So I just finished building my ryzen system yesterday and after running benchmarks i noticed my speeds were way too low CPU-Z (latest) CinebenchR15 and the AMD Ryzen Master are all reporting the same 4c8t.
Things I have tried:
Updating Bios
Disabling 2 cores (3+3) - wouldn't boot
Clearing CMOS
Booting into a fresh install of windows 10 (was using win7)
Downclocking and overclocking the CPU
Reseating the CPU
I'm out of ideas I can't find a single thread where anyone has had this same issue. Going to see if i can return it next week since it seems defective (Maybe AMD fucked up on the product line and gave it the wrong hardware ID?)
>>59972836
run cinebench and tell us how fast it goes
>>59972836
Arr Emm Aeh
Here is the AMD Ryzen Master screenshot
>>59972836
DELET DIS!!!!!!ยก!!
>>59972836
Boot to Linux and check there. Run top or system monitor to see if it's software or hardware problem.
>>59972849
Ran this earlier
>>59972870
why is your multiplier at x15?
did you update BIOS
run AIDA and check L3 if it works
it's pretty cool if one of CCX just DOA
Seems like BIOS/EFI disabled 4 cores.
Also those clocks are way too low
Just return it
>>59972912
also reset CMOS, remove battery and unplug PSU for a few minutes
>>59972889
It's definetely not just detecting it wrong, you can tell by the cache size
>>59972912
I have no idea why its at x15, i cleared the cmos like 10min ago Bios is updated (and had same issue before update) I'll run AIDA now give me a min.
Everything clearly reads 4/8
And you said 3+3 wouldn't boot. So I'm guessing just a batch that got a wrong ID
Op where did you buy your cpu?
>>59972971
This is what I'm thinking, seems logical but I don't have any way of testing this theory it does say 1700 on the chip itself.
>>59972990
A computer store in Australia M.S.Y (Branches all across aus)
Some machine probably fudged up validation and mistagged it.
Just return it, how is that hard?
>>59972836
>8 cores
>what do you need 6 cores for
>here's yours 4 cores
>>59972870
You have ryzen master set to disable 4 cores ....
>>59972961
>i cleared the cmos like 10min ago
with ryzen you have to remove battery for full clear
there is some bullshit there that prevents true clear sometimes
>>59973039
No I can't even uncheck them, i had this problem before i installed Ryzen master
>>59973056
uninstall it, go into your bios, reset defaults and then manually set the clock to 3200.
It's a problem with your mobo
>>>/g/sqt
>>59973006
It might be your motherboard .
I would contact the motherboard company what they think . It's highly unlikely that you have a faulty cpu , and that the motherboard is messing up somehow .
>>59973111
Already done all of that, tried ocing it as far as 3.8 and downclocking it as low as 2.8 (just using the multiplier)
>>59973105
It's likely reading the cache from the processor ID than a measuring API,
>>59973167
Well I don't use AIDA64 So if you want me to run something else I can, i have already spent a few hours trying to figure out what the problem is and getting nowhere.
Before I return it i want to know if its a fault with the motherboard or CPU I just have no idea..
>>59972836
Intel processors don't have this problem.
>>59973196
I'd put the blame on the motherboards, they have been less than stellar with AM4
Put it in someone else' motherboard to test you fag.
>>59973209
because they only have 4 cores to begin with unless you have a large bag of shekels
>>59973232
Easier said than done...
Going to have to bring my whole PC in because M.S.Y HATE giving out refunds and i don't want to return CPU if its a motherboard problem
>>59973279
I'm sorry your consumer right laws are so shit, I can return chips here enough time to piss everyone for the rest of week, LEGALLY at that, until I find a golden sample.
>>59973306
Abuse like you described is the largest reason we don't.
>>59973306
Lucky bastard, I have only ONCE returned something that wasn't faulty (gpu) had to say i was getting artifacts and after they ran tests for like 2 hours i had to talk them into giving me a refund.
>>59973352
>protect the corps instead of the people because there's 10 assholes who are just assholes
Australia all right
>>59973352
It depends on the store more than anything though the majority of massive retailers won't give you too much hassle but stores that have very small profit margins as they are very competitive do their best to avoid refunds.
>>59972836
HAHAHAHAHA, HE PAID 330 DOLLARS FOR A 4C/8T PROCESSOR?!
WHO WOULD DO THAT IN CURRENT_YEAR?
(i7 intensifies)
>>59972836
LIQUID COOL THIS!
>>59973006
What RAM did you buy btw? Looking at roughly the same build and I'm guessing any RAM with AMD in the name should be okay, yes?
>>59973209
Intel processors are a problem in themselves.