China mobo manu leaks | 15% ipc gain zen 2

tangoseal said:
If you down clock the Intels aggressive 5.x ghz clock to AMDs peak 4.x ghz then suddenly you see the AMD is actually faster per cycle in IPC across the board than the Intel. The only advantage really Intel has at this point in the race is gross clock speed.
Click to expand...

I have an artificial leg. I'd rather race against you with my leg on than chop one of yours off so we can compete on an 'even' basis. :whistle:

Downclocking an Intel chip so they'll be 'even' doesn't take into account that Intel chips are engineered to clock faster. If you want them to have the same clockspeed, AMD needs to engineer their chips to do so. Bulldozer and derivative chips could clock well. My brother-in-law had a highly-clocked AMD system. It didn't do him any good against the slower-clocked Intel chips that performed far better than their clocks indicated.
 
Algrim said:
I have an artificial leg. I'd rather race against you with my leg on than chop one of yours off so we can compete on an 'even' basis. :whistle:

Downclocking an Intel chip so they'll be 'even' doesn't take into account that Intel chips are engineered to clock faster. If you want them to have the same clockspeed, AMD needs to engineer their chips to do so. Bulldozer and derivative chips could clock well. My brother-in-law had a highly-clocked AMD system. It didn't do him any good against the slower-clocked Intel chips that performed far better than their clocks indicated.
Click to expand...

I'm half blind in an eye. I'd rather have my eye and no leg. Anyways...

Your missing the point.

If the AMD can damn near keep up with the Intel as its wide open and you down clock the Intel and the whopping 9% margin becomes equal it shows that if the 2700x as it sits were capable of 5.1ghz like the 9900k it would wipe the floor with the Intel.

That's the point all you Intel soldiers are missing. Probably because the bias blindness.
 
tangoseal said:
That's the point all you Intel soldiers are missing. Probably because the bias blindness.
Click to expand...

Intel soldier? Naw... When I bought my CPU/motherboard AMD had nothing I wanted. If my computer died now I wouldn't buy an Intel CPU at all. I'd be trying to decide (or, more likely, convince my wife) to go TR. But you are welcome to believe that anyone who has an Intel/nVidia system in their sig is a lemming instead of realizing that people purchase what's available to them. If AMD had a CPU and GPU that was worth buying when I needed to buy I'd have a much different system in my sig.
 
Algrim said:
Intel soldier? Naw... When I bought my CPU/motherboard AMD had nothing I wanted. If my computer died now I wouldn't buy an Intel CPU at all. I'd be trying to decide (or, more likely, convince my wife) to go TR. But you are welcome to believe that anyone who has an Intel/nVidia system in their sig is a lemming instead of realizing that people purchase what's available to them. If AMD had a CPU and GPU that was worth buying when I needed to buy I'd have a much different system in my sig.
Click to expand...

No I dont believe you are a lemming at all. I have no idea how witful, intelligent, or knowledgeable you are about whatever it is your smart in. But there is a pattern in people that are extremely anti-amd because they are just incredibly biased for Intel and really can't provide reason. Even if you aren't one of those types, for a minute it seemed as if you emulated that typical Intel attitude.

Again I am not name calling. I was only referencing the fact that AMD is par for the course for surpassing Intel as soon as they can get the clock speeds elevated a little bit more.

This is what all hope for. This is good because we need competition to drive innovation. AMD beats Intel, then Intel beats AMD, rinse repeat over and over.
 
Last edited:
tangoseal said:
it shows that if the 2700x as it sits were capable of 5.1ghz like the 9900k it would wipe the floor with the Intel.
Click to expand...
With 2700X and 9900K both overclocked to 5.4GHz both score the same in Cinbench a program that doesn't care about memory speed.
[email protected] H20 2378 cb
https://hwbot.org/submission/4010098_protoaus_cinebench___r15_core_i9_9900k_2378_cb
[email protected] LN2 2376 cb
https://hwbot.org/submission/4004853_bilko_cinebench___r15_ryzen_7_2700x_2376_cb

Any program that does care about memory speed the 2700X will fall behind by ~10-20% when at the same core clock speed and RAM speed more if you run ~4600c18 on the 9900K vs ~3600c14 on the 2700X.
Hopefully we will see much higher memory speeds with Ryzen 3000 thanks to the memory being unlocked from the infinity fabric but how much affect this and other tweaks will have on latency we have yet to see.


TimeSpy Physics cares a little about RAM latency but not as much as most games
This is the best 2700X score I could find vs a lower clocked 9900k and yet it is 24% faster
2700X@5279MHz RAM@3200 =10k
https://www.3dmark.com/spy/5471101

9900K@5104MHz RAM@4000 =12447
https://www.3dmark.com/spy/4981631

Don't get me wrong I am all for AMD smashing Intel and in some metrics they will, I just wont be getting my hopes to high that they will win across the board but I am hopeful they will at least get dam close.
 
Last edited:
tangoseal said:
No I dont believe you are a lemming at all. I have no idea how witful, intelligent, or knowledgeable you are about whatever it is your smart in. But there is a pattern in people that are extremely anti-amd because they are just incredibly biased for Intel and really can't provide reason. Even if you aren't one of those types, for a minute it seemed as if you emulated that typical Intel attitude.
Click to expand...

To me, a computer (and its operating system) are just tools. I, personally, don't care if you go AMD, ARM, Intel, POWER, SPARC, etc. as long as you're getting the processor that meets your needs. For me, that was Intel when I built my system. Now, it would be Ryzen or ThreadRipper.

Interestingly, you indicate that I exhibit behavior as someone who is extremely anti-AMD (I'm not; I'm anti-hype). There's also a pattern of people who are extremely pro-AMD. Usually, this pattern asserts itself with someone claiming that if only we could hobble Intel down to AMD's level, AMD would be better (if only this, then that). I can postulate a world where POWER will reign supreme on the desktop if only sufficient conditions exist. That doesn't make it likely to happen.
 
Galvin said:
Hopefully we'll know in 3 to 4 weeks. But don't think it'll really matter. Its either go for intels 14++++++++++++++++++ this summer. Or get something that's actually new from AMD
Click to expand...

how many pluses does it take to get to 10nm?

intel still doens't know.
 
dasa said:
With 2700X and 9900K both overclocked to 5.4GHz both score the same in Cinbench a program that doesn't care about memory speed.
[email protected] H20 2378 cb
https://hwbot.org/submission/4010098_protoaus_cinebench___r15_core_i9_9900k_2378_cb
[email protected] LN2 2376 cb
https://hwbot.org/submission/4004853_bilko_cinebench___r15_ryzen_7_2700x_2376_cb

Any program that does care about memory speed the 2700X will fall behind by ~10-20% when at the same core clock speed and RAM speed more if you run ~4600c18 on the 9900K vs ~3600c14 on the 2700X.
Hopefully we will see much higher memory speeds with Ryzen 3000 thanks to the memory being unlocked from the infinity fabric but how much affect this and other tweaks will have on latency we have yet to see.


TimeSpy Physics cares a little about RAM latency but not as much as most games
This is the best 2700X score I could find vs a lower clocked 9900k and yet it is 24% faster
2700X@5279MHz RAM@3200 =10k
https://www.3dmark.com/spy/5471101

9900K@5104MHz RAM@4000 =12447
https://www.3dmark.com/spy/4981631

Don't get me wrong I am all for AMD smashing Intel and in some metrics they will, I just wont be getting my hopes to high that they will win across the board but I am hopeful they will at least get dam close.
Click to expand...


That Cinebench comparison is not valid.
2700x is running the RAM @ 1,875MHz
9900k is running the RAM @ 4,040MHz

You say they are equal.
Cinebench does care about memory speed (at least it does for Ryzen)
What I see is the 2700x has better IPC, and is STILL faster (though within margin of error) than the 9900k despite the HUGE RAM speed disadvantage.

Also, why is the 2080ti clocked faster in the 9900k bench? Does that affect CPU score? (Genuinely don't know)


The 3dmark one is a waaay better comparison and goes to show how RAM latency affects gaming.
However in the real world I doubt a 5.2 GHz 2700x would have any trouble at all keeping up with OC'd 9900k 144hz @ 1080p in real games.
 
tangoseal said:
If you down clock the Intels aggressive 5.x ghz clock to AMDs peak 4.x ghz then suddenly you see the AMD is actually faster per cycle in IPC across the board than the Intel. The only advantage really Intel has at this point in the race is gross clock speed.

Possibly thier ringbus but small tweaks to infinity on AMDs part is rapidly closing that gap.

Lastly I too own Intel products as well so I'm not being lopsided here.
Click to expand...

Do you have any objective measurements on this ?
I would like to see them

Last time i saw an IPC comparison on 1700x the 1700 was lower than Intel on the same clock per core
or did you mean more IPC due to more cores?
 
Rockenrooster said:
That Cinebench comparison is not valid.
2700x is running the RAM @ 1,875MHz
9900k is running the RAM @ 4,040MHz
Click to expand...
The RAM speeds are a bit closer than that one is listing actual MHz the other has posted the double data rate speed, the actual clock speed for the 4000 kit is 1999MHz

Once you take timings into account 3618c12 vs 4000c15 the RAM used for the Ryzen Cinbench score is actually significantly faster than what was used for Intel.

While I don't currently have a GPU installed to show if there is a difference from GPU clock speed, the IGP score in TimeSpy phisics is around the scores I achieved with GTX1070 and Radeon VII so my guess is no it doesnt make a difference.
 
Last edited:
There are credible leaks now showing that the 16 core Ryzen 3000 is hitting around a 4600 in r15. So thats blistering powerful compared to last gen. Like a 15% increase in IPC over 2700x alone.

Just search youtube for Zen 2 leaks ... you will find credible sources. Not going to list them right now. Busy playing Titanfall 2.
 
tangoseal said:
No I dont believe you are a lemming at all. I have no idea how witful, intelligent, or knowledgeable you are about whatever it is your smart in. But there is a pattern in people that are extremely anti-amd because they are just incredibly biased for Intel and really can't provide reason. Even if you aren't one of those types, for a minute it seemed as if you emulated that typical Intel attitude.

Again I am not name calling. I was only referencing the fact that AMD is par for the course for surpassing Intel as soon as they can get the clock speeds elevated a little bit more.

This is what all hope for. This is good because we need competition to drive innovation. AMD beats Intel, then Intel beats AMD, rinse repeat over and over.
Click to expand...

I'm biased for Intel for a while because Intel systems provided a livelihood for my family for a long time. They just worked and were better, and when they got old they sold for part of their value back. So I paid more to use and got paid more to sell.

Now since I don't use the computer to do as much except play games, amd is fine. But if you wanted a reason why I loved Intel, it was because I knew they were faster and would provide the value I needed to earn a living for me and my family.
 
So when is the best estimate of when the new AMD stuff will be able to be purchased? Ryzen 7 3700X has my name written all over it, if the info is correct!:)
 
  • Like
Reactions: blkt
like this
N4CR said:
Lol yeah he probably got pussy whipped into supporting her political ideology of choice. He was pretty much 'paid level thread derailing' so doesn't surprise me to see him go with the sheckels for another job. He gave up last year after lots of us called him out for it. He'd just come in, thread crap and drag it off topic if it was anything positive for AMD, or steer the conversation to Intel (in an AMD forum lol). I bet he now works for shareblue or some other pathetic political disruption/shilling/trolling/'organic discourse' company on facebook/forums/cuckchan etc.
We have a new version of him, that trans guy/girl/it whatever that tries to make every thread about sexuality, patriarchy, fucking white males and women/gays dindu nuffin lmao. Never blocked anyone here but they're getting awfully close.
And there are others who are a watered down juan, always a few shareholders floating around here too. Usually they are easy to spot though because when AMD does something good they are not happy and usually know nothing about what they are crapping on.

Anyyyway back OT..
This is inline with expected IPC improvements. Upper end though, i was expecting ~10-12% on most workloads if all goes well. 15% might be only specific workloads if true.
4.5GHz must be all core. I would expect higher in single-dual core under PBO3 or whatever they'll have by then because we have current 2700x already reaching around there.
Click to expand...

Cool history, bro
 
tom_ozahoski said:
This guy is still here?
Click to expand...
Yes.. if you click his name, it says last online today. Could have probably figured it out yourself without asking, but hey, it was like a whole 1 click, that's a lot of effort to put in. ;).

PS. Just having fun, don't take it personal, seemed to fit this rediculous revived thread.
 
Ready4Dis said:
Yes.. if you click his name, it says last online today. Could have probably figured it out yourself without asking, but hey, it was like a whole 1 click, that's a lot of effort to put in. ;).

PS. Just having fun, don't take it personal, seemed to fit this rediculous revived thread.
Click to expand...


Yea I know but that's just too much effort to make for him.
 
You must log in or register to reply here.
Back
Top