• 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 ***

Associate
Joined
29 Jun 2016
Posts
528
why are clocks speeds on amd zen 2 so crap? if amd went from 14 nm to 7 nm in last decade why are cpu clock speeds not nearing 10gh by now?

I would say the clock speeds are excellent. AMD improved IPC enough that the clock speed doesn't need to be extreme - Look here for the efficiency disparity between Intel and AMD processors.

More importantly, the process has little to do with maximum clock speed, particularly in the comparison between 7nm and 12nm/14nm.
The uarch and density elected to be used by AMD is the dominant factor, and they have absolutely smashed it out of the park in terms of die size and performance. Also keep in mind these dies have multiple use cases from personal to enterprise processors, so they are never going to be clock speed optimised.
 

TrM

TrM

Associate
Joined
3 Jul 2019
Posts
744
I would say the clock speeds are excellent. AMD improved IPC enough that the clock speed doesn't need to be extreme - Look here for the efficiency disparity between Intel and AMD processors.

More importantly, the process has little to do with maximum clock speed, particularly in the comparison between 7nm and 12nm/14nm.
The uarch and density elected to be used by AMD is the dominant factor, and they have absolutely smashed it out of the park in terms of die size and performance. Also keep in mind these dies have multiple use cases from personal to enterprise processors, so they are never going to be clock speed optimised.

I think the die size actually works against amd here though it’s to small a surface Area for heat to transfer away from the die and not being center of cpu. I believe with socket orientation on the motherboards it’s at the bottom part on the cpu so some water blocks are not as effect as they could be.

I know amd did the io chiplet is older no and it much larger then the dies but these 3000 cpu temps doe suffer from the way they are made atm and I hope that amd change the design for future cpu’s When they die shrink again
What we have to remember is Intel's 14nm+++++++++++++++++++++++++++ is extremely mature.

Broadwell (i7 6800 / 6900 series) was Intel's first 14nm CPU, did they hit 5Ghz? 4.5Ghz at best and like a toaster oven.

7nm is brand new, give it time to mature.

That’s the one thing that will go against amd from there road map they want to jump on the latest nm and process though.

From there road map we see ryzen 4xxx being on 7nm uv process so that would be a maturing but then zen 3 proberly called ryzen 5xxx will be on 5nm and so on. I don’t think we will see maturing from amd like we have from intel.

Not saying that’s a bad thing but amd will allways be at someone’s else process and following what they manufacturer at the time . But Nvidia have shown with there gpu and other company’s have done it’s a very good strategy
 
Soldato
Joined
25 Sep 2009
Posts
9,616
Location
Billericay, UK
Intel 10nm CPUs cannot get over 4.5 either, because getting in smaller node doesn't mean better speeds.

There are limitations in relations to physics.
Higher clock speeds are generally harder to achieve now with the smaller process add to that AMD are clearly using the worse silicon for the consumer grade chips so no chance of one hitting 5ghz using normal coolers.
 
Associate
Joined
21 May 2018
Posts
123
Higher clock speeds are generally harder to achieve now with the smaller process add to that AMD are clearly using the worse silicon for the consumer grade chips so no chance of one hitting 5ghz using normal coolers.
I think intel process is far better than the one in TSMC, Intel 10nm would equal maybe 5nm of TSMC ?

All mobiles cpus when they shrank improved their speeds
 
Associate
Joined
21 Sep 2018
Posts
895
Higher clock speeds are generally harder to achieve now with the smaller process add to that AMD are clearly using the worse silicon for the consumer grade chips so no chance of one hitting 5ghz using normal coolers.

Even a 14nm is struggling to hit 5GHz with a normal cooler. A 360 AIO is not a normal cooler nor is a D15.
 
Soldato
Joined
22 Nov 2009
Posts
13,252
Location
Under the hot sun.
that is because of the crappy cheap soldering, people that did delid improved it a lot

Intel doesn't use solder but TIM on the 9900K and most of the X299 platform. (except the top chips).
And is irrelevant what Intel uses because at anything over 8 cores, delided or properly soldered (9000XE series) they hit 90C+ even with dedicated 480mm radiator.
 
Soldato
Joined
25 Sep 2009
Posts
9,616
Location
Billericay, UK
I think intel process is far better than the one in TSMC, Intel 10nm would equal maybe 5nm of TSMC ?

All mobiles cpus when they shrank improved their speeds
10nm is about the same as TSMC 7nm by all accounts. I don't know why you think Intel's is better though Intel can't get any decent consumer grade products from it besides a couple of under powered dual core chips with no onboard graphics, their miles behind TSMC at this time.
 
Associate
Joined
31 Dec 2008
Posts
2,260
Intel doesn't use solder but TIM on the 9900K and most of the X299 platform. (except the top chips).
And is irrelevant what Intel uses because at anything over 8 cores, delided or properly soldered (9000XE series) they hit 90C+ even with dedicated 480mm radiator.
9900K IS soldered....
 
Soldato
Joined
6 Feb 2019
Posts
17,465
https://www.chiphell.com/thread-2058557-1-1.html

Seems that a beta BIOS is now available for some MSI boards incorporating the new clock speed fixes, and the early reports seem to be that chips are now not only hitting their stated boost clock, but even boosting slightly higher (a couple of different 3800Xs tested there are both hitting 4.55GHz).

And how high does PBO overclocking go now?
 
Caporegime
Joined
17 Mar 2012
Posts
47,380
Location
ARC-L1, Stanton System
They did something all right... Seem to have introduced more weird clock/performance behaviour along the way.


I read in the Tom's Hardware article what they call thermal boost "issues", a strange and some what hyperbolic thing to say when they also explain this is due to 75 - 80C thermal throttling, this is normal, every CPU has a thermal cap at which they begin to throttle down frequency, these are not "issues" but normal operation of the CPU. This is why, as with any CPU you should make a good choice for your cooling solutions.
 
Back
Top Bottom