1800x or 1700X

Good question. I think the 1700X is the more sane option, but I think I'm going to grab the 1800x regardless. Just have to wait and see benches and overclockability.

>>59113924
1700X. Given the dynamic overclocking that XFR provides, I don't see the point of paying an extra $100 for another 200 mhz.

>>59113924
I ordered the 1700 myself
65w should save a bit of money on power, and my current heatsink will be overkill for it which is always a good thing
then again the most strenuous thing I'm going to be doing on my machine is muh gaymes

>>59113924
>500 dollars
none of them

>>59113924

I hear the 1800x is better.

>>59113924
intel i7 when price drops

>>59113924
Depends on how aggressive their binning process is.

since they're just different bin we'll have to wait and see how well the chips overclock. i doubt the difference is worth the extra shekels.

>>59113924
neither. the best ones will be the 6 core. these ones are pointless and overpriced.

>>59114349
I hope OC gains will be linear, I really don't feel like dropping extra $100, but if 1800x clocks absurdly higher than 1700x I'll end up starving for a week

File: boxxy.jpg (895KB, 2560x1600px) Image search: [Google]

895KB, 2560x1600px

>>59114401
>Implying that software of the future won't be written to make use of 8 cores.

shiggydiggy

>>59114401
>tfw can't afford to wait for 5 because my motherboard is about to shit itself

>>59114424
How about getting a shitty $20 replacement mainboard for your current CPU from Craigslist or whatever your local equivalent is?

>>59114418
>implying an extra $150 for 2 more cores and 2 seconds faster winzip extraction time is worth it

>>59114450
not worth it, considering I could just spend that extra $20 on an 8-core Ryzen CPU instead

>>59114418
>boxxy was like 10 years ago
Fuck I'm old.

File: 1259250022064.jpg (11KB, 289x319px) Image search: [Google]

11KB, 289x319px

Not really interested in the 8 core ones since they won't overclock as high as the ones with less cores.

The 1600x, the 6c/12t is the really interesting one.


If it can get to 4.5 ghz I might get it, else I'll jump on all the lmao's trying to dump their 7700k's right now

>>59114491
This is the best course of action.

>>59114033
Because binning
It's nice to have XFR but it's full potential lies in the 1800x

>>59114535
We'll have to wait and see how aggressively AMD is binning though. Absent additional information, $100 for 200 mhz seems like a lot of dough.

> the 6900k, the $1000 chip I cherry picked, is 10% better than $300 Intel products on the games I cherry picked
> therefore Ryzen is amazing because it's cheaper than $1000
Can you stop pretending that even you believe that?

>>59114600
Yes, make this a meme

>>59114600
I have to give it to AMD though, they did manage to make the AMD shills in here to not see they're talking about a fucking $500 chip, an amount that literally nobody in /g/ would consider even for an Intel chip.

>>59114615
The 5's are where the value proposition starts for everyday users, but the 7's are impressive in that they're matching the performance of super-high-end Intel chips that cost 2 to 3 times as much.

>>59113924
1700. All of them are overclockable.
However, you better not jump on a hype train. Ryzen 5 4core may be just the thing.

>>59114634
but seriously, we can't let AMD get the nuclear codes

>>59113924
Obvious choice... 1800 is a higher number than 1700

File: liberalism.png (641KB, 960x596px) Image search: [Google]

641KB, 960x596px

>>59114418
>boxxy

:^)

>>59114241
>65w should save a bit of money

You're talking >$15/year

If you're that worried, you shouldn't be buying computer parts

>>59114418
The majority of current software doesn't even use four cores, retard. By the time most software uses 8 cores, we'll have 32 core processors.

>>59115017
it gets hot here in the summer, the less heat my pc generates the more money I save on AC
that's not insignificant

>>59115026
> The majority of current software doesn't even use four cores
And I'm really grateful that most software isn't bloated enough to ask for more than 1 thread.
Butt! Video encoders can utilize all the cores, so if you're into it, Ryzen is a good purchase.

>>59113924

1700x. Seems like AMD is having no yield issues this time around meaning they won't be aggressively binning them.

That in mind paying an extra $100 for 200Mhz is dumb. Only a little while longer and we will know just how well they overclock.

File: 1310483412100.jpg (36KB, 413x395px) Image search: [Google]

36KB, 413x395px

>>59114971
Is that really her right now?

>>59115109
yeah

https://twitter.com/catiewayne?lang=en

File: 1346777103515.gif (2MB, 161x120px) Image search: [Google]

2MB, 161x120px

>>59115139
>https://twitter.com/catiewayne?lang=en
Why would someone make themselves look as crappy as possible with their haircut?

As a bald person this pisses me off

>>59115204
u mad Baldy McSkinHead

>>59114491

This is the smart choice.

>>59114971

Sometimes it's better to not know what happened to people.

Neither, I plan to stick with Phenom II.

>>59114343
Good goyim.

>>59113924
good question.
im looking at the 1700x because 120€ more for 200mhz is a lot.

But who knows how it's binned. If the 1700(no x) overclocks well that'd be the best option, but it probably wont, or they could sell it as a 1800x.

And I dont like the added element of unstability that overclocking gives. Even good overclocking CPUs can be fine in a stability test for an hour then crash randomly when you're doing something important.

>>59115270
I thought as much was clear from my post

>>59115408
so why not get the 1700 if you don't like overclocking?
what I'm probably doing, unless intel respond with a better deal

>>59115408
>>59116040
misread your post
1700 looks best price/performance
I heard it does overclock well, just not auto

File: chance_to_repair_100%.gif (2MB, 185x138px) Image search: [Google]

2MB, 185x138px

>>59115204
I thought it was common knowledge that Boxxy went full SJW

File: 1485796515828.png (443KB, 578x481px) Image search: [Google]

443KB, 578x481px

>>59115139
Disgusting.
I checked out her channel once when she started uploading again and she seemed like a perfectly normal girl.

>>59116158
>literally painting eye bags on your face

>>59116158
Why does boxxy look like a Mexican Now

>>59114535
Yep, and the way I see it is
>I can get the top of the line for just 100 dollars more

I'm getting a 1800x

>>59113924
Hard to say this soon, wait for reviews.
If it's the same story as it was with Bulldozer and 1800X is a way to skip the silicon lottery by guaranteeing the best binned chips it might be worth it. But it depends on how good 1700X then ends up being on average.

File: 1484757793640.gif (2MB, 500x525px) Image search: [Google]

2MB, 500x525px

Wait for benchmarks~
And then wait for the 1600X~~

>>59116356
faggot

>>59113924
Wait for reviews. That 1800X is pricey

>>59116374
Rude.

If the 1700x can reach 4.2GHz with an NH D15 and a 200 euro motherboard, I'm getting it.

>>59116374
anime website

>>59113924
you're already invested into buying depreciable tech. might as well go all the way

I got an Asus B350-Pure and a 1700x. I'll fiddle with the multiplier a little bit, but I have an H100i so hopefully XFR will take it to 4.0-4.2. I'm looking forward to reviews so I have an idea of what to expect. I know it's a damn sight better than my 8350 though.

>>59114535

Or it'll end up like the 8320 and 8350 - they both slam into a voltage wall at the same point, the more expensive one just has higher clocks out of the box.

>>59115087
Implying I use my pc for one task at a time

File: 1399249788733.jpg (32KB, 300x400px) Image search: [Google]

32KB, 300x400px

>>59115017
>everyone pays the same price for power

>>59114971

she should've just did porn

File: 38500.jpg (133KB, 640x468px) Image search: [Google]

133KB, 640x468px

https://www.computerbase.de/2017-02/cpu-skalierung-kerne-spiele-test/

we are already seeing high clocked 4 core cpus falling behind slower 6/8 core cpus. In a couple years the difference will be larger especially because CPUs at least those of Intel are not getting much faster regarding single thread performance.
I think for the coming couple years having a fast 6 core will be the sweet spot. Considering it seems these days a cpu can last as long as 5 years getting a 8 core might pay off

File: it begins.png (392KB, 3840x1440px) Image search: [Google]

392KB, 3840x1440px

Kiketel dropping prices

>>59115055
Yes it is. Running a window AC 24/7 in a room is less than $100 for 3 months.

File: LBtlbNU.jpg (41KB, 625x590px) Image search: [Google]

41KB, 625x590px

personally i care more about low power use because it generally means less noise.

>>59116776

You should see the denial of that article over on OCN - the amount of people that crawled out of the woodwork to declare a sandy vagina i5 is all you ever need is astonishing.

File: 928f000dc66a400bab27d1dc666d947c1.jpg (35KB, 499x499px) Image search: [Google]

35KB, 499x499px

Waiting for pricedrop on Baby Lake 7700 non K.
Stay mad faglords.

Anybody know which mobo will be linux friendly yet? I'm sticking with Arch and I refuse to install Winshit on my new rig. I'd rather just mod wine to suit my needs. So far all I know is Linius released a few vega and ryzen driver support with the recent Kernels. After that I'm not sure what to make of it and which mobo to go with. I wanted MSI but I heard Asus might be the goto choice for initial release.

>>59117003
autism is a serious issue

>>59117013
I know it's compatible with all mobos and the kernel deals directly with bios. I'm just concerned about driver issues and bugs.

If you have no idea wtf i'm talking don't comment.

>>59117042
no i just think you're a sperg
people who cause this much of a fuss over refusing to use something come off that way

File: 1477125067026.webm (2MB, 718x404px) Image search: [Google]

2MB, 718x404px

>>59116944
We're not mad if you want to spend more for a worse performing processor.

>>59116944
Intel never pricedrops.

>>59117143

really? >>59116788

>>59117153
>>59117143
no no they're not pricedrops, they're "competitively priced sales"

>>59117153
>>59117158
Those are all Microcenter prices. Instore pickup only, they don't ship anywhere, and those prices have been up since those chips launched.

>>59113924
threadly reminder that you MUST buy Ryzen to support competition even if intel after price drop is objectively better because otherwise you are just fucking yourself in the ass by supporting another years of monopoly.

>>59113924
I'm so ryzened I don't which one of my hands to use to masterbait!

I'm getting a 1600x for muh vidya, because it seems the best contender against the 6600k/7600k in terms of price and performance

>>59117185
that's nice but basically every Canadian parts retailer has the Intel chips on sale right now

>>59117223
Please source your claims.

>>59117242
check ncix, tigerdirect, newegg, directcanada, and canadian computers then

>>59115408
>Even good overclocking CPUs can be fine in a stability test for an hour
Nigga, don't you go full 24h when doing an overclock? I got the 3570k to 4.4GHz after lowering it from 4.5GHz. It crashed very late, like 5 or 6 hours after the meme Mersenne prime tests.

If I'm not too fussed about overclocking until quite a bit later would it be a better idea to get the R7 1700 (cooler included) than the R5 1600X which doesn't come with a cooler?

>>59117242
oh and amazon

File: Smug as fuck - Flamberge.jpg (91KB, 1280x720px) Image search: [Google]

91KB, 1280x720px

>>59116374

>>59117205
Yeah but this is one of those game-theoretical things, prisoner's dilemma or tragedy of the commons or whatever, where it benefits me personally to buy Intel if everyone else is buying AMD, but if everybody is buying Intel instead of AMD we all suffer. Most likely everyone will be buying Intel in that case.

File: CgwMR2mUgAAZ81b.jpg large.jpg (168KB, 960x1280px) Image search: [Google]

168KB, 960x1280px

>>59114971
She's still pretty cute desu

>>59117216

same for me

i dont mind buying intel counterparts for 1600x IF the price will drop by a HYUGE margin.

its so good to have some competition

>>59117252

CA prices are overpriced anyhow, nobody cares.

Fact is that the 7700k is still $340-$350 on every single store except some small retailer like MicroCenter. Then again MC has been running Intel CPU deals for years so this is nothing new.

>>59117320
>CA prices are overpriced anyhow, nobody cares.
which is why the Ryzen release is important as fuck
normally we pay anywhere between $40-100 more, even after the exchange rate. the new Ryzen chips are the USD prices converted to CAD + $5

Does it make sense to jump from a 3770K running at 4.4GHz to a 1800x?

I'll have to change the mobo, and get new ram (ddr3->ddr4) so expenses add up

>>59117373
no
wait for Zen+

>>59113924
>>59116260 This

>1800X (499$) is going against a i7-6900K (1050$)
>551$ savings
>1700X (399$) is going against a i7-6800K (429$)
>30$ savings
I would go for the 1800X

>>59117373
I'm doing it from the non HT version, but I'm looking at an entire new PC. What do you even do if you're asking this? I encode, compile everyday, and sometimes stream. The extra cores will help me greatly and since it looks like it's at least Broadwell-E, a 4GHz 8 core behemoth won't slack on single threaded for emulation or gaymes as well. There's also the thing about GPU passthrough that I could try with Zen and get rid of the dualboot.

>>59117003
I'll probably go with Gigabyte as they have win7 support listed meaning UEFI to bootloader is likely not going to cause any major issues, and the AB350M i'm looking at has Realtek ALC887 and GbE LAN which are both common enough to be supported by literarily every kernel oob. Also based on the online users manual I could do any overclocking via bios if there won't be any kernel/userspace tools for it

>>59113924
Why would you pay $200 more for a CPU that has the same number of cores, the same core frequency adjustments, and the same IPC? It's a no-brainer, the 1700X is the only reasonable choice.
>b-b-but muh overclocks
The 1700X probably clocks 200MHz lower than the 1800X. That means you're paying ~$100 per 100MHz. Fuck that noise.

>>59114491
>The 1600x, the 6c/12t is the really interesting one
Yes, but the 1800X/1700X is binned from higher silicon with better leakage tolerance. The 1600X will probably not overclock as well as a 1700X with two cores disabled at the BIOS.

>>59117738
Best to wait for some reviews and see whether there's a voltage wall or if better binning makes 1800x's more sensible

If binning doesn't make much difference, then buying a 1800x is essentially a retard tax for those who can't overclock

>>59117776
There is a voltage wall, but it's not with the CPUs. It's the motherboards that are hindering the amount of voltage you can feed through it. You could hit the same clocks as an 1800X with the 1700X, but it takes more voltage and stability is not guaranteed.

>>59116064
It doesn't. The best oc on it so far was 4.0ghz with water. And you need the X motherboard for that otherwise you'll be getting a very unstable 3.8.

>>59114680
why not buy octa? dont be the new "2core is all you need" guy

>>59115026
games will use 8 cores in the near future, watch dogs 2 already does

>>59115408
this is why i'm considering the 1700x myself with a good water-cooler and let xfr do the ocing depending on temp and what's being used

>>59114401
They're only pointless if you're not comparing them to the x99 processors

>>59114535
>Because binning

History is on the side of this being bullshit when it comes to AMD chips. The lower-clocked FX and Phenom II chips generally reached the same average frequencies as the flagship models with manual tweaking.

>>59117272
intel will never give you an octacore that clocks 4mhz+ at 450 euros

File: Tt9GQGG.jpg (79KB, 1095x821px) Image search: [Google]

79KB, 1095x821px

>>59117759
>we are already seeing high clocked 4 core cpus falling behind slower 6/8 core cpus. In a couple years the difference will be larger especially because CPUs at least those of Intel are not getting much faster regarding single thread performance.

pfff the 1600x will be a good binned chip, they will have 3.6Ghz base and 4Ghz turbo, just like the 1800x

>>59117998

>4mhz+

I don't think my motherboard could handle such clocks.

>>59118085
It doesn't work the way you think. Some of the silicon that gets sold as the 1600X are defective or failed to meet certain power requirements per frequency, thus having part of their modules deactivated to bring them under control. A 1600X and 1800X might have similar initial clocks, but they will probably not scale the same with every clock increment or voltage step. The 1800X with just six cores active will almost certainly overclock better with lower voltages than the 1600X.

1600x

>>59116415
>weebshit in a technology board

>>59118157
You know what would be great? Profiles outside of BIOS.

Imagine how great it would be being able to disable 2 cores and overclock 100-200MHz higher with just one click? Application profiles would be great, too so if you're playing older gaymes, you can disable 2-4 cores and have a higher overclock.

>>59118409
I don't think that could work with Windows OSes. Also, it could be a huge disadvantage for systems that might have a lot of daemons running.

File: reaction.jpg (121KB, 491x521px) Image search: [Google]

121KB, 491x521px

>>59118341
>Anonym
>arguing about anime on a scandanavian cheese forum

>>59118409

>he doesn't know about ryzen master

Its happening.

File: ZwY05FK.jpg (78KB, 1491x764px) Image search: [Google]

78KB, 1491x764px

>>59118409
>>59118435
>>59118487
LADS

File: 1475814697497.jpg (6KB, 160x160px) Image search: [Google]

6KB, 160x160px

1700x costs $469
1800x costs $586

in our country where average month salary does not exceed $1005

File: QGZ1ewy.gif (141KB, 287x344px) Image search: [Google]

141KB, 287x344px

>>59118533
>click on 0 Cores
>computer dies

>>59118635
>cores disabled: 0
hmmm

>>59118655
Whoops, my bad then.
Wait a Jew-baiting second, are we allowed to even choose which cores can be deactivated, or does the CPU do it automatically?

>>59118672
Doesn't look like it based on that picture, but I assume you can't disable the first two cores (cores 0 and 1), but you can disable the rest starting from the last two (6 and 7).

>>59118533

>step size 25

Now thats some fine grain tuning. It will make things interesting for those squeezing every last drop of performance from their system. I can see ryzen OC coimpetitions being down to such miniscule margins.

Thinking upon it depending on how fast ryzen's race to zero is (and iirc its lowerst core clock is something stupid like 200mhz) that makes it devastatingly good for mobile parts.

>>59118718
That sucks. At least you can do that in the BIOS if you really wanted to.

>>59113924
>AMD
neither, obviously.

What do you think Intel's marketing staff is tasting right now?
The cold barrel of a 12-gauge Winchester, or the bitter, almond-y taste of cyanide?

>>59113924
Both are btfoing intel so it doesn't matter

>>59118763
(you)

>>59116815
>I don't live in Australia therefore no one else does

File: ryzen-is-shit.png (2MB, 1920x1080px) Image search: [Google]

2MB, 1920x1080px

>>59113924
i7-7700k

>>59118907
>stock i7 7700k better than overclocked i7 7700k
what a valid and convincing benchmark you got there goldstein

File: 9wDBk1g.jpg (6KB, 250x239px) Image search: [Google]

6KB, 250x239px

>>59114971
>tuck Trump
That's cute

>>59115026
> Xbox One and PS4 both have 8 core CPUs
> nah, games won't scale, never ever

>>59113924
I7 4790k after a huge price drop in the used market.

>>59118907
ryder, not this again

>>59119130
don't bother, quadcores are now depreciated, go for ryzen

t. i7 4790k user

File: RIP in peace.png (503KB, 781x2497px) Image search: [Google]

503KB, 781x2497px

:^)

>>59119262
Intel is finally starting to not jew people as much wtf

File: 1458716115686.jpg (105KB, 480x800px) Image search: [Google]

105KB, 480x800px

>>59118533
Fucking nice.

I wish I didn't live in Greece.

>>59118777

Look up the Pentium 2 commercials.

I have a 1055T, will a 16/1700x be an acceptable and necessary upgrade?

Also: http://wccftech.com/3dcenter-ryzen-intel-cinebench-r15-comparison/?utm_source=wccftech&utm_medium=related

>>59118907
holy fuck what a blatantly retarded benchmark
haha
post more comedy gold

Watercooled 1700 system for me.

>>59119282

>As much

What did he mean by this?

>>59119393
http://wccftech.com/amd-ryzen-7-1800x-overclocking-performance/

Air cooled OC kills Kaby Lake.

File: ryzen-btfo.png (190KB, 1085x774px) Image search: [Google]

190KB, 1085x774px

>>59119374
How mad are you that it's real?

Here's another one.

The only thing Ryzen is good for is workstation (aka video editing)

Gaming and desktop usage, Intel beats it hands down, for the same price or even better

>>59114971
wtf I hate boxxy now

>>59119439
Sorry I don't care about your shill war

But do you really not see the flaw in >>59118907 that "benchmark"?
Really makes me think

>>59119358
By 1055T I assume you're talking about a Phenom 2 1055T six core? Even overclocked, the Phenom 2 series would still fall short of the new Ryzen in single core alone. Not to mention you'd be getting 6c/12t or 8c/16t compared to the standard 6c you have now. A stock Ryzen setup will easily out-perform a hot and heavy OC'd phenom 2.

File: 1487970128273.png (145KB, 648x348px) Image search: [Google]

145KB, 648x348px

all aboard the hype train!

>>59119439
There are only 2 benchmarks for the 1700x in that pic compared to 6,832 for the i7-7700k. I don't think the sample size is big enough yet to make conclusions.

>>59115017
>got new pc parts
>heavy OC
>paid 200€ more compared to 2015

fuck this

>>59113924
>390 or 390x

nigga

>>59118907
I commented on this in a previous thread, but I'm going to do so again because of how misleading those results can be to most people just wanting to knee jerk. All chips have relatively similar averages, yet the max frames are high on the intel chips and the mins are low on the AMD chips. Neither of those are very good indicators of anything for any of the chips because of how high the variance is. They're outliers in all cases. Literally useless stats.

>>59115026
Good thing you're only running 1 piece of software at a time then

>>59120208

DOS never felt so good.

>>59120197
The better IPC can cause those minimums and maximums at extreme high load when for example changing scenes. When it takes the CPU 40 milliseconds instead of 30ms for that change in scenery you get a way lower minimum FPS. It's completely meaningless without the fps graphed out fully against time.

>>59120197
Except the Intel chip gets a better average FPS too.

>>59119805
We don't know anything really until the good reviewers start putting it to the test.

But it's not looking good for AMD so far.

>>59120425
>it's not looking good
you're either blind or stupid

>>59120415
Very insignificant amount. The 7700k gets destroyed in multi-threading. Also lower power consumption on the Ryzen.

>>59113924
getting the 1700 and using the extra cash I save on a nice mobo, then will upgrade to zen+ next year and ride it off comfy

am4 is expected to last ~5/6 years right?

>>59120415
Yes, but a very small one at that...

Games using old fashion graphics APIs like DX11 are limited by single thread performance whenever they go well above 100 FPS and we know from other benchmarks that Intel has better single thread performance while losing at multi thread performance. Newer APIs like DX12 and Vulkan are supposed to fix this, meaning that when games start featuring good implementations of these APIs, the 1700 is going to end up not only closing the gap, but getting ahead.

>>59120453
Ryzen looks to be good in specific multi-threaded situations like video encoding.

However Intel seems better at everything else.

This is classic AMD all over again, adding more cores and getting beat in real world performance.

>>59117382
>Zen isn't even out yet
>wait for Zen+
hurrr

>>59120488
You may have a point when it comes to games, but pretty much everything else that's heavy on the CPU had embraced proper multithreading long ago. Even games are now getting in on it with APIs like DX12 and Vulkan not to mention the influence from current gen consoles that both have 8 rather weak cores.

>>59120488
Yes, because a slightly lower performance for a drastically reduced price is truly a horrible thing for AMD.

>>59120437
fuck off with your g memes

>>59120488
Intel is better at is insignificantly. That's the thing here. Nobody cares about a few FPS when you're already around 90 FPS.

>>59120590

>muh 144 hz gaymin'

So when do the cheap ones come out? I'm a huge sucker for the price/performance meme

>>59119439
>ebay used price vs brand new

Wow you showed him.

>>59113924
>>59114014
this.
I'm getting an 1800X because I have the money and I want a nice processor. Not pulling the trigger till I see some basic reviews on the top-tier Aorus mobo and whatever DDR4 RAM I pick (probably GSkill)

Been using the same brands since '08 and it's never let me down.
2008
>Phenom II X4 940
>Gigabyte GA-MA770-UD3 rev 1
>G.Skill RAM
>Sapphire HD4850
2013
>FX-8350
>Gigabyte 970A-UD3P
>G.Skill RAM
>Sapphire R9 270X, later to RX480 8GB Nitro+
Soon (tm)
>AMD RyZen 1800X
>Gigabyte Aorus 370 Gaming 7
>G.Skill RAM
>Sapphire RX480 later to 2X Radeon RX 5??

Shit's cash. Only failure/unreliability I've had is a single RAM stick fail on the 2008 build. That's it.

>>59120792

>Thus the command from Israel.

>200 megahertz for 100 bucks
>you can overclock the 1700x anyway

yeah totally worth it

>>59114014
>overclockability
>overclockability
>overclockability
http://wccftech.com/amd-ryzen-7-1800x-world-record/

>>59120197

And according to the benchmark the 7700k OC is worse than the stock 7700k ...

>>59115398
So when I buy intel/nvidia products I'm a jew lover.

What does it make amd/ati fans then?

What air coolers is AM4 compatible with, is there a list somewhere?
Don't live in the US so sending off for brackets or something would take months.
Confused by contradictory reports.

>>59116944
>Waiting for a pricedrop on intel stuff
You'll wait longer than any AMD fanboy ever has.

>>59121060

CoolerMaster had a list.

>>59121060
Dunno but I sent Cryorig an email a while ago and they said they were planning on releasing an AM4 conversion kit for the R1 ultimate so I'm going that route when I can.

>>59121060
Some motherboards have some sort of backwards compatibility with AM3 by having secondary holes. The only worrying thing, though, is the Ryzen package being a bit thicker compared to the old AM3 processors.

>>59121060
anything that doesn't use backplate is compatible since distance of mounting bars has not changed, but hole placement did

>>59119439
what site is this?

Thanks guys. Used to be into this stuff but its been a very long time since I've felt like anything was going on that might even be interesting to upgrade to, let alone justifiable.

>>59121125

The only board that supports AM3+ coolers is the crosshair hero. Luckily for me its the board i'm going all-in-we're-gonna-win with.

Being able to bolt a 5 year old 250w+ cooler to a new board (a socket change a year keeps the goyoim in fear) is a real money saver for me. Especially considering I can give the finger to my cooler manufacturer in so muchn that I don't need to buy their new bracket, effectively wiping out the increased cost of the hero versus other AM4 boards.

That and asus were the only brand to take AM3+ seriously so there is some loyalty there.

File: niceguy.png (248KB, 635x347px) Image search: [Google]

248KB, 635x347px

>>59113924
>mfw want to pre order but still want to wait for benches from independent reviewers and with more gaymes
I wouldn't mind buying on launch day if I didn't think that Ryzen is going to be sold the fuck out everywhere for god knows how long. Neither AMD or Intel are good at having decent availability on launch. Also fuck me for having to wait for the bracket upgrade kit from Noctua.

I'm like this close to pulling the tigger and just pre ordering.

>>59121357
i really don't believe this meme about being sold out everywhere. amd have stressed again and again that it's a hard launch with proper availability. just wait for the reviews.

>>59121357
>>59121396

Word is AMD is supplying 1 million units worldwide - I would think (at this price point) 1 million is a good supply all told. I plan to order in the next two weeks so it will be curious to see how far down the list I get pushed in my country (a country with good buying power I should add).


>gibbo you cocktease

>>59121357
You can always return it

File: Goyim I....jpg (204KB, 677x995px) Image search: [Google]

204KB, 677x995px

>>59120997
A Bad goy

>>59121396
>>59121450
A million world wide is a bit of a relief, having seen the last couple of releases from AMD's GPU division. After seeing Ryzen shoot up on Amazon's best seller list, I couldn't help but think AMD would only be shipping out small amounts. I actually remember the Intel launch of skylake, and my local microcenter got like 5 fucking chips of each SKU.

>>59121476
I very well could but that's a hassle that could be avoided.

>>59114535
wat is xfr

>>59121739
extended frequency range, so basically if your cooling is good enough it will push its clocks past its boost clock

>>59121762
ah ok, I saw that on the videos months ago just didn't know the name.

I like that honestly. Does that mean I shouldn't even overclock and just let the turbo handle the upscaling when needed? (I have apongee II cpu block I'll use for this system)

>>59121567

Always bet on Keller. Memes aside it seems AMD has gone all-in with ryzen and frankly at this price point and performance it is poised to flatout molest Intel's bottom line. If it can meet demand then Intel is going to one hell of a bad yearly report.

Remember: broadwell-E was basically a myth upon the supposed hard launch.

>>59121798

>Does that mean I shouldn't even overclock and just let the turbo handle the upscaling when needed?

In theory yes - as long as you keep the chip below whatever predetermined factors (most likely a combination of power draw and vrm temps) it will keep clocking itself up 25mhz at a time. Right now the nuances of XFR are unknown but in the consumer space its possible (again I must stress how little we know about it) it is one of the biggest technologies to hit cpus in a long time.

Again - in theory - it works like Nvidia's boost, the cooler the chip is the higher it clocks.

>>59120487
Consider the following

Canardpc said they got theirs single core to 5ghz without breaking a sweat, only held back by the motherboard.

Wattman is getting an update for per core ryzen overclocking.

So, consider this, you overclock the hell out of 1/2/4 cpus, and undervolt + downclock the others in a profile for specific games, single core should hit 4.8 to 5+ and you now have you single core performance, possibly better then kaby lake.

Now this is manual stuff, so meh on its use, but what if xfr is smart enough to understand the load and can do this on the fly?

we got some new info to add to canard's and things are getting, at the very least, interesting.

>>59121834
Yea it'll be really cool to see how it scales.

I have gotten the 1700X and the rog crosshair VI hero already on pre-order so hopefully the X chips have that XHR in them as well.

>>59117382
Wait for zen++ imo

>>59121895

Word is XFR is baked into all chips - the X on each sku (i.e 1700x and 1800x) is just a sign of binning. I would guess that the 1700 is the 8320e (to the 8350) of ryzen.

>>59121861
If that's true, it'll be awesome. But for some reason I can't shake off the idea that Ryzen will yes, be good, but not THAT good.

File: 1483602771087.png (705KB, 900x900px) Image search: [Google]

705KB, 900x900px

>>59121922

>>59120882
This. With one caveat.
Does Gigabyte Aorus support synchronization with G.Skill's RGB memory?

>>59121964

Indeed. Still its unknown how important this divide is (especially with the lower TDP of the 1700) but my gut tells me the 1700 will be the pleb chip of choice with the 1700x being the smart overclocker's chip of choice.

1800x remains for richfags and fools.

>>59122020
damn been on 4chan since 05 and never once did I see anyone almost call me smart but I guess my 1700X purchase was a good one

hopefully asus rog gives some good speeds to it

>>59121952
What i'm thinking is it will match intel's 8 core for all core oc, which it likely will if motherboards will push it.

What I have hopes for is xfr isn't just a marketing stunt and its a real tool that will oc your shit more then good enough for you.

beyond that, my hope is that it will be able to understand the load and adjust accordingly.

we got 2 gta benches right now, one of them was a leaked piece of shit, the other is up in the air if it actually was ryzen, but we know the first did not use xfr, what if the second one did?

This is literally the only reason i'm not preordering a cpu right now, is I have to know what this amounts to. I have been back and forth for the last week on which one i would get when I was set to get the 500$ black edition cpu for the last year or so (Yes, I guessed the price well over a year ago based on die size and business decisions I would make if I was in amd's position, and got my confirmation on price at the new horizon event for the minimum, and told yes, "black edition" will be 500$ at ces) first linus said 100mhz more from the 1800x, and for that fuck it, but I got thinking, if xfr CAN oc, then it could bin cpus like never before, so the lottery may be real, but then wattman getting ryzen support for oc on a per core basis, this really makes me think the 8 core is a best of all worlds chip, it can down clock and effectively shut off 2-7 cores depending on the load, and if xfr does it smartly without my input, well fuck me man, that is all I need to know. if xfr isn't smart, I may go 1700x just for the sake of the self oc, if it is, 1800x, and if the mining isn't real, ill likely get something off silicon lottery.

>>59120488
Yeah, that's more of the same, then. The times AMD beats Intel at multithreaded performance in a similar price point have been very specific and not really beneficial to even the power users. Intel beats AMD in everything important to most people consistently, except motherboard price.

>>59114491
>more
Fewer. Fewer, you prole.

File: 1329794785863.jpg (26KB, 500x377px) Image search: [Google]

26KB, 500x377px

>>59122042

ASUS were god tier for AM3+ and that has inspired confidence for AM4. It helps that the crosshaiur hero has both AM4 and AM3+ mounting sockets which is getting people salivating. Certainly ASUS has done alright by me for AM3+ and as such the hero is my AM4 board of choice rightr now with its 8 + 4 pin power support, 8 sata and AM3+ mounting. With that I can basically bolt my existing system to AM4 and the only hassle is software wise.

>>59116158
she's not a child anymore

the whole boxxy fascination is some seriously pedophiliac shit

>>59122137
so comfy

>>59122204

> some seriously pedophiliac shit

You mean perfect for /v/ and /b/?

>>59122229

All we need is some sort of crosshair hero extreme and AM4 is DONE like dinner.

>>59122011
No idea. RGB is fun and all to play with but I don't bank my purchase on whether or not it's enabled or not so RGB isn't really a consideration for me so I haven't researched much.

>>59122137
>It helps that the crosshaiur hero has both AM4 and AM3+ mounting sockets which is getting people salivating.
Holy shit, I didn't even notice this. Why are they not promoting the backwards compatibility in their specs? Do you know if it's just the Crosshair, or if the X370 Pro will be capable of this as well?

>>59122406

A quick look (and based upon the oc3d vid) its only the hero that has both mounting. The hero officially lists AM3+ in its specs, the prime does not.

File: CpRzopqUsAA5vs8.jpg large.jpg (231KB, 960x1280px) Image search: [Google]

231KB, 960x1280px

>>59122204
She's like 25 you retard

>>59122432

>if there is grass on the field, play ball.

>>59122432
the fuck?

Im ten years somebody grows up a little

about time to give people a little respect or be treated like a shithead

>>59122452
*in ten years

>>59120746
3 months from now, probably alongside vega

>>59120746
How cheap are we talking?

I think r5 is q2 and r3 is q3 or 4 I think

>>59120415
How can you even tell since it there isn't an overclocked benchmark for the AMD chip? An OC on the 1700 may push it above or equal to the 7700k.

File: My body is ready ryzen.png (500KB, 1487x757px) Image search: [Google]

500KB, 1487x757px

>>59119358
my computer

>>59122442
>if not, turn'em around and play in the mud.

>>59121798
If its done right, it will be able to tell the load, and oc accordingly, either by shutting cores down, or by downclocking them to nothing,

ryzen master already has 0/2/4/6 core disabler, so when you want a 4 core or 2 core monster, click it and possibly xfr takes it from there.

for me though, i'm considering having a 2 core profile for when I sleep at night that downclocks as low as I can go, the enabling normal clocks and shit when I wake up, a very nice feature considering windows likes to rice my current cpu to max frequency at 50-75% use when it thinks it has down time, will be nice to sleep in a room, and wake up in the same room, but not have the room be near 90 when its only 70 outside.

>>59122020
for all we know the binning is real, personally going to likely go with the 1800x just so I know what i'm guaranteed and remove as much of the lottery as I can.

>>59122137
I want to see the asrock taichi first, as it seems to be the board to beat when it comes to co capabilities, if my cpu will oc itself, I want the motherboard to be the last thing on my list to worry about.

File: speccy.png (91KB, 796x874px) Image search: [Google]

91KB, 796x874px

>>59123133
I'll throw my scores up for a realistic comparison.
Note* I was running youtube and had a bunch of tabs open

>>59123296
god, i can only laugh at how much my cpu and ram suck now a days, yet if I had ddr3 I may not even be considering the initial zen upgrade and may have held off for zen+

>>59121861
I don't think XFR will be able to do it. Maybe improved versions a year or two from now.

>>59123405
XFR is per core, so it will be interesting at the very least.

>>59123315
That's the real question. Zen+ is so close to the first release that it's tempting to prolong the wait.
Good thing I'm not into pleasure denial.
I'm sure you can pick up a cheep second hand 8XXX setup for an in between.

I have a 3570K; it sucks, I can't even overclock to 4ghz without overheating it.

Should I consider a zen upgrade, or stick it out until zen+?

Looking at things performance wise, it doesn't really seem like any of the zen chips are any more than 15-20% faster than my cripplechip

>>59113924
I'm considering the 1700x much more, but I don't plan on getting it until the summer when funds are more flexible. I'm betting on the biggest gap between the 1700x and 1800x being binning. I don't really need the extra few hundred Mhz as much as I just need an incredible multithreaded monster. As long as I can get 4.0Ghz on it I'll be happy.

>>59123443
wtf?
My 3570K is 4.4ghz with a 212 evo and a $90 asus board

>>59123467
>"K SKU's"
>"better binning"
Intel does it again!

>>59123473
what

>>59123473

Given Intel only makes laptop chips and xeons they really do bin in a way that all the fail chips become the dekstop chips.

>>59123405
xfr has something like 20+ sensors per core, and is per core oc, even if it doesn't automatically figure out what the fuck you are doing, manually shutting 2-6 cores down and have xfr still running should be a thing.

If this is able to get a stable oc at any core count, that would be fucking amazing.

>>59123417
from what I can tell, amd is just adding to the fpu to make it on par with the skylake-e that will have avx2 and avx512

Its possible there will be some integer work done, but its going to largely be an fpu expansion.

Once that is done, amds cpus will be no way inferior to intel

>>59123443
>>59123467

Its calld the silicon lottery for a reason.

as for get it or not,

1) it will be better then your cpu all around just on an all core oc
2) you can disable 4-6 cores and likely manage 4.5+ghz and apparently with higher ipc then kabby, beat it out.
3) if you do anything that eats threads, like use chrome, you will have a better all around experience.

but you make a good point, your cpu isn't destitute, your call, but I would honestly say wait till the 6 cores come out, not to get a 6 core, but by that point all the mother boards will be benched, and we will know exactly what the binning is like on the 8 cores, if the 1700 is just as good as the 1800x or if the 1800x is in a class of its own, and by then any issue that is likely there with memory will be gone.

>>59123539
is a 2500k destitute enough to get Ryzen+?

>>59123539
>from what I can tell, amd is just adding to the fpu to make it on par with the skylake-e that will have avx2 and avx512
>
>Its possible there will be some integer work done, but its going to largely be an fpu expansion.
Thanks for the info. No more looking to the hoRyzen+. I'll grab what I can, when.

File: 1454963546625.png (282KB, 827x877px) Image search: [Google]

282KB, 827x877px

>>59113924
Intel fanboys are much greater in number and are up-voting their 7700k processor in the user benchmarks website. It's once again at the top of the charts. This just shows how many more Intel fans there are than AMD fans. How does it feel boys?

>>59123609
fanboys are a bunch of retards

File: 1475311921307.gif (295KB, 700x704px) Image search: [Google]

295KB, 700x704px

>>59123609
>cognitive dissonance is rife in the world.
Thanks for the news.

>>59123563
Honestly, sandybridge gives you enough performance its hard to justify an upgrade at all, but then amd put out an 8 core cpu and gave you a real upgrade path that isn quad to quad.

wait for benches, see how xfr works, and wait for motherboard benches/reviews, chose then.

unless you game at 120+hz, or render video the upgrade is hard to justify, especially sense your cpu had an average oc of 4.6-4.8.

personally, and keep in mind my computer >>59123133

I use my computer like it has 32 threads and 64gb of ram, and I have used it this way since it had 3gb of ram and had xp on it. My upgrade is as much quality of life as it is I can really use the cores, if you don't push your system outside of gaming, its hard to say, but you may be getting close to end of life on the motherboard so a legitimate need may come up for you soon.

>>59123594
don't quote me on what they are doing, this is just their engineers speaking about it that they left some stuff on the table for rysen that they didn't want to leave off for ryzen+ this could mean many things but considering they have jerry rigged avx2 and no avx512 that xeons and skylake e will have, that is a possibility, along with an improved memory controller as it seems that its going to have issues from the get go on ryzen, nothing game breaking, but issues none the less.

>>59123609
hard to be a fan of amd cpus for the last, what was it, 6 or 7 years.

gpu wise, can't complain, they have treated me better then any nvidia I had and lasted FAR longer then they had any right to, my main reason to look forward to vega.

but that said, amd said they were very competitive with ryzen, they said very very competitive with vega... fingers crossed they mean they are prices competitively, or they are very competitive in performance (as in unfucked drivers, or the way they render just gave them the boost that nvidia got with the 900 series)

>>59118124

Mine gets unstable with 4.9mhz overclock

back at 4.2 until I watercool, fuck.....

So all ryzen chips are soldered on.

Can we speculate the temps of the 1700-800x

>>59124125
<60'c

>as long as you are using XFR

Would the 2 extra cores be worth it?

>>59124465
Yes

>>59124465
No

>>59113924
Intel.

>>59124465
No, the 1600X is going to be an absolute monster in price/perf.

>>59124465
It depends on what you're comparing the 8-core with

>>59124489
>>59124551
t. poor subhumans

You can disable cores on the fly for higher overclocks btw. If you have the money there's no reason noto get a 1700x/1800x.

File: Bob.jpg (5KB, 171x171px) Image search: [Google]

5KB, 171x171px

>>59114014
Girls just wanna have fun.

>>59117252
https://u.nya.is/jdqfvk.mp4

Curious...

>>59118085
It'll be the best of the rejects.

>>59117003
Most boards will be fine. Just don't use the ASMedia USB 3.0/3.1 ports for now, only the ones that come directly from the chipset.

>>59116546
Take care with the VRMs on this one. It's a 6+1 phase arrangement, and even if they are backed by digital IR VRMs, you might find that your maximum is around 4.2-4.3GHz, which XFR won't get you to.

>>59119358
A R5 1500 is three to four times as powerful depending on the benches. Anything more powerful than that will just be more gravy on the plate. You're long overdue for an upgrade.

>>59125265
Well when it comes to disabling cores, there's variance all over the place. For example, it could be a chip where 3 of the 4 2core modules were really good but the 4th had a flaw in it which causes it to need to be disabled.
So it'd OC quite well even though it's had an entire core disabled.

And chips like that are the 1600X I bet.
If you separated their processors into pockets of probability, assuming some kind of distribution of chance, you'd find it lines up with their product line pretty well.

>>59118907
Kaby Lake has a clock speed advantage at stock clocks, even under heavy load. The R7 1700 has to clock at 3.7GHz or less during gaming benchmarks.

Like, what exactly are you trying to prove here? AMD directly admitted that Broadwell-E levels of IPC was their goal. Currently they're slightly above that, and only clock speed is holding them back thanks to a slightly immature process not made for desktop CPUs.

>>59120488
>However Intel seems better at everything else.

Only if you're benching against Kaby Lake or Skylake, which have clockspeed advantages thanks to a more mature silicon process. Against Broadwell, AMD is on equal footing. Against Haswell, it has Intel beat.

You're reaching quite far to find something wrong with the performance on offer here.

>>59125454
The R7 1700 has 8 cores with a 65w tdp rating. This processor seems to be better utilized for budget workstations aiming for low power consumption. I wouldn't be surprised if we get much better single core performance from the lower end chips since they would have less cores to deal with.

>>59125478
less corns?

nigga wtf. gtfo. im getting a ricin becuz of the corns. future is now. more cornes == future

File: 1478101374006.jpg (382KB, 1920x1560px) Image search: [Google]

382KB, 1920x1560px

Why shouldn't I just buy the 1700?

whats the fucking difference between the 1700, 1700x, 1800, and 1800x

I really need new pc right now (my main is laptop which is dying), I'm going for more expensive X370 MOBO + 1700 w/ stock due to tight budget then swapping out for zen+ with water cooling meme later

>>59125501
WE don't know, you'll have to wait until the NDA is lifted and actual reliable benchmarks come out.
Personally I'd trust the tech report's the most since they're transparent about testing and kinda a small enough site for intel to ignore.

>>59125478
Agreed, but it also seems like a decent all-rounder. That average FPS being on par with an overclocked Kaby Lake is quite good, and I'd love to see the frametime charts for benches like these. I'll be testing this myself when my review samples come in.

>>59125501
xfr, it's basically auto-overclocking
They're better binned so the x processors are also guaranteed to higher frequencies than the non-x chips. In fact, if you try hard enough, you might be able to reach clocks higher than the auto-overclock.
Meanwhile, the 1700 is rated at a 65w tdp compared to the 1700x at 95w tdp

>>59125501
1700 is maybe just downclocked because it was unstable on higher clocks?

>>59125501
We don't know yet. Presumably the 1700 has trouble going over 4 - 4.1GHz, but that's just one sample.

1700x/1800x most likely overclock better.

>>59125501
>whats the fucking difference between the 1700

65W TDP, 3.0-3.7GHz range, 8 cores, 16 threads, full cache. XFR might boost it to 3.8GHz, and it might clock up to 4.0, but only with a good sample and a good board.

Mostly equivalent in gaming performance to a stock-clocked Haswell-E chip and maybe Skylake/Kaby Lake in a lot of games.

>1700x

Better binned than the 1700. 95W TDP, 3.4-3.8GHz window, with XFR boosting that maybe to 3.9GHz. Might hit 4.1-4.2GHz under water with a good sample and a good board.

Beats out Broadwell-E in benchmarks, might beat it in games, definitely gives Skylake a run for its money.

>1800x

Top binned part, 95W TDP with a 4.0GHz boost clock. Will probably run up to 4.3GHz on good samples under water with a decent board.

Generally beats Broadwell-E into the grave. Will still lose to Kaby Lake in single-core benchmarks which scale with clockspeed.

Ryzen's voltage regulation and power requirements are much more strict than Bulldozer derivatives, so a board with at least an 8+2 phase setup will be OK for a lot of overclocks.

Both the 1700X and 1800X lack a bundled cooler, but this will change eventually, when both are bundled with the Wraith Max. The 1700 is bundled with the Wraith Spire.

>>59117440
1700x price pains me so

>>59125592
>Both the 1700X and 1800X lack a bundled cooler

WHY THO

File: Speccy64_2017-02-26_02-31-07.png (169KB, 1232x1028px) Image search: [Google]

169KB, 1232x1028px

>>59123296
got me stomped on diskmark even after I get my 4 disk raid going.

>>59123443
Sorry for your loss.

>>59125439
They have 2x 4 core modules. Another anon previously stated that the only way for the binning process to work is if the modules are symmetrical in core count. So through any number of processes of elimination, they would be determining the best 3 cores per module to leave functioning.

The thing that's making the 6 cores so tasty other than the price, is that they're getting >2Mb l3 cache per core. They really are looking to be the oh so sweet spot.

>>59125645
If you're going to spend 4-500$ on a unlocked cpu, you generally have no intention to use a stock cooler.

File: comparativo-ryzen-7-1700x-vs-intel-core-i7-6800k_chamada.jpg (62KB, 640x360px) Image search: [Google]

62KB, 640x360px

>>59117440
>1700X (399$) is going against a i7-6800K (429$)
it's 38% faster.

File: my build.png (21KB, 503x435px) Image search: [Google]

21KB, 503x435px

>>59125658
>If you're going to spend 4-500$ on a unlocked cpu, you generally have no intention to use a stock cooler.

That's fair I guess

Now all I need to do is sell my PC so I can use that cash towards a new one with Ryzen. How much do you guys think I can get for this? I'm not sure if I want to sell the whole thing, or just sell the mobo, cpu, ram, and graphics card, and then keep my PSU, hard drives, and case.

the gpu is a 290x btw

>>59125645
there were leaks of them with Wraith Max stock cooler, but looks like those bundles won't be ready at release

>>59125645
I'm guessing lack of volume. AMD has reportedly been making a million R7 chips for this launch. I don't think their cooling partner who makes the Wraith Max is able to meet around half a million orders in time for the launch, which is why two OPNs exist for the 1800X and 1700X - one specifying a bundled cooler, and the other one wothout.

Also, maybe AMD wants to chop out the price of the cooler, but keep the price the same for launch so they can earn more money from launch sales.

>>59125730
>but keep the price the same for launch so they can earn more money from launch sales.

And to add to my previous point, later on they could reintroduce the 1700X and 1800X with the Wraith Max bundled in, without a price change. That would be a huge PR coup for them.

>>59125501
1700 can't clock for shit, 1700x can clock but has high leak currents, 1800x is the shit but AMD would rather make Naples out of these and sell for higher profit margin thus high priced flagship.

1700 - cheap workstation
1700x - 8c for those with more sense than money
1800x - 8c for those with more money than sense
1600x - the enthusiast gayming price/perf king of the 2017
1500 - good enough
rest - peasants

>>59125658
The L3 works idenpendent of the cores and can either be the full 16Mb or half disabled at 8Mb

>>59113924
if the pricing would be correct for AMD, like the 1700x for 300$, the 1800x for 350$ and the 1700 for 225$... but right now it's a pile of shit and an i7 7700k is better for the price.

File: crystal dick mark.png (642KB, 803x682px) Image search: [Google]

642KB, 803x682px

>>59125658
>got me stomped on diskmark even after I get my 4 disk raid going.
>even after I get my 4 disk raid going.
Anon-kun, you're making me blush.
Good to see that it doesn't weigh disk performance too much. I was umming an erring about my rig because of how high a score that got.
Guess we'll bash digital dicks again come Ryzrn. Cheerio.

File: 1485535781217.jpg (27KB, 480x360px) Image search: [Google]

27KB, 480x360px

>All these posts referencing XFR
XFR is just another +100MHz on top of the boost clock that activates whenever it's stable. They're calling it XFR because it depends heavily on binning. Calling it a second boost worth 100MHz would probably be illegal since some chips will never actually activate XFR due to binning issues. It's literally nothing more than another p-state.

>>59125921
>XFR is just another +100MHz on top of the boost clock that activates whenever it's stable
>just another +100MHz
Citation?
>sponge.jpg
Oh, it's /v/ damage control.

File: 1482097056421.jpg (563KB, 1920x1280px) Image search: [Google]

563KB, 1920x1280px

>>59125944
>Citation?
Just wait and see.

>/v/ damage control
I'm on a Phenom II right now and will probably get a 1600X. I personally believe that Ryzen has surpassed Kaby Lake in IPC and we'll see that after March 2. I'm as AMD fanboy as it gets, and I'm just saying the truth.

>>59118907
fake and gay

File: DiskMark64_2017-02-26_03-37-07.png (98KB, 400x320px) Image search: [Google]

98KB, 400x320px

>>59125690
As is you shouldn't have an issue selling it for 5-600$. Could probably get more, all depends on how morally driven you are, and how stupid the buyer is.

>>59125861
That makes sense and more in line with my original thoughts / speculations. Someone else had been raving about how it was tied to the cores, and so it would have to be partially disabled during the rebinning process.

>>59125910
Well, it would take something like 8 HDDs to come close to that, and as I'm planning on a raid 10 for redundancy, I actually only get half the write speed on top of that. But those 4k speeds seem kinda low. Here's my currently only ssd because I pulled my 480GB to go into a laptop since I have the hdds for the raid now. Just have to get a drive cage modded into my air 540, then the next step is a 128gb nvme ssd for cache and page filing. The itch for moar never goes away.

>>59125944
According to linus tech tips after the launch conference. https://www.youtube.com/watch?v=3rUndzpdo1I

File: N7xTXdi[1].png (99KB, 628x879px) Image search: [Google]

99KB, 628x879px

Hmm

>>59126062

Fucking '11' with minimal bios fuckery and OS interference.

>>59126062
why would you use hardware option instead of changing the oc stage in bios?
It does the same thing right

>>59126125
Bios might give you flexibility for reaching higher overclocks at lower voltages. This knib seems more like some sort of preset for guaranteed clocks.

File: Disk life.png (56KB, 715x828px) Image search: [Google]

56KB, 715x828px

>>59126044
>But those 4k speeds seem kinda low.
Yeah they are, think it came down to the stripping size I chose. If I went larger KB's then I'd be around 32-34 read 210 write. (guesstimate)
They're cheep SSD's I'm using. Kingston U400. Was more cost effective to go RAID 0 on two drives than one larger.
>I'm planning on a raid 10 for redundancy
Yeah, that's a fair precaution. I'm not fussed in doing that because I only have generic shit on my raid array. So if failure happen it's only really the time of an OS install that I'm out.
But here's my S.M.A.R.T so you get an idea what you're in for. Picture Related.
>The itch for moar never goes away.
I hear that. SSD's are nerd crack. I want my whole system to be on them.
I've got a few HDD's that were meant for additional storage that are unopened because I just can't go back to that.

>>59126062
>these go to 11

>>59114284
Notice the lack of replies

>>59126194
because it's a retarded remark?

File: DiskMark64_2017-02-26_04-12-20.png (99KB, 400x320px) Image search: [Google]

99KB, 400x320px

>>59126158
Well, my stuff will be mostly expendable / recoverable. But once you get a couple terabytes of shit, you don't want to have to redownload it all over again. I'm already mentally preparing my self for horrendous access times since HGST 7k4000's have notoriously long spin up time I've discovered. I feel like I'm going to be disappointed, though. This drive is a lot slower on sequential than the other one I ordered with it.

>>59126313
Yeash, how can they even sell those. Then again I'm awashed in new tech.
If ya don't mind a bit of noise, use 10k RPM scuzzy drives.
>don't want to have to redownload it all over again.
My stuff is just normal back ups. Nothing fancy. Just a few drive I swap in to pull shit them whipe my array.
Only thing I can see myself re downloading in the TB range is a bunch of games I 'll never play.

>>59113924
1700X fits better to my wallet

>>59125861
I'm still of the opinion that disabling a core on Ryzen also disables the L3 cache controller. I don't see why the R5 1600X needs 16MB of L3.

>>59125876
$350 for stock performance on par with a Core i7-6900K? $225 for a R7 1700? That leaves AMD with little room to manoeuvre for future price wars with Intel, and it leaves them no space for a 6C/12T, 4C/8T, 4C/4T, and 2C/4T rollout with at least a $50 price difference between different core counts.

AMD does not need to undercut themselves with pricing, and they don't need to be seen as the budget alternative anymore.

>>59126125
Easy overclocks for people who don't know what voltage to set, and just want to twist the dial, run some stress tests, and settle on a level that's OK with temps.

File: 1479362888960.png (364KB, 599x563px) Image search: [Google]

364KB, 599x563px

>>59114343
>a literal 4 core processor

just fuck my shit up senpai

>>59126062
So according to all info gathered
1800x 4.4 Ghz all 8-core overclock
1700x 4.2 Ghz
1700 4.0 Ghz

>>59115026
oh yes the majority doesnt use one thread..

meanwhile on the real world that multitasking is a thing says otherwise..

>>59119439
>Comparing the quad and single core scores from our samples shows a relative 4.3% boost on the single core score so it’s possible that turbo wasn’t fully engaged
>windows reported a peak clock of 3.5 GHz
Not an argument till more benches, especially XFR'ed ones will come