• 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.

Core 9000 series

everytime intel bring out another faster cpu than there already amd beating chip it is funny watching people who just love amd saying well if they just do this or do that on the next new amd chip they will have intel beat. they been saying same thing for best part of 15 plus years now and going further downhill.it aint happening.
 
andy_mk3 said:
Finally an interesting post :D

That looks awesome! I really think the 9900k is finally going to be a worthy upgrade to my 2600k. I'm hoping a balls out upgrade will last me another 6+ years.
Click to expand...
I'm thinking the same, just with the 9700K - looking forward to seeing some proper reviews with concrete numbers.
 
SiDeards73 said:
Infact thats about the one thing AMD does better than Intel right now, AMD's virtual core if you want to call it that, is actually faster than Intels virtual core, i think i read somewhere that Intels is about 18-25% performance of a physical Core, where AMD's is about 35-50% or something.
Click to expand...
Very surprising given that Intel had, what, a 15 year head start on their SMT implementation? AMD somehow manages to outperform Intel's version in their very first attempt and also, as far as we know, do it in a more secure way considering OpenBSD is going as far as recommending disabling Intel's SMT altogether for security reasons.

I wonder if they just never touched HyperThreading at all since the Nehalem days, falsely assuming it was not something that could be noticeably improved? I mean up until Haswell they can be forgiven for focusing on single threaded performance since they actually got somewhere with that approach but beyond that it's surprising they never bothered tweaking it.
 
DragonQ said:
Very surprising given that Intel had, what, a 15 year head start on their SMT implementation? AMD somehow manages to outperform Intel's version in their very first attempt and also, as far as we know, do it in a more secure way considering OpenBSD is going as far as recommending disabling Intel's SMT altogether for security reasons.

I wonder if they just never touched HyperThreading at all since the Nehalem days, falsely assuming it was not something that could be noticeably improved? I mean up until Haswell they can be forgiven for focusing on single threaded performance since they actually got somewhere with that approach but beyond that it's surprising they never bothered tweaking it.
Click to expand...

Pretty sure AMD's SMT is based on IBM's SMT, would explain some of the reason if that is true. Done some digging around and Intels is supposedly *upto* 30%, but on average is around 5-30%. Still cant get a definitive figure on AMDs, seen reports it as high as 48% on some scenarios.
 
~>Dg<~ said:
everytime intel bring out another faster cpu than there already amd beating chip it is funny watching people who just love amd saying well if they just do this or do that on the next new amd chip they will have intel beat. they been saying same thing for best part of 15 plus years now and going further downhill.it aint happening.
Click to expand...

It's equally funny that you think I love AMD lol. I don't think there is another member here that has been so critical of ryzen.
 
Did a few people miss that the 9700K is 8 core 8 threads? It's 8 threads and scoring slightly higher or on par with the 2700X and higher than the 8700K (12 threads). The i9 9900K is the one with 16 threads.
Not sure how this ended up being another HT vs SMT discussion since the bench result is with a chip without HT.
 
AndreiD said:
Of course, but somehow people managed to make it a HT v SMT discussion despite the 9700K not having HT.
Click to expand...
Beren's statement on performance differences in single- and multi-threaded tests triggered it, I assume. Maybe SMT isn't the only driving factor to AMD's apparently superior multicore scaling then?
 
You must log in or register to reply here.
Back
Top Bottom