• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD Zen 2 (Ryzen 3000) - *** NO COMPETITOR HINTING ***

Lets not forget Adored TV said the chips would be 5ghz. And then when they were announced tried to backtrack and said "Yeah but I meant they would overclock to 5Ghz". Both total bust.
Trouble is we don't know what the leak source was. Could've been wishful thinking by a random employee, could've been that 5 GHz max boost was their original plan until that changed for some reason (poorer yields, poorer clockers overall, keeping more back for TR/EPYC), or it could've just been total BS to keep the hype train running.

Overall, the clock speed is, once again, the sour note on these chips. I think we all need to realise that it's very possible there won't ever be chips that clock as high as Coffee Lake Refresh, at least not in the silicon era. Nodes just don't normally get that "mature"; the only reason 14nm has limped on for so long is AMD's disappearance from the market for 6 years.

IPC improvements are in the long run much better than pushing clocks anyway. I think best case scenario with Zen 3 would be another 10-15% IPC increase and a ~200 MHz clock bump across the board. More realistically though, I'm expecting ~5% IPC and ~100 MHz.

128 megabits = 16 megabytes though.
Motherboard vendors tend to quote megabits...
People on this forum, and on the internet in general, are absolutely bloody awful at being specific with units. "MB" is not the same as "mb" or "Mb", and if you're a computing enthusiast you need to learn these things, IMO, to avoid getting yourself confused.

Basically the "128 Mb" BIOS chips are the small ones that might end up ditching features in exchange for supporting newer CPUs.
 
I think it is safe to ignore all benchmarks already created until we have a clear indication from all reviewers that their CPUs are functioning as they expected. Right now we know that not to be the cases.
I'm waiting patiently before placing any orders. I reckon late August is when I buy, though I may start sourcing components earlier since it will be a complete new build.
 
I have found several yourtubers benching stock ryzen 3600 non x vs stock 8700k(so 4,2ghz vs 4,3ghz) using b450 boards and it either goes toe to toe or beats the 8700k.. Now i'm sitting here, skeptical, cause i want this to be true, but it seems to bloody awesome to be true. I don't need more cores. 6c/12t is plenty for me until next year but i would like the IPC improvements as im currently bottlenecking a bit in 1080p.
 
The most important thing is to maintain 1:1 infinity fabric, 3733MHz will be unlikely, very few CPUs can achieve this, but 3600MHz is quite achievable and 3200MHz now easy, then just tighten timings best you can, we have found 3200 12-14-14-28-1T possible and 14-14-14-36-1T at 3600 using Samsung B-Die.

Hey @Gibbo

I have a set of these https://www.overclockers.co.uk/team...-channel-kit-black-grey-tdged4-my-089-tg.html already on my x470 board, have another set coming today with my 3900x. Do you happen to know what dies are on these? Reckon I 'll see more than the advertised 3000mhz with some tweaking?
 
My fairly early stepping doing rounds at 4K on the Strix. Plug and Play.
unknown.png
 
In contrast to AMD's own presentation where they 'showed' a 69% difference between PCI-E 3 and 4 this article shows there is virtuly no difference at all (as we already knew)
There’s a big difference in some content creation.

In Eposvox’s testing Navi was beating the 2080ti when both were using Ryzen 3000 with pcie 4 for rendering.

Time stamp 8:47

https://youtu.be/JfEjU8Zx9Ps
 
Back
Top Bottom