stimpy88This one really needs to be more than just a 5-7% IPC improvement.

It really doesn't. The 7nm process will allow quite some frequency improvement. Both more cores at higher frequency, and turbo boosts close to 5 GHz. We can expect 10%+ improvement in frequency as well. Then there's the probability of AMD adding even more cores to their top tier CPUs. So all in all we'rw talking 15% better performance or more as well as possibly more cores.

evernessinceIns't that what the 2800X is going to do? I don't think AMD even needs to wait for next gen ryzen for that goal. .

No, it isn't. We don't even know if the 2800X even exists. And if it does, you're naive to think it will be magically better than the current 2700X. If anything, it'll do 4.4 GHz with stable voltages at best.

evernessinceI almost spat coffee all over my keyboard, so thanks for that! :laugh:


As a first-gen Ryzen owner, I'll be skipping the current Ryzen 2XXX series, but I'll be all over this upcoming Zen 2 bunch. I expect around 5-10% IPC gain over Ryzen 2XXX, all-core boost clocks at 4.2-4.3GHz and boost clocks at 5GHz.

It seems AMD needs another breakthrough to truly break that glass ceiling on IPC. Mind you, they're not that far behind Intel and 7nm should help close the gap even further but I'm talking about at least a parity match and better.

That's not entirely true. AMD is only around 4-5% behind Intel in IPC with Ryzen 2. They only need 5% IPC improvement to reach parity/slightly surpass them. I am however pretty sure that AMD will aim to surpass Intel in IPC, as Intel will already have 8 cores Coffee Lake CPUs with high frequency ready this autumn; AMD has to be as good or better in clockspeed, as well as providing better IPC, in order to provide a CPU that's competitive with Intel's 8 core Coffee Lake. A goal that's entirely realistic for AMD to achieve.

Intel's superiority, however, is stiill in better cache latency, which means better game performance.

La MentheThat's not entirely true. AMD is only around 4-5% behind Intel in IPC with Ryzen 2. They only need 5% IPC improvement to reach parity/slightly surpass them. Intel's superiority, however, is still in frequency and better cache latency (which means better game performance).

images.anandtech.com/doci/12625/Cache Latency vs Intel_575px.png

No.

La MentheIt really doesn't. The 7nm process will allow quite some frequency improvement. Both more cores at higher frequency, and turbo boosts close to 5 GHz. We can expect 10%+ improvement in frequency as well. Then there's the probability of AMD adding even more cores to their top tier CPUs. So all in all we'rw talking 15% better performance or more as well as possibly more cores.

It's probably very safe to say that the GF 7nm process will be nowhere as good as Intel's 10nm, and I will put my life on the fact that we will not be seeing an 8+core Ryzen2 running air cooled at 5GHz anytime soon. For the next few years IPC is king for AMD, as Intel will match them on core count soon, and probably still better them on clock speeds. So AMD does need to aim and reach higher than the current Intel architecture, and settling for minor increments that result in deliberately staying below the IPC of what the current Intel architecture is producing is not what a company that very nearly went bankrupt a couple of years ago should be doing. This will only lead to a repeat of what AMD were doing when Intel released the Core2 architecture, which nearly destroyed AMD, and has taken them 10 years to counter. If they "settle" again, then AMD will not survive another 10 years of being significantly behind Intel. This time round, AMD no longer have a strong GPU division to back their financials up and provide turnover, unlike before as once again, after miss managing, they have lost to nVidia, and show no signs of being competent or able enough to do anything about it, AMD cannot afford to repeat past mistakes, it will end them.

La MentheThat's not entirely true. AMD is only around 4-5% behind Intel in IPC with Ryzen 2. They only need 5% IPC improvement to reach parity/slightly surpass them. I am however pretty sure that AMD will aim to surpass Intel in IPC, as Intel will already have 8 cores Coffee Lake CPUs with high frequency ready this autumn; AMD has to be as good or better in clockspeed, as well as providing better IPC, in order to provide a CPU that's competitive with Intel's 8 core Coffee Lake. A goal that's entirely realistic for AMD to achieve.

This is critical for AMD, and just think about not taking into account Intel's next architecture! Intel are starting to wake up to AMD and the threat they once again pose. However they could crush AMD easily, and only by AMD aggressively pushing their architecture forward will AMD win over Intel.

stimpy88It's probably very safe to say that the GF 7nm process is not as good as Intel's 10nm, and I will put my life on the fact that we will not be seeing an 8+core Ryzen2 running air cooled at 5GHz anytime soon. For the next few years IPC is king for AMD, as Intel will match them on core count soon, and probably still better them on clock speeds.

GloFos' 7nm, from the specifications I have read, is in fact about as good as Intel's 10nm. That being said, Intel is said to start with 10nm+ , which is most likely better (but still much closer to GloFo's 7nm than what Intel's 14nm+ was to GloFo's 14nm, when Ryzen launched in 2017). Whether GloFo's 7nm is as good as Intel's 10nm is irrelevant; you need to compare it GloFo's 12nm or 14nm process, which the last two Ryzen processors were one. Do you honestly believe their 7nm will provide less than 10% possible frequency increase over their 12nm process?

The upcoming 7LP 7nm FinFET process will, according to GloFo, provide the possibility to achieve around 30% performance improvement over 14nm, or 20-30% over the current 12nm process of the Ryzen 2 (which I believe was a quick and dirty transition from AMD, as they have the majority of their focus on Ryzen 3 / Zen 2). Traditionally, the actual clock speed improvements are half of those stated to be possible (because of various factors, like architecture limits), so my I guess is at least a 10% frequency improvement here -- around 4.8 GHz turbo on some cores, and all cores at somewhere around 4.3 GHz is very, very plausible.

stimpy88For the next few years IPC is king for AMD, as Intel will match them on core count soon, and probably still better them on clock speeds. So AMD does need to aim and reach higher than the current Intel architecture, a

This is true, and AMD will already at least match Intel in IPC with Ryzen 3 next year; that fact is pretty sure. But it's certainly not the 15% IPC you claim....

How much more improvement they will do in IPC further down the road remains to be seen, although I doubt it'll be as good as Zen 2. There's supposedly going to be two major revisions to the architecture, with Zen 2 next year being the first one. The second one is Zen 5 (Zen 4 name is being skipped) in 2021. AMD are said to have a whole new architecture ready after that as well, from what I've heard.

In the question of clock speed, I think AMD have a greater base from which to start from than Intel. Frequency increases have their limits, and Intel can increase the frequency only so much, before they need to stop. AMD, on the other hand, still have pretty conservative clockspeeds in comparison, and have greater potential of performance increases as we move along. A potential AMD needs to take advantage of: single-digit IPC superiority is not enough to make up for their clock speed deficit, if they want to perform better than Intel Core.

Unfortunately I can't find the pic anymore, but there was a comparison between all 3 process nodes, Intel 10nm and 7nm from GloFo and TSMC, and Intel's was inferior in every aspect they showed.
I think the major point of comparison was SRAM cell area or smth like that.

ShurikNUnfortunately I can't find the pic anymore, but there was a comparison between all 3 process nodes, Intel 10nm and 7nm from GloFo and TSMC, and Intel's was inferior in every aspect they showed.
I think the major point of comparison was SRAM cell area or smth like that.

laszloActually Intel 10 nm is almost same as Amd (GF) 7 nm....

but sound better 7 than 10 no?

Intel doesn't look bad to me... Slightly (in amounts that make almost no difference) worse in areas that don't really matter, and better in some others... Intel just need to fix the process first.

evernessinceas the 5775c has better gaming performance than the 7700K and 8700K.

7700K - yes, in most cases 5775c manages to outperform it by a whisker. When comparing oc vs oc it actually manages to squeeze a couple percent more than overclocking does to 7700K. 5775c only hits 4.3GHz in most cases, but that's 20% over stock 3600MHz. 7700K oc'd by 20% has to hit 5000MHz. Plus you can oc edram on 5775c, that's ~3% gained.Not much, but it all adds up,.

But compared to 8700K - no way 5775c even matches it.

ShurikNUnfortunately I can't find the pic anymore, but there was a comparison between all 3 process nodes, Intel 10nm and 7nm from GloFo and TSMC, and Intel's was inferior in every aspect they showed.
I think the major point of comparison was SRAM cell area or smth like that.

The comparison showed a very, very slight inferiority. Either way, Intel will start with 10nm+, as their 10nm is reported to provide performance deficit (that is, lower clock speeds) over their current 14nm++, and even upcoming 14nm+++. I'd be surprised if it wasn't better than GloFo's 7LP 7nm.

There's also still doubts about GloFo being able to provide 7nm early enough for AMD -- unless of course AMD are planning to release Zen 2 in the summer of 2019, or maybe even later.

cucker tarlsonThis is the one we've been waiting for. Ryzen 1 was just a prelude.

Eh, you can literally say this for every generation to the next. Zen1 did well and I'm sure Zen+ will do well.

Waiting for Zen2 is quite a ways out. If someone needs a new build now, they should just build now.

AMD doesn't want to pull ahead of Intel, that would cost them their underdog status. I can't imagine Intel letting that happen anyway. By the time AMD reaches 5GHz boost, Intel will be at 5.5 or more. So it's better for them to keep in position, right behind.

FxEh, you can literally say this for every generation to the next. Zen1 did well and I'm sure Zen+ will do well.

Waiting for Zen2 is quite a ways out. If someone needs a new build now, they should just build now.

Im waiting on TR 2 3950.

eidairaman1Im waiting on TR 2 3950.

It sounds like you have a heavy focus on workstation tasks and gaming performance comes second.

FxIt sounds like you have a heavy focus on workstation tasks and gaming performance comes second.

I just build them to last a very long time. I was on an Athlon XP before my FX, my FX has a new home to replace a Friends P4 rig.

eidairaman1I just build them to last a very long time. I was on an Athlon XP before my FX, my FX has a new home to replace a Friends P4 rig.

Nice. Then that makes two of us getting ready to upgrade. I am going to re-purpose my FX rig for my wife's telecommute/youtube needs. I'm gonna pull the trigger with Zen+. With AMD sticking to AM4 for quite a while, I can always throw in a Zen2 if I become tempted beyond self-control :D

FxNice. Then that makes two of us getting ready to upgrade. I am going to re-purpose my FX rig for my wife's telecommute/youtube needs. I'm gonna pull the trigger with Zen+. With AMD sticking to AM4 for quite a while, I can always throw in a Zen2 if I become tempted beyond self-control :D

Im not upgrading till next year though lol.

Fot me to know this chip hits 5.0 air tells me it has plenty of life left, it will be put back to 4.2GHz till 5.0 is truly needed lol

This is starting to become very fun to watch - AMD is getting so damn aggressive about new product launches, and Ryzen 3 could bring a similar level of performance uplift to what Ryzen 1 brought.

If 7nm Ryzen comes out before Intel gets (real) 10nm/8-core products out... They will curbstomp Intel harder than possibly even SandyBridge crushed AMD.

The Quim ReaperJust give us overclocks in the 4.5-5Ghz range and you'll pull sales away from Intel quicker than Donald Trump burns through brain cells.

I don't think you have to worry about losing any brain cells because you don't seem to have any with comparisons like that.

Anyway, unnecessary political comments aside, AMD is showing it to Intel in a big way. I'm really loving me some competition, finally.

btarunrpresents the company with opportunities to improve several silicon-level specifications, such as the number of cores per CCX, the IPC of each core, the core-count of the die, the cache hierarchy, and the overall energy-efficiency

Could we be seeing 10 to 12 cores with Ryzen 2!?

thebluebumblebeeCould we be seeing 10 to 12 cores with Ryzen 2!?

It is quite possible we could have 12 or 16 (I would guess 12 is more likely).

Actually one of the latest rumors is that AMD will separate the controllers, I/O, and core's onto their own separate dies; and then they will also make 3 and 4 core CCX's. The idea is that they will then have a 12nm I/O in the center, and then add 7nm 3 or 4-core CCX's as they wish around the other parts of the CPU.

TLDR - AMD is likely to have more than 8 cores with Ryzen 3, and they will likely make it even cheaper to produce than before by splitting up the CPU's into smaller components.

thebluebumblebeeCould we be seeing 10 to 12 cores with Ryzen 2!?

While I wouldn't be opposed to that, I hope that they focus on IPC above all.

FxWhile I wouldn't be opposed to that, I hope that they focus on IPC above all.

Consider a 7nm die is around half as big as a 14nm die. Thus they could double the cores and still "focus on IPC."

The most prevalent rumor is that the top desktop Ryzen 3 will have 12-cores... and that would mean they did mostly focus on IPC. ;)

Captain_TomThis is starting to become very fun to watch - AMD is getting so damn aggressive about new product launches, and Ryzen 3 could bring a similar level of performance uplift to what Ryzen 1 brought.

If 7nm Ryzen comes out before Intel gets (real) 10nm/8-core products out... They will curbstomp Intel harder than possibly even SandyBridge crushed AMD.

Here we go again with this nonsense....You truly believe Ryzen 3 will bring as big of an upgrade as Ryzen from Excvator (55% performance uplift)? What planet do you come from? Your claims bear no realistic probability; it's not even close! You are claiming an improvement on an already existing architecture is supposed to bring larger performance improvements than Ryzen did from a 6-year-old architecture (which was already inferior for its time).

You claim that Ryzen will be "so far in front of any Intel offering it will be a painful mirroring of the xxxdozer vs Core days". Even very conservatively, we're talking about another 30%+ increase in IPC here, if it's gonna surpass Core in any way to actually make up your description. Meaning AMD will increase IPC by 30-40% and clock speed by 15% (again, both numbers are conservative).

So this slide by AMD



Is, clearly bollocks, huh? It's pretty evident that you, Captain_Tom, know better than AMD themselves, and AMD's graph showing around 15% overall performance uplift by Zen 2 is massive standing. Instead, the increase will be, as you said, like this:



Yeah...no.

It's people like you who are responsible for creating these exaggerated and stupendous hypes about AMD products, that always end up completely failing.

bug^^^ Even 22nm has raised enough problems for TSMC they had to eventually scrape the whole node and go from 28 to 14nm.
The bad news is, while silicon is hitting its limits and we come up against challenges we foresaw, we still have nothing to turn to once silicon is no longer viable.

There may be nothing in production at this time, but I would argue that the future is, potentially, very bright for electronics. There are a number of exciting research projects in the works of which a sampling is here. I even read somewhere that a research team managed to construct a transistor made of three atoms; unfortunately, I cannot find the reference ATM.

Captain_TomIt is quite possible we could have 12 or 16 (I would guess 12 is more likely).

Actually one of the latest rumors is that AMD will separate the controllers, I/O, and core's onto their own separate dies; and then they will also make 3 and 4 core CCX's. The idea is that they will then have a 12nm I/O in the center, and then add 7nm 3 or 4-core CCX's as they wish around the other parts of the CPU.

TLDR - AMD is likely to have more than 8 cores with Ryzen 3, and they will likely make it even cheaper to produce than before by splitting up the CPU's into smaller components.

One of the things that AMD beat Intel at years ago was putting memory controllers on die. It would be interesting to see AMD go the separate route, again, and how that would affect overall system performance.

La MentheHere we go again with this nonsense....You truly believe Ryzen 3 will bring as big of an upgrade as Ryzen from Excvator (55% performance uplift)? What planet do you come from? Your claims bear no realistic probability; it's not even close! You are claiming an improvement on an already existing architecture is supposed to bring larger performance improvements than Ryzen did from a 6-year-old architecture (which was already inferior for its time).

You claim that Ryzen will be "so far in front of any Intel offering it will be a painful mirroring of the xxxdozer vs Core days". Even very conservatively, we're talking about another 30%+ increase in IPC here, if it's gonna surpass Core in any way to actually make up your description. Meaning AMD will increase IPC by 30-40% and clock speed by 15% (again, both numbers are conservative).

So this slide by AMD



Is, clearly bollocks, huh? It's pretty evident that you, Captain_Tom, know better than AMD themselves, and AMD's graph showing around 15% overall performance uplift by Zen 2 is massive standing. Instead, the increase will be, as you said, like this:



Yeah...no.

1) So your response is to post an outdated roadmap with no specific markers on the Y-axis? LOL ok, great rebuttal! /s

2) Is English your first language? You seem to not understand what the word "could" means. I said Ryzen 3 could be as big as an uplift as Ryzen 1 was vs Excavator (~50% boost). I am only basing this "could" on recent rumors AND GloFo's own performance estimates:

www.extremetech.com/computing/263286-sitting-globalfoundries-talk-7nm-euv

^^^ The current estimate is that 7nm products can use 60% less power than 14nm for the same performance, or it can also offer a 40% boost at the same power consumption. There is also up to a massive 45% reduction in die size (almost half the size!). So even if we were to be insanely conservative and assume the end result is half as good as expected, we would get enough room to add 2-6 more cores and increase single-threaded performance by 40%.


3) So then let's get this straight - you think adding 50% more cores, and a massive IPC boost will not equal a 40-55% overall performance uplift. Well then I guess we can agree to disagree!


TLDR: It is not insane to think the R7 3800X could be a 12-core with 20% higher IPC and 4.8GHz clocks. That would destroy Intel's current plan of just slapping 2 more cores on Skylake again.

Captain_TomIt is quite possible we could have 12 or 16 (I would guess 12 is more likely).

Actually one of the latest rumors is that AMD will separate the controllers, I/O, and core's onto their own separate dies; and then they will also make 3 and 4 core CCX's. The idea is that they will then have a 12nm I/O in the center, and then add 7nm 3 or 4-core CCX's as they wish around the other parts of the CPU.

TLDR - AMD is likely to have more than 8 cores with Ryzen 3, and they will likely make it even cheaper to produce than before by splitting up the CPU's into smaller components.

I've seen something similar in one of AdoredTV's videos (16 minute mark, for context):



Supposedly, the middle chip will be everything except the cores and the other 4 chips will be only the cores. Obviously, there should be infinity fabric in each chip to connect them.

Whether or not this is true or complete bollocks, i don't know ... but it is plausible. This should not increase core count, unless they also have CCXs in the corners, which isn't beyond the realm of possibilities: it could effectively double the core count.