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

[Murphy]

CPU A cycles 50 instruction per clock and is clocked at 1Ghz while CPU B cycles 25 instructions per clock but is clocked at 1.5Ghz.

Which CPU is the higher performance part?

CPU A?
CPU B?

I thought so, you don't understand what a cycle is do you? If you did you wouldn't have just said CPU A clocks (aka:cycles) 50 instruction per clock and is clocked at 1Ghz.

I did say cycles are also know as clock-speed but it seems you missed that part, obviously in your example it would be CPU A but that would be missing the point as you wouldn't say a Xeon Phi is suitable for playing games on but it has roughly the same IPC as Ryzen so saying Clock speed is irrelevant as IPC is all that matters is perhaps a little misleading, like i said the two are intrinsically linked.

 

[KungFuSpaghetti]

Here is how i look at it.

I'm not expecting any Ryzen 3000 to beat the 9900K by any significant margin in gaming, if at all in fact, what i am expecting to see is by any reasonable measure gaming performance parity, given there is only a 15% difference between the 9900K and the 2700X, its very achievable, by the looks of it they will achieve that, at least.

That's good, AMD don't have to beat Intel in gaming performance metrics, if they can match Intel then Intel can no longer boast about having "the fastest gaming CPU" what's more AMD are likely to be offering "9900K equality performance" for a whole lot less making the Intel part utterly irrelevant.

I'm not expecting the ES model final production version to be faster than a 9900k in gaming, because realistically, it wont be, especially at 65w. The 105w or whatever it is I would expect to be though based on the CB results and the 12 and 16 core SKU's that bit quicker again.

 

[Perfect_Chaos]

The fastest chips should be at least a little quicker than Intel in gaming if aiming for icelake though shouldn't they? Or one would think.

 

[nashathedog]

The fastest chips should be at least a little quicker than Intel in gaming if aiming for icelake though shouldn't they? Or one would think.

Personally I think we'll get more of the same with Intel's top gaming cpu's remaining ahead, If we can get 9900k gaming performance with the next gen Ryzen desktop flagship we're doing well, People need to remember that Intel's underlying cpu architecture is a lot more mature than Zens, Intel's had the time to make it shine & even though AMD's polishing there's as we speak it's going to take time.

 

[humbug]

I thought so, you don't understand what a cycle is do you? If you did you wouldn't have just said CPU A clocks (aka:cycles) 50 instruction per clock and is clocked at 1Ghz.

I did say cycles are also know as clock-speed but it seems you missed that part, obviously in your example it would be CPU A but that would be missing the point as you wouldn't say a Xeon Phi is suitable for playing games on but it has roughly the same IPC as Ryzen so saying Clock speed is irrelevant as IPC is all that matters is perhaps a little misleading, like i said the two are intrinsically linked.

I don't know what IPC Xeon Phi has compared with Ryzen and frankly your detractive curve ball is nonsense.

it would be CPU A

Well done.
Yes that is exactly right because it is doing more work for every clock cycle, if you agree with that then you must also agree that the clock speed alone is not relevant, if Ryzen 3000 is 100% the gaming performance of the 9900K then it is that, what clock speed that is achieved at is irrelevant.

 

[CuriousTomCat]

I've zero doubts that had that test instead been gaming benchmarks, then the results would have certainly favoured Intel.

Cinebench is a very good indicator of IPC. I think everyone agrees that the Ryzen 3000 CPU on demo during CES was clocked lower than the i9900K. Therefore, the Ryzen 3000 has a higher IPC than Intel and that's a fact. An absolute fact.

However, there is no guarentee that this will translate to games. If your favourite games have been optimised and written for Intel, then Intel will probably have the highest framerate in those games so Intel will be the best gaming CPU for you. Games aren't always an indication of which CPU is faster. I'm certain that Ryzens IPC gain will close this gap in games and will equal Intel at the very least.

If it's 0% IPC improvements that's not great, as the 5.0/5.1ghz SKU's will be barely any quicker for gaming

If it's 0% IPC improvement, then a 5.0GHz CPU does 25% more calculations per second than a 4.0GHz CPU. That's a big improvement.

I thought so, you don't understand what a cycle is do you? If you did you wouldn't have just said CPU A clocks (aka:cycles) 50 instruction per clock and is clocked at 1Ghz.

I did say cycles are also know as clock-speed but it seems you missed that part, obviously in your example it would be CPU A but that would be missing the point as you wouldn't say a Xeon Phi is suitable for playing games on but it has roughly the same IPC as Ryzen so saying Clock speed is irrelevant as IPC is all that matters is perhaps a little misleading, like i said the two are intrinsically linked.

I think that Humbug's example proves he knows exactly what a clock cycle is and it's correlation with IPC. Your answer is correct and I'm sure Humbug knows it.

 

[humbug]

I think that Humbug's example proves he knows exactly what a clock cycle is and it's correlation with IPC. Your answer is correct and I'm sure Humbug knows it.

Somehow he agrees with me but still has a problem with it, go back over the last two pages and people are arguing that Ryzen 3000 must be high clock speed, seemingly higher than the 9900K, my argument to that is basically this, that is not necessary if the IPC is the same in gaming or higher and provided plenty of examples to show that the 900K's gaming performance advantage comes purely from its higher clock speed, to the tune of 17% higher clocks that make up around 15% higher performance, the engineering sample is 15% faster than the 2700X, how it achives that seems to be the contention, somehow its good if its IPC and not good if its clocks speed with another arguing the opposite.
It seems to me looking for a negative in that which simply does not exist, the performance is there, that is all that matters.

Its the same performance "YEAH BUT BUT BUT CLOCK SPEED"

 

[Murphy]

I don't know what IPC Xeon Phi has compared with Ryzen and frankly your detractive curve ball is nonsense.

If you understood IPC you would.

Yes that is exactly right because it is doing more work for every clock cycle, if you agree with that then you must also agree that the clock speed alone is not relevant, if Ryzen 3000 is 100% the gaming performance of the 9900K then it is that, what clock speed that is achieved at is irrelevant.

Yea, unfortunately that's not what you initially said, need i remind you that you said "Clock speed is irrelevant ^^^^^ the FX-9590 was an all core 5Ghz part, it was still crap, IPC is all that matters."

It was only later that you added the 'only' caveat.

 

[humbug]

If you understood IPC you would.

Yea, unfortunately that's not what you initially said, need i remind you that you said "Clock speed is irrelevant ^^^^^ the FX-9590 was an all core 5Ghz part, it was still crap, IPC is all that matters."

It was only later that you added the 'only' caveat.

I'll rephrase that then, Clock speed on it own is irrelevant, again the FX-9590 was a 5Ghz part and yet its gaming performance was junk.

This is the point, clock for clock Zen+ is = to Coffeelake in gaming, it is 15% slower in games because its clocked that much lower.

The Ryzen 3000 engineering sample was measured at 15% higher performance, what clock speed that was is irrelevant, it could have been 1.2Ghz for all anyone should care because be that as it may its still 15% faster than its predecessor.

Don't get so hung up on clock speed, a CPU with 100% performance vs 100% performance of the other CPU is 100% performance vs 100% performance if one is running at 1.2Ghz and the other at 8.5Ghz they still have the same performance.

He misunderstood what i was saying, as did you, so i spelt it out better.

The Ryzen 3000 clock speed is irrelevant, the only thing that's relevant is its performance, nothing more.

 

[Murphy]

He misunderstood what i was saying, as did you, so i spelt it out better.

The Ryzen 3000 clock speed is irrelevant, the only thing that's relevant is its performance, nothing more.

If two people misunderstood what you were saying that probably says more about what you said than how it was understood by others, granted you added caveats and expanded on what you initially said but just because you've done that it doesn't make what you initially said anymore correct, perhaps it's best if we just accept that what you said was incorrect but has been corrected through feedback.

 

[humbug]

If two people misunderstood what you were saying that probably says more about what you said than how it was understood by others, granted you added caveats and expanded on what you initially said but just because you've done that it doesn't make what you initially said anymore correct, perhaps it's best if we just accept that what you said was incorrect but has been corrected through feedback.

What was incorrect? when talking about Clock Speed irrelevance the conversation was clearly about Ryzen 3000, i was responding to precisely that, if despite this you made other assumptions then that's your problem.

Failing that, well... now we know you agree with my IPC analysis.

Next time take the time to read carefully.

 

[Thekwango]

nvm, it's scary around here.

Get a notification that nashathedog has quoted my post - then see this?? Really hope it’s not me you think is scary

Really intrigued now by what you wrote originally??

 

[Murphy]

What was incorrect? when talking about Clock Speed irrelevance the conversation was clearly about Ryzen 3000, i was responding to precisely that, if despite this you made other assumptions then that's your problem.

Failing that, well... now we know you agree with my IPC analysis.

Next time take the time to read carefully.

Can't you even follow your own conversation? Why are you asking me what was incorrect, I've already provided you with a link to what you said that was incorrect, as you're finding it tricky allow me to emphasis what you got wrong "Clock speed is irrelevant ^^^^^ the FX-9590 was an all core 5Ghz part, it was still crap, IPC is all that matters."

If you can't understand that clock speed is intrinsically linked to instructions per clock, you know what with it being in the name and all, there's really nothing more i can say.

 

[CuriousTomCat]

If two people misunderstood what you were saying that probably says more about what you said than how it was understood by others, granted you added caveats and expanded on what you initially said but just because you've done that it doesn't make what you initially said anymore correct, perhaps it's best if we just accept that what you said was incorrect but has been corrected through feedback.

It's possible that a 4.5GHz Ryzen 3000 CPU has the same FPS in games as a 5GHz Intel CPU, we just dont know yet. Too many people are obsessing over clockspeed. Two pages ago, DG was saying how important it is that AMD hit's 5GHz to stand a chance of beating Intel, like clockspeed is all that matters in games. This train of thought winds me up so I thought Humbugs "clockspeed is irrelevant" comment was justified in the context of the discussion and I'm surprised people took it so literally as if to say he has a disregard for IPC.

IPC is far more important for AMD to get right than clockspeed.

 

[humbug]

Can't you even follow your own conversation? Why are you asking me what was incorrect, I've already provided you with a link to what you said that was incorrect, as you're finding it tricky allow me to emphasis what you got wrong "Clock speed is irrelevant ^^^^^ the FX-9590 was an all core 5Ghz part, it was still crap, IPC is all that matters."

If you can't understand that clock speed is intrinsically linked to instructions per clock, you know what with it being in the name and all, there's really nothing more i can say.

You're still not getting it.

Ok, this is what started it.

AMD have matched Intel's best performance at an unknown clock speed, and its an engineering sample with unfinalized clocks.

With that in mind doesn it matter what the clock speeds are? do you agree it doesn't, they only thing that matters is the performance?

what do you think i was talking about there? its in the middle of a conversion about the CES performance of Ryzen 3000

So when people are talking about its clock speed what do you think i was talking about when i said "its irrelevant"

Cue the rest of the conversation where you got involved.

Clock speed is irrelevant ^^^^^ the FX-9590 was an all core 5Ghz part, it was still crap, IPC is all that matters.

I'll add to this...

2700X stock (4.025Ghz) vs 9900K Stock (4.7Ghz) clock displayed in the OSD. GTX 1080TI 1080P

The 9900K is clocked 17% higher.

There are no results in this so i have taken the averages from the OSD in random places, no shenanigans you can see the run troughs side by side for a true performance comparison.

9900K is faster (on the right) in all, percentage difference in (Brackets)

Ryse of the Tomb raider: 188 - 213 (+12%)
GTA-V: 132 - 153 (+16%)
AC Odyssey: 103 - 116 (+13%)
Forza 4: 156 - 177 (+13%)
Witcher III: 141 - 162 (+15%)
AC Origins: 105 - 113 (+8%)
FC5: 118 - 127 (+9%)
FA4: 106 - 122 (+16%)
Arma III 92 - 112 (+22%)

So that's 10 to 20% with most around 15% to the 9900K, with 17% higher clocks.

The fact is Zen+ gaming performance is excellent when measured by clock speed alone, 15% slower than the 9900K purely based on clock speed advantage 9900K, this engendering sample has 15% higher performance than the 2700X.

 

[Potatowithearsontheside]

Somehow he agrees with me but still has a problem with it, go back over the last two pages and people are arguing that Ryzen 3000 must be high clock speed, seemingly higher than the 9900K, my argument to that is basically this, that is not necessary if the IPC is the same in gaming or higher and provided plenty of examples to show that the 900K's gaming performance advantage comes purely from its higher clock speed, to the tune of 17% higher clocks that make up around 15% higher performance, the engineering sample is 15% faster than the 2700X, how it achives that seems to be the contention, somehow its good if its IPC and not good if its clocks speed with another arguing the opposite.
It seems to me looking for a negative in that which simply does not exist, the performance is there, that is all that matters.

Its the same performance "YEAH BUT BUT BUT CLOCK SPEED"

I understand fully what you're saying, and I agree that ultimately it doesn't matter where the performance comes from, only that it is there. However, I think that you may be missing something in your analysis; IMO the comment about it being "bad" if it us all clocks is more likely a reference to the fact that we know that the ES wasn't on final clocks, and that there'd likely be little room for further clock increases, but also by its very nature the IPC of this gen architecture is unlikely to be further improved ahead of launch. The "bad" reference simply implies that if it is all IPC gain then there's likely more scope for further improvements ahead of launch. *Edit: read...by increasing clocks.*
Your analysis is correct if the ES proves to be similar to the production SKUs.
It's more a perspective thing I think.

 

[Potatowithearsontheside]

Or as a quick summary...
"At the end of the day..."
"But it is not the end of the day yet."

 

[~>Dg<~]

CPU A cycles 50 instruction per clock and is clocked at 1Ghz while CPU B cycles 25 instructions per clock but is clocked at 1.5Ghz.

Which CPU is the higher performance part?

CPU A?
CPU B?

we get this but from the demo just recently the amd chip could have been overclocked or at max potential to be equal with a stock intel chip. we get the arguement but intel in general oc more than amd. so if the amd can match ocd then we already know they behind intel because they always oc more. you know what we mean. if the amd chip is max 4.5 and matches a stock 9900 then we already know they behind in games.

 

[Potatowithearsontheside]

It could also have been massively underclocked, so...?

 

[~>Dg<~]

so lets just keep guessing until its out.

based on 15 years being behind you think it would be underclocked to match a 9900k ?

i love postivity but that is some leap.

on that basis 10 core 5ghz underclocked cpu faster than anything intel for £200. not bad.