even providing counterarguments, let alone good ones. Mods pls.

>>59213338
ive only read the GN 1800x review where they apparently had their BIOS sorted out. 1800x got mostly destroyed in every game, one of them even had an i3 chasing it. does the 1700 do gaymen better than 1800x?

>>59214725
yes its weaker on game but in synthetic benchmark and encoding it beats 7700K

if you can afford buying 1800X for sure you also can afford to buy top of the line GPU and gaming at minimum 1080p resolution which is I see the performance on par between 1800x and 7700K

>>59211541
>>59212225
Here, I'll make it really simple for everyone:

The 1600x is an 1800x but not as good at multithreading. It might even overclock a little better.

The money you save buying a 1600x instead of a 7700k can go to a better GPU

An extra $100 on a GPU will increase the performance you see in videogames far more than a 10% upgrade from the 7700k.

The only reason to buy a 7700k is if you've already maxed out on the GPU and don't care about the price:performance you're getting.

>>59214802
Nice shilling, AMDrone. You were already told how actual people bench the CPUs. GamersNexus did that, and poozen sucks at gaymen.

>>59214355
Except in most cases Ryzen is actually faster.

http://www.phoronix.com/scan.php?page=article&item=ryzen-1800x-linux&num=4

Take a look at these loonex benchmarks. reks in most cases.

more people need to do minimum framerate tests

>>59214797
Because AMD are not some huge conglomerate and only have limited resources.

>>59214797
They don't want to turn 8 core processors into 6 core processors by slicing 2 perfectly fine cores off. Which is good. Nothing is wasted.

Sorry dude.

>>59214792
>any tests with SMT disabled?

There was one I saw that showed a ~10% average performance gain in games with SMT off, I don't remember what site it was from though

>>59212749
Already an identified problem and it's due to how windows handles threads.

The fix is on microsoft's end, so expect it never.

>>59214852
install gentoo

>>59214852
Microsoft are Intels bitch

File: 1481832345298.jpg (10KB, 186x356px) Image search: [Google]

10KB, 186x356px

>>59211541

nobody talking about what instruction set ryzen is going to have? no new instructions? improved latencies? SSE improvements? AVX512? what about BMI sets? reeee

>>59212852
>Did I fuck up too badly?

In my honest opinion: Yeah, you did.

1600x would've been cheaper, same/better performance, and better multithreading/future-proofing.

>there waiting to release the full lineup and I fucking hate that.

Well, given the bios issues, it's obvious they already released a little early. You should at least be happy that R7 is acting as a beta test for the R5 release.

>>59214867
What did you expect from a /g/aymen board?

>>59212999
the human eye can't see more than 4 cores

File: 1488490873994.png (148KB, 763x930px) Image search: [Google]

148KB, 763x930px

I'm satisfied with my purchase. Happy to get 90% of the performance for 60% of the cost and to support a competitor. Games are using Intel-optimized code. It should only get better from here on out.

>>59214878
1600x is not even out, stop recommending shit that doesn't even have any reviews

File: 1465691780359.jpg (42KB, 471x599px) Image search: [Google]

42KB, 471x599px

>>59214897
just wondering as a current game developer, what is "intel optimized code?"

>>59214885
Dayum. I ain't human dehn. I'm surprised I can see 32 of 'em.

kyle bennett, boyfriend of roy taylor and now being paid by AMD to produce a VR game series says he doesn't recommend the cpu at all for gaming.

http://www.hardocp.com/article/2017/03/02/amd_ryzen_1700x_cpu_review/6

>First and foremost, for most HardOCP readers, you are going to find that Ryzen simply has a huge clock deficit to overcome compared to Intel. If you are looking to build a system for desktop gaming alone, there is simply no way to suggest that Ryzen is your CPU as the Intel 7600K and 7700K still hold a great advantage especially if you are overclocking.

>If you are building a PC today that is going to be used for nothing but desktop gaming, I would suggest you buy a 7600K or 7700K and overclock those and enjoy the performance you will be getting with those.

so if you are going to do _nothing_ absolutely positively fucking _NOTHING_ but encode shit all day in handbrake and run cinebench benchmarks then Ryzen is the way to go. if you are going to play gaymes then get a i7-7700k.

>>59214897
You are getting 90% of the performance for 150% of the cost.

You are literally paying more for worse performance.

>>59214423
It's because games don't ever cap more than 3 cores. This is basic knowledge and also the reason why a massive amount of threads is meaningless compared to single core performance.

>>59214811
THERE IS NO 1600X

GO AHEAD AND TRY TO BUY ONE

>>59213235
>7600k
what are you doing?
OCing 7350K will give you speeds of stock 7600k

>>59214924
It's because Witcher 3 only uses DX11. Vulkan and DX12 will fix that issue.

>>59211595

you mean the 7700K.

the 6850X isnt exactly a gaming cpu.

i do agree that Intel is better for gaming, altough the difference isnt that great.

i for one would go for a 7600K with an overclock, and then just put the money into a better graphics card instead. since any cpu beyond an overclocked 7600k is overkill for gayms.

to OP. the 1700 is still the best buy on amds lineup. it is basiclly a downclocked 1800x with 65w tdp instead of the 95watt. meaning if you OC it, it should be close or surpass the 1800x. and its hell of a lot cheaper.

but again for gaming, then the 7600k is still the king.

i can add one more thing as well. im planning a mini itx build, and the Asus z270i is a fucking godly motherboard. it even has dual m.2 slots. so again no reason to go AMD.

>>59214850
hmm
well it still doesn't explain the weird core utilization on 1700
gonna see if I can get a ryzen my self later then compare code with my 5960x to see which instructions actually cause bottlenecks

File: ashes.png (76KB, 601x830px) Image search: [Google]

76KB, 601x830px

>>59214937
No it's not, plenty of DX12 games have been tested.

Ryzen loses in all of them.

File: 2.png (60KB, 925x906px) Image search: [Google]

60KB, 925x906px

>>59214909

>>59214827

>>59214933
Terrible idea.

Need a high-end mobo and cooler. End up spending more money then cheap mobo and cooler with i5. i3 7350k + Price ($168-$179) = Garbage

>>59214946
it's hyperthreading all over again

disabling smt improves performance. shit needs to be optimized

>>59214360
How are tech illiterates like them in the business of reviewing tech?

>>59214804
I read the same review as you and they never mentioned having fixed that particular issue.
They mentioned having updated to a day 1 BIOS update, that day 1 update does not have that particular fix.

The only board with that fix is the Gigabyte board at the moment, ASUS haven't pushed theirs yet (at least at the time of that review).

>>59214816
No, fuck you, their benchmark has the BIOS Bug, now fuck off.

>>59214937
Wishful thinking. In theory, sure, it's possible, but even fucking Id didn't make their Vulkan (different API I know but same principle as in Dx12) code utilise 16 threads fully.

It's constant baiting and switching with AMD. They've been behind competition for quite a fucking while now, and all we get "yeah now it's worse but soon with vulkan/zen/dx12/woopdie doo new drivers".

The inconsistent performance is infuriating, it's exactly like owning a SLI/Xfire setup used to be. Great performance is possible, but quite often it's just shit.

Wait for Zen2 basically is what this thread is. This is a new architecture, it needs to be optimized.

>>59214946

Reasons already addressed in thread:
-shitty bios on mobos (except Gigabyte's atm)
-intel optimized code fucking with Ryzen
-etc.

The shitty bios problem is definitely affecting memory clocks (downclocking and crap) and possibly Ryzen's version of SMT.

>>59214919

Of the 6900?

Or are we still using 7700 as some sort of competitor?

>>59214956
Don't Z170/H170 mobost cost exactly the same as while B150 are cheaper by a small margin?

File: patch.jpg (31KB, 276x250px) Image search: [Google]

31KB, 276x250px

>>59214827
Lisa-sama will never shill her cores on /g/
(´･ω･`)

File: 1488518738453.jpg (12KB, 217x280px) Image search: [Google]

12KB, 217x280px

>>59214827
>>59214955

It's like Gameworks but for Intel CPUs.

File: Cover.png (405KB, 640x640px) Image search: [Google]

405KB, 640x640px

>>59214995
They need to get their shit together and make Zen2 actually overclockable. If 5GHz becomes manageable, then they've got my interest.

>>59215001
>-intel optimized code fucking with Ryzen

Not to necessarily fuck with Ryzen, it's just that the devs optimized it for Intel's HyperThreading... which works slightly differently than AMD's SMT.

Now it's up to lazy devs to optimize and patch.

>>59214971
ok thanks i'll try and track down some reviews using the gigabyte.

>>59214995
This thread is pretty clear cut that 1800x is best 8core on the market at the moment, I don't understand where you got the "wait" from

>>59215039
None of these idiot reviews have tried disabling 2 or 4 cores to see how it would affect the overclock... give it a day or so.

>>59215033
I highly doubt it is intentional optimization for Intel but just how it was because it was the only architecture worth a damn at the time. Ryzen is new and still has bugs to shake out.

>>59215049
but muh gaymes

>>59215061
Gaymers are cancer

>>59215054
>buy 8c16t cpu

>disable half the threads

Congratulations you now have a $500 Haswell i5.

>>59215056
Well that's literal optimization for Intel.

But yeah, I think we're at the mercy of game devs to make sure their titles utilize AMD's SMT properly... just like they're finally utilizing Intel HyperThreading properly over the past few years.

Plus BIOS and OS driver updates might help in a few titles.

Ryzen is by no means bad a gaming though.

>>59215061
Games work fine when you don't stress the CPU, when CPU is stressed you have huge dips, SMT has problems, for sure but shouldn't affect anyone who isn't playing on a 100$ budget.

>>59214725
there is flaw in this testing method from 1999 when games were cpu bound

>>59215072
No you nigger, disable 2 cores on a 1800x to simulate overclocking a Ryzen 5 1600x

>os patch needed
microsoft remembers 7, r-right?

>>59215080
It's not AMD's SMT that has issues. There titles expected Intel's SMT implementation.

HyperThreading and AMD's SMT work slightly differently and that's probably what is causing the issue.

It took game devs long enough to properly use Intel HyperThreading. It will take a bit of time for them to use AMD's SMT properly.

>>59215082
it's more of a way to see how well the CPU will handle future gpus but that point is moot when you compare 7700K with 1700 because the 1700 has more cores and that's what is gonna be used in the future.

File: v5GFX9T.png (147KB, 984x792px) Image search: [Google]

147KB, 984x792px

Results are everywhere.

Wait for muh patches.

File: ggxcuqu.jpg (315KB, 1842x1066px) Image search: [Google]

315KB, 1842x1066px

>>59215122

>>59215065
this. wtf is up will all the "rgb" leds on everything these days.

>>59215118
This sounds very similar to what people said five years ago with sandy bridge.

>>59215153
they cost bugger all.

>>59215118
nah, I just think that frames is not correct metric for cpu testing

From little graphs of frametimes i've seen ryzen smooths things out compared to 7700k or even 6900k, I'd like someone to make whole frametime test like they do for gpus.
Probably nothing will happen though since "testers" so stuck up to their methodology it has to be something beyond measure significant to change their mind.

>>59215167
they are eye cancer.

>>59215178
Your opinions are noted but disregarded as you can turn them off.

>>59215135
what the fuck is going on?

save me form itnel overlords already! I'm okay with 5% but not okay with 50%. No more intel CPUs, noooo.

>>59214816
Most people on /g/ do more than JUST game. If you want to feel like a badass for getting people to buy intel CPU's, maybe you should go to /v/. You'll have more success there.

>>59211541
To me, Ryzen is a pretty awesome choice, because I run a lot of threaded shit, as well as multiple VMs, etc.

MORE CORES is good for me.

For the absolutest high-end there is some Intel Xeon that would arguably "be better" for me, but that's so much more money than I want to spend. I would rather spend it on more RAM, more SSDs, etc. Because those are things I actually need.

> But what about iGPU? Do you want to have an external GPU?
Well, yeah, I do. External GPUs allow me to have loads of monitors at high resolutions, which is something I am willing to spend money on.

I know that you can drive 2 4k monitors for a desktop with Intel's onboard stuff, but it's not exactly optimal.

where are the 4k/1440p tests? all seem to be done on 1080p

>>59215231
most poor people cant afford 1440p so its tested at that
and its easier to get the game choked on the cpu at 300 fps

>>59215231

I've only seen a few but it was interesting.

One site had the 1800x losing in every game in 1080p, and suddenly pulling ahead by 10FPS in about 30% and tied on another 50% once you're at 4K...

Performance is so different when you compare these reviewers you know some shady shit is going on.

>>59215200
Save your worries and just get an Intel instead of this waiting game of patches and optimisation promises.

>>59215266

in all seriousness tho. if you are planning on buying an 8core cpu, then 1440p/4k is a big thing.

i mainly play starcraft, but its in 4k, so for me its not a big thing, but im still curious if performance is different.

>>59211541

The transition from dual-cores to quad-cores occurred around 2006-2007 a time without YouTube. Now, as of 2017, YouTube is the biggest internet video service ever.

I'm sure there are a lot of budding content creators that might want to start out with something beefy, but reasonably priced. The Ryzen processors are perfect for content creators on a budget.

I could see content creators with 4 core, 8 thread i7s switching over to an R7 1700 for the benefits it reaps when it comes to video editing software.

These people may come in droves. Especially when the R5s and R3s start to kick in. Strong with great bang for the buck. That's what should matter most in the end.

>>59215287
Well, YouTube was around, but not in full force though.

>>59215041
Not only that, there is code literally running loops and calling sleep functions.
Intel literally got into court and lost recently.

>>59215135
>>59215200
sooo
I've read some interesting shit
apparently there are major issues ASUS boards. Some have BIOS updates that give +20% performance or something. Gigabyte boards seem to also work much better than ASUS boards.
Looks like there are some major motherboard issues

>>59215231
at 1440p it either better than 7700K or exactly the same
dishonored 2 is 3-4% better for example

so I don't know, I either stay on i72600 forever or go 1700x right now because I've got 1440p monitor

initial smear is on CPu already though, not everyone would read 10+ reviews like I did

well based on the joker review with the gigabyte mobo im thinking 1700 is a fantastic option. assuming reviewers werent all sent really good chips and normal retail chips can OC to 3.9

Im curious as to how a 1700 @ 3.9 does against a 1800x @ 3.9 in nongaming workloads. if it's similar theres really no reason to get a 1700x or 1800x lol

>>59215324
Yup. All the mobos have shitty bioses right now (except for Gigabyte atm). Many people experiencing downclocking ram and see Ryzen's SMT being fucked with.

Problems like these ended up staggering Ryzen's momentum...

>>59215413
its the same chip
it will run at a higher voltage
1800x could posssibly overclock higher with newer bioses

>>59212608
You're typically running more than one application you moron.

>>59213152
That market segment, and I know it because I work in it, uses i7s for laptops, and Xeons for workstations and servers, and don't care about how much they cost.

Ryzen has the potential to be huge here, because Excel is king, and 10 years ago, Excel 2007 started to enable multi-threaded processing. This means that every beancounter currently on a dual-core i7 is going to hear about the new processor that can run Excel with 8 cores instead of 2, and thus 4 * 0.85 = 3.4 times faster. They will want that so badly that you have no idea.

And then they'll see the pricing, and they'll point out to their boss that it's the "cheap" upgrade, and that they really need new computers soon, and they shouldn't use old outdated tech and please oh please daddy can we get the new toy?

It might very well happen.

>>59213955
Since when were games developers good programmers?
Yes. There are a few people who are really good programmers and who make games. Carmack for example.

But ye average games programmer is not better than ye average programmer. And he has shittier working conditions, with worse pay, worse benefits, worse methodologies, and worse tooling.

Games, from a craftsmanship perspective, are made by enthusiastic mediocrity.

It's not the CPU's job to be massively multithreaded, it's the GPU's: >>59214968
AMD kids are illiterate in technology.
They pretend a CPU is a GPU....

>>59215460
Then where are all the single-core CPUs with stellar performance?

An alternative hypothesis: You're an idiot and need to kill yourself.

>>59214164
Let me tell you a little secret.
Do you know how you write a standard fuckparty enterprise Java program these days?

> Step 1: set up a threadpool
> Step 2: Just throw that needs to be done at i.
That's how you do it these days. I assume my fellow ENTERPRISE PROGRAMMER brethren can back me up here with other languages.

Doom 3 had Carmack struggling with using 2 threads. Watch Dogs 2 now just scales to whatever number of threads it has. Our fellow programmers who maek gaems have finally figured out how to threadpool it seems.

This ridiculous idea that we'll be stuck with dual-threaded games etc. is banking on the incompetence of the ones making the game engines. If UBISOFT could do it, fucking EA should ba able to too.

> Intel says
> Pay no attention to my mobile i7s.
> http://ark.intel.com/products/97541/Intel-Core-i7-7567U-Processor-4M-Cache-up-to-4_00-GHz