I lost the silicon lottery. I need 1.4v to run 3.9ghz.

Same but on a 1700. Saved $280 for 100mhz less, though

rip same

>>61309352
What board do you have and what ram speeds you running?

I could run 4.0, 4.1 if i went up to 1.45 on an asus b350.

Shitabyte x370 k5 gaming can't even get 3.8 because it has no fucking LLC and the BIOS is basic as fuck. can't even train my memory.

Let me get back to you once I find how much difference that 100MHz matters

Need 1.425v to get 4.0 GHz, can't get 4.1 even on 1.5.

>>61309352
I get 4ghz at 1.37 on my 1700

File: maximus-v-gene.jpg (1MB, 1500x1426px) Image search: [Google]

1MB, 1500x1426px

I won big five years ago. I own a dreamy 3570K that get's to 4.6 really easily, and 4.8 with still manageable temps.

I want to stop being a threadlet though, so now I'm super nervous about next time I have to play the great lottery.

File: 1288367953014.jpg (26KB, 303x288px) Image search: [Google]

26KB, 303x288px

>>61309917
>get's
What the fuck was that?

>>61309352
>80% yield
>op still lost
wew.
maybe shit mobo?

>>61309917
Just buy a couple then return them

my 1600 does 3.7ghz on 1.4v, can you feel my pain ?

>>61309352
hahahahahahahahahahaha
you lost :)

>>61309957
>what is binning

>>61309701
This, in what real world situation would 100 MHz make a difference?
overclockers are cancer

That's what you get for buying Ryzen. Worst part because Intel has been releasing same IPC for last 5 years they also went full retard and honest to god if I had to buy a CPU now I'd simply not buy one at all. Anything involving AMD is garbage, Intel on the other hand tries to have 3% gains by setting my computer on fire. Thanks Ayymdrones.

File: vrbJic9.jpg (112KB, 720x960px) Image search: [Google]

112KB, 720x960px

>>61310319
>I-Intel never loses the silicon lottery
>W-winner in every box!

Fuck off, you're obviously a fanboy even when you pretend not to be.

tfw R5 1500 sits at 3.95Ghz at a conservative 1.28 v

>>61309352
Just use a lower clockspeed. I run my 1700 at 3.7/1.28v.

>>61309352
>>61309819
>>61309844
>>61310213
>>61310398
Do all you Ryzen owners mind terribly posting Geekbench 4 results?

>>61310449
>geekbench

>>61310449
Why would I install that garbage?

File: 1499454016996.jpg (37KB, 400x491px) Image search: [Google]

37KB, 400x491px

>>61310449
>geekbench
Nice try sand merchant

>>61310625
>>61310627
>>61310655
Geeze sorry. I just thought "hey it's gratis"

>>61310213
>>61309352
How about 3850-3875 MHz? The 25MHz increments might allow you to find some sweetspot where you are close to 3.9GHz but can drop the voltage a bit.

>>61310672
Get back to /soc/ you cancer.

>>61309352
Thats pretty much the norm for a 1700.

File: cpu-z.png (17KB, 403x402px) Image search: [Google]

17KB, 403x402px

>>61310449
No, but you can have this cpu-z

File: 1358606392967.jpg (22KB, 262x228px) Image search: [Google]

22KB, 262x228px

>>61310759
According to CPU-Z's records someone has reached 4.9ghz with the R5 1600

>>61310672
Geekbench is an awful "benchmark" with some very transparent biases. No one in their right mind would use it for anything other than comparing two chips of identical architecture and identical software platforms.

>>61310791
They probably did too, for about one minute, on liquid nitrogen.

I got 4ghz with high bclk

>>61311011
5.3 all core oc is the highest official 8 core I know of from ryzen.

File: 1.79.1.png (15KB, 403x402px) Image search: [Google]

15KB, 403x402px

>>61309917
I'm in the same boat wanting them threads but >>61310759 that single core

>>61311228
ryzen kick ass on that, before someone "pay" them to nerf the score

overclockers are the biggest faggots out there

>let me buy a non-top of the line CPU and buy a $220 motherboard $70 heatsink fan and some special Overclocking memory so I can overclock my CPU by 400mhz!

or you could've just gone with a higher end component and used OEM cooling to get the same result, much cheaper too

>>61310319
>bought a g3258
>only 4.3GHz at 1.25V
>4.5GHz at 1.35V
>at that point it sucks electricity as much as my 45nm athlon ii x4 at 3.7GHz
>mine is not terribly bad, some reported worse.

>>61311388
yeah but these are people who think overclocking makes them hackers

>>61311388
>i have no idea what i'm talking about

>>61311388
Retard

>>61311464
>>61311478
>let me buy a 1600 because I can overclock it to 4 ghz, easily!

>$230 for CPU
>180 for riced mobo
>80 for a huge lump of heatsink + fan or $100 for a closed loop set up
>$140 for special OC friendly ram
>$60 for more case fans because you want your CPU to run 0.05C cooler
Or you could have gone iwth a 1600x
$260 for CPU
$80 for motherboard
$0 for heatsink fan because you're using the OEM
$80 for RAM because you're normal


you guys are the retards

>>61311446
yeah like 90% of the ppl overclocking and getting retarded aftermarket watercoolers don't need it at all

>>61311593
>1600x
>"OEM" HSF
>being this ignorant
Just stop

>>61309672
>Buying a Gigabyte motherboard

What the fuck did you expect

File: 1488544970262.jpg (31KB, 387x461px) Image search: [Google]

31KB, 387x461px

>>61311593

>>61311658
the 1600x turbos to 4.0, it should be good enough for your video games, boy.


OEM/Stock is widely accepted as appropriate nomenclature
>>61311676
haha nice reaction image, did you fall into a coma in 2011 or something?

>>61311681
>OEM/Stock is widely accepted as appropriate nomenclature
Not either of those guys but you're an idiot and you haven't realised it yet. I'll leave you to do some frantic googling.

>>61311681
>1600x comes with its own cooler
Dumbass faggot
Do a little Google before you shitpost

>>61311681
see
>>61311464

File: 1300044776986.jpg (17KB, 250x250px) Image search: [Google]

17KB, 250x250px

>>61311388
>OEM cooling

>bought a 1700x with the single most expensive x370 mobo , and I run stock settings .

I'm happy .

>>61311867
>and I run stock settings .
Don't you mean OEM settings?

>>61311867
>single most expensive x370 mobo
c6h or x370 titanium?

>>61311593
None of the X ryzen cpus come with a stock heatsink.
None of the K/X intel cpus come with a stock heat sink.
As you can see the highest tier cpus from both companies don't come with stock heat sinks. Do you see why you're a retard now?

>>61311934
I guess . Other than putting my ram at 2933 mhz I haven't touched default xpu settings . It's very fast as it is and I don't want to deal with temperature and voltage

>>61311942
The titanium because that's the one that's most expensive one, didn't you read his post?

>>61311942
The meme titanium with the mediocre vrms . I wanted a taichi but they were out of stock for months .

>>61311934
The OEM settings are the stock settings you stupid nitpicking retard

>>61311934
funny

>>61311989
You still haven't googled what OEM means and how "OEM settings" can vary across manufacturers by the sole reason of what OEM means? You dumb retard

>>61312025
OEM/Stock are interchangeable

shut the fuck up

>>61311951
>Replying to the bait

>>61311974
>out of stock for months .
You lucky
Taichi with shit bios wont do any better

Should I delid my 4770k? Sounds scary

>>61311951
my point still stands. You can buy a cheap $30 heatsink fan on your X series Ryzen CPU instead of opting for a seemingly "cheaper" non X variant and spending $60-80 on an aftermarket HSF and a motherboard with more stable components for overclocking.

>>61312068
>OEM/Stock are interchangeable
Stock means whatever came with your equipment, from settings to attachments. Like a car.
OEM means somebody who makes a part to be sold by somebody else.
Literally two different things. An OEM part does have stock settings, but that "stock" setting would and could be different things across different resellers. Just accept it, you are an ignorant baboon who meant "stock" but instead said "OEM"
>I'm a ricer, so my Civic no longer has OEM engines
Apologies in advance to any /o/tists

File: DERAmm5U0AA40Fg.jpg (112KB, 734x692px) Image search: [Google]

112KB, 734x692px

>>61311593
You don't need OC-friendly RAM at all, you OC via multiplier these days, there is no FSB anymore, anon.

With Ryzen though, 2933 or 3200 MHz DDR4 improves perf though, regardless of OCing CPU or not. People actually OC the RAM to those speeds with slower modules too, tho.

What you have a point in, is that particularly R5 1600 is best for milder OC, where you use the stock heatsink it has and which is not the worst (and yeah, R5 1600X lacks any), and you pair it with some reasonably cheap B350 motherboard, so that you don't waste money that could have gone towards R5 1600X or even R7 1700.

Your OC will be smaller but hey, if you get sa 3650 or 3700 MHz, still good.

File: 1498026128469.jpg (13KB, 319x331px) Image search: [Google]

13KB, 319x331px

>>61312114
It's a sppoker

>>61312131
Lel to match the stock hsf on the 1600 you will have to shell out more than $30

File: 3cd94mB.png (928KB, 2560x1440px) Image search: [Google]

928KB, 2560x1440px

>>61312131
I'm OP. I'm still running the 1.4v on my 1600, on a $30 212 Evo cooler. Temps are similar to pic related. My Gigabyte B350 motherboard cost around $100.

>>61312223
>stock hsf
Don't you mean OEM HSF?

>>61312246
No

>>61312384
But they are interchangeable

File: 1452274458515.jpg (92KB, 768x783px) Image search: [Google]

92KB, 768x783px

OEM POSTER OR STOCK POSTER,


WHO WAS IN THE WRONG

>>61312512
>tfw cryptocucks either made GPUs overpriced or out-of-OEM

>>61312545
OEM systems with popular crypto cards aren't overpriced though:)

>>61312097
What's wrong with the Taichi? From what I've heard it's the best AM4 motherboard.

I have an NH-D15. Should I get a 1700 and OC it or just get a 1700x?

>>61312677
Get the 1700. X version is for people scared of OCing.

>>61312636
Assrock got shill here alot, Taichi got good vrm and that about it

they already eol X370 Gaming K4, i think it best to stay way for this company

>>61312697
Great English, Gigashit shill

>>61312677
1700x and OC it

>>61312818
I would take ASRock over Gigabyte actually, I think. Gigabyte always fucks up things in the BIOS.

I got F2A82X-UP4, and it didn't get any updates after mid-2013 IIRC and it acts funny - for example it sets fans to max and activates the optical disc drive to max speed if there is a disk inside when you hibernate the PC. Lots of noise before the power is finally shut.

Stilt guy says stay away from them, too. It probably isn't a bad brand (just not ideal), still better than MSI. ASRock and GB might be on par, some issues there, other issues here.

I used FM1 ASRock mobo and it was sorta fine.

>>61312695
>>61312936
not that anon but is there an difference in terms of overclocking from the 1700, 1700x, and 1800x

i heard 1700 is most optimal.

my 1600 hold 3.7 with 1.2375
stay jelly guys

>>61312984
The K5 is one of the worst AM4 overclocking boards, which has soured a lot of people's opinion of Gigabyte. It's embarrassing how bad the BIOS is for a $170 board. I had a $120 Intel board with better features than it.

>>61313040
they have the best b350 mb though the gaming 3 as the same option as much more expensive x370 mb

>>61309352
My 1600 can hold 3.7Ghz at 1.28v with Prime. I set it 1.293v just to stay safe.

File: 1442107956381.jpg (62KB, 600x600px) Image search: [Google]

62KB, 600x600px

>run Cinebench multi once
>"Fuck it it's stable"

is my R7 1700 @ 3.9ghz 1.3v (i honestly havent tested lower voltage just tried this as a baseline and was expecting to have to bump it) winning the silicon lottery?
I can't go higher than 1.325 if i recall (my bios has some bug where pumping more than 1.325v makes the CPU clock down to 1.55ghz)

File: o0700100012696567516.png (127KB, 700x1000px) Image search: [Google]

127KB, 700x1000px

>>61309352
My 1700 does 4.1ghz @ 1.375v :^)

Anybody else in the same boat as OP but don't mind because you see your CPU as a stopgap until Zen+?

File: valuables secured.jpg (1MB, 1280x966px) Image search: [Google]

1MB, 1280x966px

>>61313202
I can run cinebench on my 1800x at 4.1 at 1.4, but its in no way stable. The thing is barely stable at 4.0 at 1.4, I'm convinced that one of the cores is absolute shite and keeps throwing WHEA errors.

>>61312984
>>61313040
>tfw you have a k7 for the BCLK generator
>tfw it's sub par power delivery might be the reason for your 1800x not holding an overclock.

fuck you gigabyte release the fucking new bios already

>>61313916
my 1700 was an upgrade from my 8350, and probably gonna hold me over til Zen2 (7nm) because the IBM 7nm process is MADE to run at 5ghz, with just Zen+ IPC improvements (they predict 15%) and 25% higher clock speeds, I'm gonna be Randy in the internet-less episode of southpark.

>>61309352
>Needing to OC an AyyyyMD CPU to make it usable

lmfao

>>61314025
I returned my k7 for the crosshair vi because the bios is so lackluster, doesn't even have llc or any advanced features for overclocking

>>61314173
k7 has llc at the very least, now. It's not horrible but it's certainly not the best

File: 1325452613326.png (158KB, 500x281px) Image search: [Google]

158KB, 500x281px

>>61311593
I bought a 1600 and OC'd it to 3.7GHz on stock voltages and stock cooler.

Where is your god now?

>>61314150
http://cpu.userbenchmark.com/Compare/Intel-Core-i5-7500-vs-AMD-Ryzen-5-1600/3648vs3919

>>61313794
Nice
What board, it 1.375v bios or cpuz ? and what Soc voltage

>>61314150
>Needing to delid an Intel CPU to make it OCable
>Needing to buy a K series CPU and Z270 motherboard

Kill yourself.

>>61314150
i5's are pretty terrible unless you overclock nowadays.

>>61314398
and before some asshat comes in "HUR +8% SINGLE THREAD AYYYYMD BTFO"

These are stock clocks.
Lets see, i5 at 3.4ghz, r5 at 3.2ghz
3.4/3.2 = 1.0625.
3.4 is a 6.25% higher clock speed.

Just ignoring that ram speed effects ryzen benches quite a bit (2400mhz -> 3200mhz put my r7 1700 from just below to slightly over an i7 6950x in multithreading and an increase of 10% in single threading)

That means that 1.75% of that speed on the i5 comes from *something* in the computer, even *IF* you attribute it to IPC from the arch, you can't OC the 7500 at all as far as im aware, whereas 4ghz is common on every non-8 core part, and pretty common on the 8 core parts.

Then theres the multi-core performance, and the fact the CPUs cost virtually the same.

File: 214_keks.png (30KB, 454x534px) Image search: [Google]

30KB, 454x534px

>>61314499
they both boost stock you dumb fgt
r5 1600: 3600mhz+50mhz xfr
i5 7500: 3800mhz 1 core turbo
3800/3650=1.04
the i5 is clocked 4% higher giving it 4% better ipc according to this test
>4ghz is common on every non-8 core part, and pretty common on the 8 core parts.
Not on the stock cooler or a cheap b350 with no LLC. Zen efficiency also goes in the dumper at the high vcore required to reach the fabled 4ghz.

>>61309352
This is likely a problem with your motherboard and not the CPU.

>>61311388
I overclock for the single core perf. Sometimes I play games via Wine and need to hammer out tons of converted graphical calls to keep my GPU fed. It's air cooling though so I only get an extra 500MHz. Free perf though.

>>61312114
It's easy. I've done the vice, wood block, and hammer method 5 times. Never damaged, and always quick. Hard part is scraping the damned glue off. My temps fell by 10c by replacing with CLU.

>>61312984
Maybe when they're not dead. I've seen 60% of their boards just fail for no real reason so far. Never again.

>>61309352
i got my xeon e3 1230 v5 to 4.5ghz at 1.36v. did i win?

File: 1476471183765.jpg (57KB, 1000x1000px) Image search: [Google]

57KB, 1000x1000px

Running my Ryzen 1700 at undervolted stock settings (offset of -1.1V) and 3200Mhz memory
Runs well

>>61313118
Gigabyte has teh worst B350 boards, at least considering VRMs

>>61315238
you have no idea what ipc is do you, dumb fuck

>>61314025
The K7 and gaming 5 have really solid 6+(2x2) VRM

Is the crosshair 6 a meme? I've had good experiences with Asus in the past

>>61312974
The 1700 still measures temps differently afaik. In a good way for overclockers.
Other than that, the other chips may r may not be easier to get a high clock on because they're binned for better stock clocks.

>>61310449
dude what the fuck, my bone stock 1600 spanked the 6700k/5960x in cpuz 1.77 then I updated because the cpu family was unknown in that build just said "amd k17" or sth

now the tables are turned and the intel processors get a big boost (and all the fairly recent AMD processors are expunged from the reference charts)

>>61317035
Intel's pockets go deep

>>61316990
Well it's not so pretty, but I think it's a pretty good board.
https://www.youtube.com/watch?v=3xfLYLRsPZ0

>>61317035
That just sent chills down my spine

>>61311388
>>61311593
standard and x versions are identical except you just have to manually overlock the standard one. It's a marketing trick for retards like yourself. They both overclock the same except x comes without a cooler and is more expensive you fucking braindead imbecile.

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

174KB, 696x678px

>>61314419
>What board,
Asrock X370 Killer SLI/ac
>bios
2.50
>what Soc voltage
default

>>61316833
*-0.1225
And I'm running a gigabyte X370 gaming 5

File: enough.jpg (39KB, 700x500px) Image search: [Google]

39KB, 700x500px

>>61317035
Is this confirmed?

>>61310319
> Intel on the other hand tries to have 3% gains by setting my computer on fire.
> Thanks Ayymdrones.
How do you even live with that logic.

>>61317035
|
|>
|
|3
|

>>61318155
They nerfed the scores because Ryzen "overperforms"

The branch predictor is that good.

File: january jones.gif (2MB, 364x433px) Image search: [Google]

2MB, 364x433px

How far can you OC the r5 1400 with the stock cooler and get respectable temps?

>>61317087
>>61310281

>>61318237
I got 2.2PetaFlops outta mine but it ran hotter than the sun and destroyed the martian atmosphere causing a mass migration to earth and the fucking lefties made us bring the nigs with us so its all going to shit here anyway also im a time traveller.

File: specs ryzen.jpg (353KB, 973x940px) Image search: [Google]

353KB, 973x940px

3.8 @ 1.325v is the sweetspot for me on a 1700x.
I tried 3.9 briefly but voltages where too high to my liking to be fully stable (1.38-1.39v).

File: puff.jpg (56KB, 408x491px) Image search: [Google]

56KB, 408x491px

>>61316170
>play games via wine

Not doing too bad, I got 1.3v for 3.8Ghz on my 1700

awwwww ye.

>>61318340
1.38-1.39v is fine, what cooler you using

>ryzen 1800x [email protected]
I won the lottery?

>>61318425
Thermalright le Grand Macho RT.
I may bump it up 100 Mhz at some point but the higher temps and fan noise doesn't seem worth it. It's pretty much silent now.

File: 1800x.png (49KB, 849x621px) Image search: [Google]

49KB, 849x621px

>>61318463

>>61318529
>52W @ 3.3ghz 1v
notebooks with full ryzen when? I need this.

>>61309352
>not waiting for the 2nd gen ryzen where they sit on 4.5-5ghz stock volts
muh dood

I'm in-between, need 1.35v to run 3.9ghz on a 1700

>>61318529
how do one make this, are it even tested for stability ?

>>61312231
How does the stock cooler compare to the EVO?

>>61309394
>>61309352
>>61309440
ryzen is pretty bad for overclocking, people bought it for stock clocks and many threads... so if you wanted to OC, why buy one?

>>61314333
What are your temps during idle and prolonged load?

>>61317035
the scores were fucked due to the new architecture, and they fixed tit.

>>61317035
Yeah they fucked the shit out of it. Don't use CPU-Z's benchmark, it's trash. http://www.guru3d.com/news-story/new-cpu-z-upgrade-lowers-ryzen-performance.html

Can you overclock ryzen with boxed stock coolers

>>61319423
The 1600's boxed cooler is enough for some OCing.

>>61312818
What's bad about Gigabyte, and how is Asrock better?

>>61319438
Gigabyte's B350 boards have junk VRM

>>61318155
just use an old version and find out
http://www.oldapps.com/cpuz.php?old_cpuz=48

>>61319448
Does Asrock offer the best B350 boards then?

>>61319551
Avoid asrock
https://forums.anandtech.com/threads/asrock-x370-gaming-k4-is-already-eoled.2509665/

>>61310672
>using any software that uses ICC
What kind of faggot would do that?

>>61319575
Asus then? Last time I asked /g/ was shitting how "ANUS" is so the worst overall for AMD. MSI?

1.38v @ 3.8 here lmao
got the 1600

>>61319609
MSI only has one good B350 board.
ASUS also has one good B350 board
Asrock has a few.
Gigabyte ones are all shit

>>61319661
Of all these 4 manufacturers which one B350 board would you personally recommend?

>>61319683
Asrock or ANUS because the MSI one is too overpriced and enters into X370 price ranges

>>61318237
1400 stock cooler is thinner than 1600 and 1700 ones. Like half as thick. Because of that and the fact that the shitiest crystals go to for 1400, 3.5 or 3.6 if you have luck. With noctua u12s (or similar performance cooler) it can hit 3.8 on 1.35 as it is not smart to push it more. Get 1600. They have good ones here and there, 1/3 of them can make 3.75 on 1.250 or 3.85 on 1.300 and that is good enough. Friend of mine have really good r7 1700 that can make 3.8 on 1.175 and is cold as death.

>>61319697
Thank you. Is there anything negative I need to know about Asus?

>>61319734
No, it's just a joke from the launch where ANUS's most expensive motherboard(also most expensive AM4 motherboard) were bricking themselves left and right, BIOS updates fixed that.

>>61319423
>stock coolers
you mean OEM coolers?

>>61319711
Other than RGB lightning are all the stock coolers from 1600 onwards the same?

>Buy i5 4670k, mobo, RAM, 212 EVO, case for $175
>not even stable 4.1GHz stock volts. Don't want to overvolt due to noise. Settle for 3.9GHz undervolted.
>buy i7 4790 system for $250
>swap processors and flip system for $330
>now can't even overclock. 4 core 3.8GHz @ 1.007V.

Jewed by Intel again.

>>61312189
This is probably One of the Most autistic posts I've ever seen on /g/
>stock != oem

>>61320566
you literally just added a tl;dr

>>61309352
Holy crap, I'm running my 6700k on 4.5 ghz with only 1,27V

Glad I didn't fell for the AMD meme

>>61321937
>hurr hurr Intel never loses the silicon lottery
You're retarded.
>>61319835

>>61321951
The difference being 6700k's hitting 4.5 ghz are fairly common while Ryzen chips that go over 4.1 ghz on air cooling without crazy voltages are extremely rare.

>>61321976
>4.5 GHz
That is the boost clock on a i7-7700K, you aren't exactly breaking any records here.

>>61319365
Dude stfu, stop pretending like 90% of intel chips can OC to 5 GHz. Both AM and intel only have a few chips that can overclock more than 200-300 MHz.

File: 1499656993053-g.jpg (96KB, 1280x720px) Image search: [Google]

96KB, 1280x720px

My 1700X will do 3.8ghz at 1.3v which is what I'm happy with. However 4.0ghz is unstable regardless of how much voltage I throw at it. On the flip side, I can do the stock clock of 3.4ghz on all 16 threads while heavily under-volted. Prime95 3.4ghz 16 threads @ just under 1.0v if manually set with voltage LLC set manually as well.

Same with my ram. My kit is only 2400mhz at CL14 at 1.2v. with 1.25v I can get the timings down to CL12 at the same speeds.

This is all on an Asus ROG crosshair with 1102 BIOS if it matters. If you bought Ryzen to be an overclocker, you bought into the wrong platform. Underclocking on the other hand is nice.
>Mfw 16 threads all encoding 1080p Blu-ray remuxes while also gaming AAA stuff at the same time with no fps drops or stutters.
>mfw doing all that and the CPU uses less than 70w, stays under 60C, all on a single tower, single fan $35 cooler.

File: allow me to interject.jpg (72KB, 960x727px) Image search: [Google]

72KB, 960x727px

>buy CPU
>OC it
>return it if the result is not good enough
This is why consumers have rights, to abuse them to get the perfect overclock. Don't tell me you're too beta to return a CPU.

File: [HorribleSubs] Kobayashi-san Chi no Maid Dragon - 10 [720p].mkv_snapshot_03.26_[2017.03.15_14.21.42].png (393KB, 462x497px) Image search: [Google]

393KB, 462x497px

>>61322594
>Cucked by restocking fee

>>61322415
Yeah but the i7-7700K requires a $300 AIO and even then it still gets pretty hot. People have been OC'ing the R7 1700 to 3.9 GHz on the stock cooler it comes with.

>>61322624
>$300
What? I have Ryzen and even I know that's BS. A $100 cooler sure ( still unacceptable) but don't go exaggerating.

>>61322584
Aren't you afraid of the PC not booting if you underclock too much? I know you can physically reset it on the motherboard but it seems like a hassle. How do you know how low to go?

File: leverpostej.jpg (47KB, 417x500px) Image search: [Google]

47KB, 417x500px

>>61322618
>Buying from a store with restocking fees
>Not having consumer rights in the first place
You must be a burger, freedom is great huh?

>>61322694
I meant undervolt not underclock btw

>>61322415
The issue with Ryzen is you likely can't even overclock to the XFR clock with an X CPU. A 1600X will likely only get to 4.0 despite having a 4.0 boost and 4.1 XFR. You get much higher multi performance but slightly less single, you shouldn't have to make a tradeoff like that when overclocking even though it's small.

>>61322766
You're right, 1600X is not really an overclocker CPU. It's pretty much there for people who don't want to bother with overclocking.

>>61322694
When I over/under clock, I stress the shit out of the component before deeming it stable. All I did was keep knocking the voltage down by 0.1 increments and stress test with prime95 until failure. Then when I feel I found my upper or lower limit, I stress test for 12+ hours

Make sure to have your load line calibration set to extreme or strict to ensure your board doesn't boost or droop even a tiny bit from the voltage you have set. Once you find your low and it passes a 12-16 hour test of prime95 blend, I consider it stable. You can then re-activate the AMD stepping and down clock at idle settings. At idle, the voltage drops as low as the CPU needs determined by the board. It just never exceeds the 1.025v ceiling I have set. (99% of the time the chip uses under 1.0v, but I've seen occasional bumps to 1.015v when under 100% load for a few weeks, so I set an extra 0.025v for safety.)

>>61322815
>just prime

>not prime + OCCT + IBT + a few hours of x264 encoding

I've seen plenty of times Prime going for a few days but one of the others crashing in a few hours.

>>61319365
Ryzen overclocks just fine, it just hits a wall at 4ghz for this first generation.

>>61322866
Well I've been encoding 1080p rips of the entirety of the Dexter series as well as Hannibal and Breaking Bad. All on super HQ present with a couple tweaks. CPU has been x264 encoding for 2 weeks now. All while playing GTA 5 at the same time. If there were problems, they'd have shown themselves by now.

>>61322511
I've yet to have an Intel chip not go at least 500mhz over stock. I don't know what you are going on about but most unlocked chips with both Intel and AMD prior to ryzen had a lot of oc room.

>>61323374
No they didn't. All intel and AMD do is bin chips to function at 1.4V in certain high clocks. They can never know which batch has chips that OC to say 5GHz at 1.3v so it's extremely hard to say how many chips can be OC'd at all because they never fucking test for that shit.

It's all a lottery like everyone says and you've just been lucky.

>>61319734
Been running the asrock b350m pro and it had 3200mhz memory since beta agesa 1.0.0.4 without issue, has the m.2 slot above the pcie slot as well.

>>61319575
Has at actually come to anything? Just looks like they're consolidating their product line by dropping the k4.

Best B350 for OCing? Asus?

>>61324648
MSI Carbon > ASUS Prime = Asrock AB350 (the white one)

>>61309352
Please. I cant't even get a stable OC of 3.65GHz

>>61324685
= all other MSIs

Any thoughts on the Gaming 3? Mine overheats @ 4.0 1.4v with Ryzen 1600. Crashes like once every two days.

About to pull the trigger on a W500 refurb:
1920x1080
i7-3720QM four coars 3.6GHz boost
1x8GB DDR3(should i fall for the 16GB meme?)
240GB SSD(adding a 500GB for extra hentai)
HD Graphics 4000+NVIDIA Quadro K2000M

What do you think /g/? Also do i need to ask them what WiFi card is in there, or face terrible peril when i install loonix?

>>61327641
Can't you replace the wifi card if it doesn't work?

>>61327653
You're right, it's in the express slot.

>>61327400
Stock cooler ? try lower the voltage or mhz

>>61327641
>16GB meme

Not a meme it's basically required for anything demanding

>>61316964
this. it's not as strong as say the crosshair or the tachi, but it's no way inefficient for ryzen overclocking. its more than ample enough.

i had the gaming 5 but did replace it with a crosshair. simply because of the bios.gigabyte bios is severely lacking to the point of being absolutely embarrassing. i'm not sure how it is now, but when i had it, it lacked such basic features as disabling onboard audio.

File: Screenshot_20170707-233809.png (1016KB, 2560x1440px) Image search: [Google]

1016KB, 2560x1440px

>MSI b350M Mortar Arctic
>R5 1600X
>Ram at 2933 so far.
>Can only hit 3.8ghz with 1.385 volts
What the fuck did I do wrong? Did I lose the lottery or is my mobo fucked?

File: elmo.png (107KB, 250x253px) Image search: [Google]

107KB, 250x253px

>>61330208
You lost

>>61330208
You lost, although I've heard rumors that X370 is better at overclocking than B350.

>>61330315
>>61330642
Alright gents, I guess this is my life now.

wtf I was planning on getting 1600x because it's supposed to be better binned

>>61330855
What's your RAM voltage?

>>61330876
1600X is better if you don't want to bother with OC. It's top end is basically 1600 when OCed. Not saying you can't OC it but silicon lottery applies no matter what.

>>61330208
check your ram sub timing, it sometime mess with core overclock

>>61330208
Try overclocking it without overclocking your RAM at the same time

I got 3693mhz across all 6 cores on the 1600x at work, fucking hits 81C according to Speccy and HW monitor while running an Aida64 load on a 120mm AIO. Ambient temperature of 34C Motherboard is a Gigabyte AB350M Gaming 3

Basically a 100mhz overclock.
Running 16*2 32GB @ 2400mhz as anything over 16gb basically shits the bed. Purely used as a video encoding rig.

>>61332504
Undervolt it, prob can get 1.28 or lower at that clock

>tfw 4025MHz @ 1.375V

>>61331029
>>61331010
Will do, gonna report back in.

>>61332504
20°C offset on X skus nigger, ryzen cpus straight up turn off above 75°C

File: 1600x.png (43KB, 1544x533px) Image search: [Google]

43KB, 1544x533px

>>61333128
They already addressed the issue with the new bios update for X series and I'm on AGESA 1.0.0.6

AB350M gaming 3 has settings going straight up to 90C on the F5 bios

Max temp for Ryzen is 95C it says it right on their site.

>>61333208
at which temps are you idling and what is the room temp?

>>61309939
>>61309917
>get's
What the fuck was that?

Y'know, get's ya everytim.

I got my 1600X to 3.9 stable and if I try 4.0 with a bit of an overvolt I can't even get into BIOS. Had to flip the jumpers to reset everything. I suppose 3.9 with 2933 RAM is acceptable for now.

>>61333128
It's always been 95C

>>61327641
>WiFi card
Why do people use WiFi cards? what are they for? whats wrong with modems?

>>61333267
high 40s or low 50s room temp is 31~34C

>>61309352
>1.35 on 1700x all the way to 3.9GHz, maybe more

Feels ight, temps start to get out of control (not too crazy though) above 3.9GHz even with the voltage staying at 1.35 - 1.373.

>>61311867
Consider running it at 3.8 - 3.85. Also, I hope you've got you memory freq /cl in order.

>>61312974
1700 is for cheap work stations / high end bestbuy PCs, 1800x is for mid level workstations, 1700x is for normal consumers with a reliable OC.

>>61322584
You should update your BIOS; 1201 adds some new profiles and fixes some important bugs

Ok so help me out /g/. I got my 1600X to boot at 40 multiplier and 1.3875V. Windows shows that my speed is at 4.0 but CPU-Z and HW monitor shows all cores are stuck at 2.2. When I dialed down the multiplier back to 39, CPU-Z and HW monitor shows the correct clock speed of 3.9. What is causing this random 2.2 lock?

>>61333851
same thing happen to me, had to overclock bclk to get 4ghz. think it just bios bug

>>61334116
I tested it more and it's only 40 multiplier that causes this problem. I think I have to play with base clock now. Hopefully this gets fixed soon.

File: overclock.jpg (80KB, 400x396px) Image search: [Google]

80KB, 400x396px

you could run it at 1.4v 24/7 just fine

>>61319785
The metal part is the same, the fan is different (no rgb) Don't know if fan vendor is the same.

>>61328830
>Stock cooler ?
He probably meant OEM cooler

>>61333829
New agesa 1.0.0.6 out on new 14xx bios for CH6

Sandy bridge Master-race. My i7-2700k was stock 3.5 but I clocked it to 4.8. 6 years old but still much faster than the latest generation.

>>61309917
I could get 4.8Ghz on my 3970X
If I ever upgrade I want to try out liquid nitrogen to see if I can hit 5.0Ghz+

>>61336669
Please keep reminding these plebs. They need to know!

>>61336850
Keep lying to your self about how great that POS cpu is

>>61309352
I need 1.319V for 3.8Ghz on my 1700, can boot 3.9Ghz with 1.45V but it isn't stable.

This coupled with some other issues makes me regret buying ayyymd. Ryzen was not ready for launch.

>>61337437
here's another reminder: stop using exclamation points and use CAPS instead. How new are you to the internet in general to not know this.

I get 3.8 GHz at 1.32 V and need like 1.385-1.4 V for 3.9 GHz on my R5 1600.


I just run it at 3.8 because the performance difference is minor but the voltage/temperature difference is massive.

>>61309352
I need 1.4 to run 5ghz.

>>61337551
OH NO LESS THAN 5% DIFFERENCE IN PERFORMANCE FROM 4 GHZ!!11!!!
HOW HORRIBLE

File: 1317464430091.jpg (12KB, 472x472px) Image search: [Google]

12KB, 472x472px

Why does /g/ shill so much for AMD?

The clockspeeds and voltages I'm seeing here are just horrendously bad

File: didiwinthelottery.png (26KB, 398x396px) Image search: [Google]

26KB, 398x396px

>>61309352
is this gud gyus

>>61338340
Not bad, wraith cooler ?

how your ram timing

>>61322415
Actualy no, the 7700k boots to 4.5 ghz on one core out of the box, I run a 6700k on 4.5 ghz on all cores.

With a stable oc at 1,27v 4.5 ghz I could probably push it to 4.7ghz if I wanted, but why bother, it stays nice and cool, power consumption is great and it's already powerful enough

File: ryzen win.png (1MB, 1680x1050px) Image search: [Google]

1MB, 1680x1050px

>>61338416
no, using a H11Oi Corsair AiO. Ram runs at 2999Mhz. cant get it to run at 3200mhz.

>>61338331
4.1 is the max that Ryzen gets to even with good silicon lottery.

>>61338535
Yes, and from what I've seeing not many people even reach that, and if they do they run them at high as fuck voltages

>>61338331
The clock and voltages may seem "bad" but it is punching higher at Intel products that run at higher clocks.

Zen+ / ZenX will knock Intel out. AMD is likely going to have 7nm before Intel in 2018/2019. Intel is just getting around to 10nm for launch next year while GlobalFoundries has a working 7nm FinFET node.

Zen is already a strong platform minus the clocks. Zen+ platform will improve ipc + clock and a 7nm node will increase performance an additional 40% or costs by 40%.

Picture a 8 core 16 thread Zen+ selling for $250. Intel's server market will collapse.

AMD is sprinting while Intel is now just getting off the line.

https://www.globalfoundries.com/news-events/press-releases/globalfoundries-track-deliver-leading-performance-7nm-finfet-technology

>>61338527
2999mhz with 1.0.0.6 BIOS ?

>>61338670
>The clock and voltages may seem "bad" but it is punching higher at Intel products that run at higher clocks.

It really isn't though, the only reason they beat some products of Intel in multithreaded applications is because they simply have more cores.

For people who focus purely on gaming though, Intel is still the go to brand, and anyone who suggests otherwise is only looking at cherrypicked results

>>61319367
30C idle, 75C load

>>61338671
yes

File: 1499876545757-g.png (334KB, 679x702px) Image search: [Google]

334KB, 679x702px

>>61338700
>>61338700
>they simply have more cores
>For gaming Intel is still the go to brand
>cherrypicked results

I'm talking profits and that is the high volume server cpu market. Intel is already losing contracts to EPYC in the market. What I showed was AMD moping the floor with any 10nm Intel CPU either in the price side or performance side by a 40% margin.

This is being posted by a 7700k owner.

>>61338834
I don't really care about any of that outside of the fact that I'm happy that AMD is gaining back some market share so we get back competition

>>61338551
But they're good at sub 4ghz so it doesn't really matter.

>>61338900
Good at what?

File: 1499704890153-b.gif (987KB, 229x176px) Image search: [Google]

987KB, 229x176px

>>61338846
>I don't really care about any of that
So you don't care about a potential 40% performance jump not counting the additional clock improvements and platform improvements?

Ok then.

>>61338914
I don't care about server cpu's no

>>61338932
What about 40% ipc for gaming performance?

You do understand what a 50% die shrink means right? Either 2x the transistor count (dependent on platform) or 2x the number of dies per wafer if yields match.

Ryzen has been blessed with exceptional die yields. If that can translate into the 7nm node expect all Intel CPUs to be gutted in 2018/2019/2020.

>>61338670
>>61338834
>>61339108

Someone archive this for the future when /g/ was right again.