AMD-hired Agency in South Korea Teases AMD Ryzen 7 3700X, Ryzen 5 3600X

[Space Lynx]

Of course Intel is faster than AMD currently(10%ish IPC advantage, massive clockspeed advantage & latency advantage), but this article isn't about 2700x's, it's about 3700x's. If the rumors about clockspeed & IPC improvements are true, then intel's advantage in frame rate may disappear completely. Rumors are just rumors however, we'll see when they release the products next year.

As for the difference between Intel & Ryzen currently @1440p, it's little-to-nothing unless using a $1,200+ 2080 Ti, and occasionally the $700 2080(1080 Ti). Those GPU's are not main-stream products, and anything lesser than those @1440p get virtually identical FPS in most games, as the games become GPU bound.

you are right about 1440p on max fps. but not min fps, AMD still struggles there. i concede though, 10 fps when you are rocking 130 isn't a huge deal anyway. i hope 7nm amd cpu and gpu impress me. im still keeping an open mind about that being my next build hopefully spring/summer 2019... unless AMD decides to keep being slow as crap

 

[TheoneandonlyMrK]

I remember the times when amd was ahead of Intel.especially 1 core 1 ghz days. Make no mistake those days AMD itself was charging premium. Neither companies are innocent. Competition is good for us consumers as they have to improve and make prices competitive.

True , ,i don't know why you got -1'd.

But if there is good competition across the price points there is room for that top price part, Intel proved this adequately.
Both having good parts which Intel no doubt have should allow consumers to benefit most ,not many of us need that 1K sku.
But it's ok it exists, like multi thousand pound CPUs and GPU, so long as that price points in the extreme im ok with that, it's the mid range price hike that erks me.

 

[Super XP]

by that time AMD will be on Zen 3/4 with perfected Chiplet design and 5nm EUV

Intel is in serious problem, from what we know their 10nm uarch is still monolithic, and by the time intel launches first 10nm product AMD will move to Zen 3/2+ with 7nm+ EUV (second gen 7nm)

so if intel with late 2020 product can only beat AMD early 2019 product, its nothing to be happy about

Intel having serious problems is a great thing. AMD went through this for many years, and they came out of it with a very successful ZEN design. Intel Hired the legendary CPU Architect Jim Keller. This tells me they are desperate.
Hopefully AMD takes the lead for several years to come till Intel come out with a come back product.

 

[Totally]

True , ,i don't know why you got -1'd.

But if there is good competition across the price points there is room for that top price part, Intel proved this adequately.
Both having good parts which Intel no doubt have should allow consumers to benefit most ,not many of us need that 1K sku.
But it's ok it exists, like multi thousand pound CPUs and GPU, so long as that price points in the extreme im ok with that, it's the mid range price hike that erks me.

I disagreed because at AMD's height the fx-57 was $1050, the next chip down was the fx-55 at $850 while Intel's best was an Pentium 4 EE coming in at $999 that performed worse than either. Those are chips that aren't intended for the mainstream so I don't understand what crimes they're guilty of? Specifically AMD, what they didn't charge $1500 for the fx-57 like Intel would have had they been the ones in that position?

 

[m4dn355]

Intel having serious problems is a great thing. AMD went through this for many years, and they came out of it with a very successful ZEN design. Intel Hired the legendary CPU Architect Jim Keller. This tells me they are desperate.
Hopefully AMD takes the lead for several years to come till Intel come out with a come back product.

After that Amd again hires Keller to create zen successor and share Intel designs in order to beat them to the ground.

 

[Super XP]

After that Amd again hires Keller to create zen successor and share Intel designs in order to beat them to the ground.

Lol you never know.

But I have a suspicion that Jim Keller helped AMD even beyond ZEN, before he left. That's what AMD is setting up right now, for after ZENs generation.

 

[Vya Domus]

You people give way, way too much credit to just one person.

 

[m4dn355]

You people give way, way too much credit to just one person.

Let's say Lisa, Keller, 20+ engineers and marketing team made the impossible.

 

[Metroid]

12nm 2700X released on April 19, 2018 $329, 7nm 3700x will be released on January 20, 2019 $299. That is less than a year, which means that if you are not happy about it then there is something wrong with you.

 

[Space Lynx]

12nm 2700X released on April 19, 2018 $329, 7nm 3700x will be released on January 20, 2019 $299. That is less than a year, which means that if you are not happy about it then there is something wrong with you.

where did you get that date from? lol January 20th? is this official?

 

[Patriot]

Please stop posting a single unverified source and citing adored tv... if you actually watched his content, he says to take a grain of salt, this is from an anon email not his normal trusted sources.

 

[bajs11]

been using intel CPUs since I was a kid back in the 90s, the original first gen Pentium
but maybe its time to switch
getting tired of that intel inside sticker anyway

 

[kings]

It seems that the leaks are false!

South Korean Ryzen 3600X and 3700X contest is pulled after AMD is notified of its existence

Steve Walton of TechSpot and Hardware Unboxed reached out to AMD for comment and was told that the contest was not sanctioned by AMD and that it has now been pulled. He also made several remarks on the recent rumors from Reddit and AdoredTV, stating

The Ryzen rumours are fake, pretty certain at this point. I've got info off the record so I can't released [sic] it, but yeah people are going to be upset when they really shouldn't be... don't assume Ryzen will be anything like Rome.

 

[Patriot]

It seems that the leaks are false!

South Korean Ryzen 3600X and 3700X contest is pulled after AMD is notified of its existence

Yeah, expecting >8c on am4 is a recipe for disaster. 7nm gives 20% higher clocks on same power package, OR half power usage at same clocks. Not both, and not with doubling the cores.
Don't expect miracles and price cuts... They may be the underdogs but they are not going to get rid of the need for threadripper.

 

[Vya Domus]

They may be the underdogs but they are not going to get rid of the need for threadripper.

Even if they would double the core count they can do the same to the Threadripper CPUs as well. That's not really a valid reason.

 

[Patriot]

Even if they would double the core count they can do the same to the Threadripper CPUs as well. That's not really a valid reason.

clocks are more important than 16c on am4, they cant crank clocks and double the cores... especially when that niche is covered by threadripper.

 

[Vya Domus]

clocks are more important than 16c on am4, they cant crank clocks and double the cores... especially when that niche is covered by threadripper.

Again, that isn't that out of place. Managing higher turbo clocks isn't that big of a deal even with 16 cores. An 8 core Zen 2 CPU with let's say ~4.1 base clock and ~4.7 single core boost seems pretty likely, right ? Those numbers sound much bigger than really are, that's just a ~400 mhz bump over what we already have. Using two of those dies to get that 16 core Ryzen that still has higher clocks is a perfectly viable path, that was the whole point of this new chiplet design.

And the niche that Threadripper resides in can be expanded further by a new socket and 64 cores.

And by the way your claim that 7nm either brings 20% better clocks or half the power isn't right either, it's 20% better clocks and half the power (well, 40%) . And that's according to TSMC themselves. Another thing to realize is that who knows what sort of power saving design choices were made for Zen 2 and the move to that 14nm I/O die, who know how much power can be saved there as well. Saying that 16 cores on AM4 is a recipe for disaster has no basis within the context of a supposed radical redesign on new node.

 

[Patriot]

Again, that isn't that out of place. Managing higher turbo clocks isn't that big of a deal even with 16 cores. An 8 core Zen 2 CPU with let's say ~4.1 base clock and ~4.7 single core boost seems pretty likely, right ? Those numbers sound much bigger than really are, that's just a ~400 mhz bump over what we already have. Using two of those dies to get that 16 core Ryzen that still has higher clocks is a perfectly viable path, that was the whole point of this new chiplet design.

And the niche that Threadripper resides in can be expanded further by a new socket and 64 cores.

And by the way your claim that 7nm either brings 20% better clocks or half the power isn't right either, it's 20% better clocks and half the power (well, 40%) . And that's according to TSMC themselves. Another thing to realize is that who knows what sort of power saving design choices were made for Zen 2 and the move to that 14nm I/O die, who know how much power can be saved there as well. Saying that 16 cores on AM4 is a recipe for disaster has no basis within the context of a supposed radical redesign on new node.

Thats not what amd is claiming or what they have managed with the mi60

 

[Vya Domus]

Thats not what amd is claiming or what they have managed with the mi60

That's a different, existing product, ported on a new node.