16-Core Ryzen 9 3000 Series ES Sample spotted & Zen 2 is a Memory OC Beast, DDR4-5000 Possible! (Updated)

[Neftex]

@valdyrgramr this is pointless, let me just say the world doesnt work on 100% certainty. hope youll learn this soon because you cant get proof of everything you want

[Neftex]

1 minute ago, valdyrgramr said:

You could just quote me, and again I don't blindly trust people rather you like it or not.  Blindly trusting people who can't prove shit is the reason why millions of people have been killed.   

im pretty sure not trusting people killed millions as well

[Neftex]

Just now, valdyrgramr said:

No, that was not trusting proof.

what proof does a agent have if he just heard someone planning invasion while undercover? were back to trust

[Neftex]

4 minutes ago, valdyrgramr said:

Nice derailing, but um the agent has to use proof to be trusted.  

i like youre utopian view

 

EDIT: im not the one who derailed cpu leak into killing people

[LAwLz]

2 hours ago, Neftex said:

but youre saying his leak isnt true because you dont have a proof. it goes the other way too, wheres your proof the leak isnt true?

No it actually doesn't.

There is a thing called "burden of proof". The burden to prove something is on the one making the claim. If I claim I have a pink elephant which vomits pure alcohol in my living room, would you believe me? If not, then do you have any evidence that I don't have it!? If I make a statement like the one above, it is my responsibility to prove that my claim is accurate.

 

Also, saying something might not be true is not the same as saying it is false.

Just because someone chooses not to believe in something, does not mean they automatically believe the opposite. There is a third position you can have which is "I don't know". If I told you "my brother works for AMD and my sister works for Nvidia", would you believe me? I probably wouldn't if I were you, but I would not automatically assume that statement was false either. I would be rather indifferent about it and think that it could be true, but it might not be. The only way to know is if proof supporting either side was presented.

[DildorTheDecent]

Big zooms.

 

16 core itx rig would be boolin'.

[Neftex]

3 hours ago, LAwLz said:

No it actually doesn't.

There is a thing called "burden of proof". The burden to prove something is on the one making the claim. If I claim I have a pink elephant which vomits pure alcohol in my living room, would you believe me? If not, then do you have any evidence that I don't have it!? If I make a statement like the one above, it is my responsibility to prove that my claim is accurate.

 

Also, saying something might not be true is not the same as saying it is false.

Just because someone chooses not to believe in something, does not mean they automatically believe the opposite. There is a third position you can have which is "I don't know". If I told you "my brother works for AMD and my sister works for Nvidia", would you believe me? I probably wouldn't if I were you, but I would not automatically assume that statement was false either. I would be rather indifferent about it and think that it could be true, but it might not be. The only way to know is if proof supporting either side was presented.

guess ill have to repeat myself

6 hours ago, Neftex said:

hes not obligated to prove any of it, this isnt a court (and even in court there are cases where the defendant has to prove he didnt do it instead of the accuser proving he did it)

 

5 hours ago, Neftex said:

yes he could be right or wrong, and thats why you look at the record and see if hes reliable. if his previous leaks are mostly true then you can assume he has good verified sources and consider the next leaks also most likely true.

 

if someones previous leaks are 100% correct, do you just disregard his new leak as "could be right or wrong"? i wouldnt, unless its batshit crazy and other reliable leaks say otherwise.

 

again, he doesnt care if you believe it and neither do i. i dont even care for this leak itself

 

[leadeater]

3 minutes ago, VegetableStu said:

we're 15 days away from the keynote ._. don't be biting each other on whether the crystal ball is right or not

My crystal ball is better anyway, it has RGB.

[Derangel]

To all the yammering about fake/real proof/not proof crap: All leaks, regardless of the reliability of the leaker, should be take with a grain (or more) of salt. Not only does that leave a person open to the chance of the leak being wrong (whether intentionally or not) but it also keeps hype a bit more in check. Zen has been outstanding, but that doesn't mean things can't go tits up or fail to meet expectations. Massive overhype from the community makes it very likely that a product will never be able to meet the over-inflated expectations driven by consumers and cause the product to be "disappointing" even if it is a good product.

 

Discussing potential leaks like this is fun and its interesting to engage in "what if its true" trains of thought, but we should be careful not to let that lead to "this is what it must be" thoughts or to even expect the leaks to be 100% true. There are so many "leaks" and rumors floating around for Ryzen 3 that it would be impossible for all of them to be right and even ones that seem "reasonable" could be utter bollocks.

[RejZoR]

2 hours ago, leadeater said:

My crystal ball is better anyway, it has RGB.

Wow, poor dude. Mine has ray tracing and projects realistic RGB on the crystal ball surface.

[leadeater]

 

Tech Jesus has spoken, click the link to find out more. Yes that is bait and it's time stamp about Zen2.

[Stefan Payne]

15 hours ago, valdyrgramr said:

Isn't it Zen, Zen+, Zen 2, Zen2+, or are they just skipping straight to Zen 3 and not doing tuning this time?

there's 6nm wich uses the 7nm Libarys, so Zen2+ is highly likely.

[cj09beira]

40 minutes ago, Stefan Payne said:

there's 6nm wich uses the 7nm Libarys, so Zen2+ is highly likely.

i dont think we will get a zen 2+, my guess is zen 3 will use either 6nm or 5nm and will be out next year,

[Stefan Payne]

34 minutes ago, cj09beira said:

i dont think we will get a zen 2+, my guess is zen 3 will use either 6nm or 5nm and will be out next year,

Zen3 will be a normal shrink, wich will be 5nm.

 

Zen2+ will be a slight improved Zen2, moved to the very similar "N6" Node

https://www.anandtech.com/show/14290/tsmc-most-7nm-clients-will-transit-to-6nm

 

Money Quote:

Quote

N6 uses the same design rules as N7 and enables designers of chips to re-use the same design ecosystem 

So basically its equal to 14nm GF -> 12nm GF.

 

 

There is no reason to not use it, if its available!

It could give AMD some advantages (higher clock, lower power)...

[Stefan Payne]

2 minutes ago, VegetableStu said:

AMD hasn't mentioned anything about a Zen2+ though o_o maybe they might use it for Zen2 APUs?

Because that would "Osborne" their Zen2.

You don't talk about it if you don't have to and the Product isn't really on the Market...

You only talk about things that you want to sell. Not about things that you might eventually sell next year.

 

And that's when it will be released: Mid-End 2020.

Though its possible that I'm wrong and it won't happen and they move directly to N5.

 

We'll have to wait and see...

But its far far away anyway.

[Stefan Payne]

1 minute ago, VegetableStu said:

just thought of another possibility (although the substrate mounting might be an issue): the single-die ryzens are released earlier with 7nm, and the dual-die ryzens might come out much later with 6nm

 

mmm so much speculation, LOL

From the Anandtech Article it looks to me that N6 should be cheaper than N7 because they moved towards EUV and thus need less steps.

 

The next gen APUs, after the 12nm Picasso, might be N6 and never see N7. Though N7+ might also be possible, if TSMC makes AMD a good offer for that process...

[S w a t s o n]

38 minutes ago, Stefan Payne said:

Zen3 will be a normal shrink, wich will be 5nm.

 

Zen2+ will be a slight improved Zen2, moved to the very similar "N6" Node

https://www.anandtech.com/show/14290/tsmc-most-7nm-clients-will-transit-to-6nm

No such thing as Zen2+, this I am sure of.

 

I do agree that Zen3 could move to 5nm but it has different design rules, 6nm comes out after 5nm so that wont be a thing. 7nm+ is either still the plan for Zen 3 or AMD will go all in and will redesign for 5nm but I'm not sure they will, no need to risk a new node (even though 5nm seems to be coming along nicely)

[schwellmo92]

2 hours ago, S w a t s o n said:

No such thing as Zen2+, this I am sure of.

 

I do agree that Zen3 could move to 5nm but it has different design rules, 6nm comes out after 5nm so that wont be a thing. 7nm+ is either still the plan for Zen 3 or AMD will go all in and will redesign for 5nm but I'm not sure they will, no need to risk a new node (even though 5nm seems to be coming along nicely)

7nm+ also requires re-design AFAIK. 

[cj09beira]

Just now, Jack_of_all_Trades said:

 

To hell with logic and practicality, hype trains run on dreams and unreasonable expectations and we have so much of that. CHOO CHOO mofos!

5ghz boost isn't based on ignoring logic, we will know soon enough

[S w a t s o n]

1 hour ago, schwellmo92 said:

7nm+ also requires re-design AFAIK. 

Zen 3 was built on 7nm+ but I have heard conflicting reports of it requiring new designs rules and not vs 7nm. Logically it should not because it's a true successor for 7nm and 6nm is based on 7nm+, 6nm is confirmed to have the same design rules as 7nm

[Trixanity]

5 hours ago, Stefan Payne said:

Zen3 will be a normal shrink, wich will be 5nm.

 

Zen2+ will be a slight improved Zen2, moved to the very similar "N6" Node

https://www.anandtech.com/show/14290/tsmc-most-7nm-clients-will-transit-to-6nm

 

Money Quote:

So basically its equal to 14nm GF -> 12nm GF.

 

 

There is no reason to not use it, if its available!

It could give AMD some advantages (higher clock, lower power)...

[Stefan Payne]

1 hour ago, Trixanity said:

 

Yes, and??
That doesn't mean that they couldn't update Zen2 with N6, wich should be cheaper than N7, at no real effort...

[cj09beira]

1 minute ago, Stefan Payne said:

Yes, and??
That doesn't mean that they couldn't update Zen2 with N6, wich should be cheaper than N7, at no real effort...

that would only be true if they can afford to delay zen 3, which i would not advise to, 

[ravenshrike]

27 minutes ago, cj09beira said:

that would only be true if they can afford to delay zen 3, which i would not advise to, 

If the rumors regarding on-die memory changes and 4 way SMT are true for Zen 3 they might end up going with a Zen 2+ refresh for Ryzen 3, 5, 7, and 9 while putting out EPYC and Threadripper on Zen 3 until the process matures.

[S w a t s o n]

1 hour ago, Stefan Payne said:

Yes, and??
That doesn't mean that they couldn't update Zen2 with N6, wich should be cheaper than N7, at no real effort...

6nm wont be out until after 5nm, long after Zen3 exists. It's a bit odd but 6nm is 7nm++ meant for people that want the best perf on N7 design rules. AMD wont be waiting on nodes, they know Intel's 7nm is coming along well