6-core Ryzen 5 1600X Benchmarks shows 50% faster than Core i5 7600K

Someting is wrong i get more multithreaded score @less clocks, check if you disabled CnQ, Turbo Boost, C1E and CPP on BIOS to get the real sobre, i got 7916@4.00GHz

 

I'll believe it when I see the reviews.

 

2.5x in mono and 3x in multi, sounds great but The Bench does'nt represent real world results, your CPU is getting low scores compared to FX CPUs but in real world your CPU has better results per clock/per Core

 

AMD FX 8350@4.8 1399/9503


Not bad at all

 

Right now I have only a couple profiles saved in the bios and both have the turbo enabled, one to 4.0 and one to 4.5GHz.
I don't have the time to try and modify stuff now, but this last one works with CPU LLC set to High, Cpu power response set to Fast and every other voice of the DIGI+ Power Control set to Auto or default values.
The offset overclock is set to + 0.05V.
C&Q, C1E, SVM and Core C6 State are enabled, the HPC mode is disabled and the APM master mode is set to Auto.
The higher cpu Vcore is 1.464V according to hardware monitor.
Probably disabling the turbo I would need to change something, but I did this in the past, both via bios and using Overdrive, and the cpu was doing fine in stress tests
The Asus M5A97 Evo Rev.2 is a great motherboard for overclock, but the bios it's a pain in the ass, because if I want the cpu frequency to scale down in idle I'm forced to overclock with the Offset mode. I freaking hate it. Overdrive does a better job, but in my experience isn't a very stable program.

 

My highly tuned 2600K does 1822 / 7614 with CPU-Z 1.78.

That would mean you need a SandyBridge running @5GHz with aggressive memory timings to catch-up with ST Ryzen at 3.7GHz. This is on par with latest Intel CPUs, which I find a bit unrealistic and I wouldn't make much assumptions based on just one (and note very good) benchmark...

On the other hand, if it is representative result, which will be confirmed by other benchmarks, I will have no other choice but to finally upgrade to 8/16 core system

 

And my weak cheap cpu for your reference

 

I really hope this is true.

 

My 2700K@4.9ghz gets 1850 single threaded, not bad for a 100 year old CPU.

 

Okay, good info, looks like I'm going AMD for the first time in a while!

Going for the 1700 variant though.

Very cool!

 

Looks like I'll have to upgrade ASAP :infinity:

 

4930k @ 4.2 with 1866 mem singlethr:1821 and for multithr:11029 windows 7 home pr 64b

 

im reaaally sceptical, its two different pc's with different os. no info about cpu on cpuz screen. sketchy af imo

 

moar benchs!
https://www.techpowerup.com/230736/amd-ryzen-1700x-1600x-1300-benchmarks-leaked

 

Maybe you have a Core disabling tech enabled like C1E or CP6

 

They definitely knew what they were doing back in 1917. :thumbup:

 

Actually, Ryzen cpu's look to have lower TDP than Intel ones. at least on 8 and 6 cores..

If you want to know the real performance by core of an architecture, you need to normalize it by mhz.. ( or instruction / cycles ).. power consumption have nothing to do with it. But if you want to know the performance / watts.. this time yes. ( and as i have said, i dont think the watt metric is in favor of Intel this time around ).

If i set my 4930K to 5.3ghz ( who is barely the wall for this one under H2o ). im pretty sure that thoses single thread performance will look way worst.... This dont mean that the IvyBridge architecture was vastly superior of Intel Skylake.

 

Yeah my uncore is at 4.4GHZ, Ram is 16GB 1866 DDR3, not overclocked as I don't like to risk it with RAM just not worth it for real world applications.

 

Great! I'd love to not default to Intel!

 

per watt means absolutely nothing. Yes, if you care about energy consumption, then it "sorta" means something, but it has absolutely nothing to do with performance.