We're going to need some pics of that if you don't mind. HWInfo logs are even better.
to give an idea of performance/OCs
my 1800X does 4.2 @ 1.45ish all cores
4.3 or so on 6c12t
4.4 on 4c8t
whereas the 1700/1700X seem to be only able to do max OCs of the same or worse, but with more volts
semi interesting:
1600X doesnt overclock as well as my 1800X does, the 1600X tops around 3.8 perhaps 3.9 on all cores on the same cooling that the 1800X can do a 4ghz daily on
1500X is an odd ball, and I haven't messed to much with overclocking with it yet
1 Like
As far as I know both the X370 Killer SLi/ac and X370 Gaming K4 support 2933Mhz at the very least.
http://www.asrock.com/mb/AMD/X370%20Taichi/
http://www.asrock.com/mb/AMD/Fatal1ty%20X370%20Gaming%20K4/index.asp
That is horrible concerning the 1600X.
I hope this is just a case of Silicon Lottery, and not some AMD Nerf.
the chips are binned differently, you wouldn't expect a $250 CPU to perform as well as the $400 and $500 ones would you lol, the more expensive ones are binned a lot better
the ones that failed to be binned as the 8core good OC chips, became the 6/4 core chips as far as i can tell
except the 1500x that one is odd
i have to delve into OCing it more
in my opinion: i wouldn't bother with bigger oc's than 3.9 anyway. The performance gains are minimal anyway, as stated above you'll get more performance for higher frequencies on memory. Wendell did a nice video
I saw some notation where 3000 will be reduced to 2933, but another spec that lists what I assume are profiles for 2667.
I would certainly expect a chip that lists a stock boost of 4000Mhz to OC to that level, and the pricing is for 2 less cores, no?
the boost is on 1 core 4 ghz, that is a whole different ball game than all cores at 4 ghz
good point, but usually when you are being served same cpu but with less cores they were likely disabled due to poorer/below standard of those 8 core binning
Current Draw on Ryzen can get really high with higher voltages with an all core OC. It isn't stable with an all core 4GHz OC. Maybe if you freeze it with Dry Ice/ Ln2 to reduce the internal resistance but not at the binning that they are in.
I'm hoping that as an early adopter I'll get an 1800X that had 2 perfectly good cores disabled just to fulfill inventory.
Then I will rest easy.
and I quote: "hope is mother of stupid."
just get 1600 or 1700
I thought that if you employed the PState method you would alleviate much of what you describe?
Why would I abandon hope, and why would I not spend $30 more for 1600X to get a better guaranteed Boost Clock and higher XFR to play with?
because 30$ is like 2-3 days of lunch money.
I'll eat Clif Bars, give me my X!
That's not how P-States work. P-states do not control individual cores. they control the scaling governor of the CPU. Which in turn autonomously scales each CPU core individually according to that profile. Now when you boot though, there are points during which the CPU is in max performance mode and if all cores clocks aren't stable it will still crash.
Same thing happens at POST, the system tests cores at max perf and if any of them fail. No POST.
1 Like
Alright, you guys got me excited now!
Fedex tracking says my new Mainboard is due by the end of the day!
1 Like