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

[S w a t s o n]

1 minute ago, Dan Castellaneta said:

I would say that considering how many leaks have come out that have been disproven in general (not really by this source but just in general), it's always good to hold a decent amount of skepticism. It will be very interesting, though, if AMD drops this bomb in a few months. I'd just be interested in how well it holds up.

Well this is just an ES leak anyways, not sure we will ever see it actually confirmed or disproven but if I was going to trust any leak, this guy is up at the top of the list.

[MadDuke]

3.3 Base with 4.2 boost on 7nm performance node  compared to  2950x and 3.5 base with 4.4 boost on power saving 12nm?   

This makes absolutely no sense

[Drak3]

1 minute ago, MadDuke said:

3.3 Base with 4.2 boost on 7nm performance node  compared to  2950x and 3.5 base with 4.4 boost on power saving 12nm?   

This makes absolutely no sense

It's an engineering sample. Those are typically clocked lower.

[MadDuke]

5 minutes ago, Drak3 said:

It's an engineering sample. Those are typically clocked lower.

Don't remember a single case with an AMD CPU where that was the truth.

[Tristerin]

 

[Jurrunio]

2 hours ago, Jack_of_all_Trades said:

How lower are usually engineering samples compared to retail ?

There are the optimistic and pessimistic side of things. First gen Ryzen for example, first leaked ES runs 2.8 base and 3.2 boost, and the final product does 3.8 to 4 when overclocked. Roughly 20-25% lower than what it could do. Same apply to the Zen 2 leaks, then it could hit 5. However Zen+ ES and retail pretty much clocks the same.

[S w a t s o n]

6 minutes ago, MadDuke said:

Don't remember a single case with an AMD CPU where that was the truth.

Are you trolling?

Almost every CPU ES goes like this, you can find lots of Ryzen 1 ES leaks with lower clocks than release...Epyc, Epyc 2 as well
 

See:

1 minute ago, Jurrunio said:

There are the optimistic and pessimistic side of things. First gen Ryzen for example, first ES runs 2.8 base and 3.2 boost, and the final product does 3.8 to 4 when overclocked. Roughly 20-25% lower than what it could do. Same apply to the leaks, then it could hit 5. However Zen+ ES and retail pretty much clocks the same.

 

[MadDuke]

2 minutes ago, Tristerin said:

 

So, it just confirms what I'm saying.

 

 

The leaked Enginerring sample was:  3.4  base and final version was 3.4 base

 

And this leak for the new part is 3.3 base on a new node compared to 3.5 on an old node?  Please

[cj09beira]

28 minutes ago, MadDuke said:

3.3 Base with 4.2 boost on 7nm performance node  compared to  2950x and 3.5 base with 4.4 boost on power saving 12nm?   

This makes absolutely no sense

 

25 minutes ago, Drak3 said:

It's an engineering sample. Those are typically clocked lower.

dont forget tdp, am4 wont be able to dissipate as much

[Drak3]

1 minute ago, cj09beira said:

 

dont forget tdp, am4 wont be able to dissipate as much

AM4 can still dissipate enough heat for a 16C chip to achieve the same or slightly higher clocks.

[MadDuke]

1 minute ago, cj09beira said:

 

dont forget tdp, am4 wont be able to dissipate as much

Actually. That is not true. It would be true if by some miracl it was a 32Core64Thread Part.  Nothing is preventing AM4 with dissipation with this design. Nothing at all. 

[cj09beira]

Just now, MadDuke said:

Actually. That is not true. It would be true if by some miracl it was a 32Core64Thread Part.  Nothing is preventing AM4 with dissipation with this design. Nothing at all. 

just saying am4 wont get 180w cpus

[MadDuke]

3 minutes ago, cj09beira said:

just saying am4 wont get 180w cpus

Just saying.   180W was on 12nm.   and it was clocked higher than this "leak"

[Jurrunio]

9 minutes ago, cj09beira said:

just saying am4 wont get 180w cpus

You can get a 2700X to push more power than that, it all comes to how aggressive factory specs are

[sazrocks]

44 minutes ago, MadDuke said:

Don't remember a single case with an AMD CPU where that was the truth.

Might want to take some Prevagen then.

https://www.kitguru.net/components/cpu/matthew-wilson/amd-zen-engineering-sample-benchmark-leaks/

https://www.tweaktown.com/news/53101/amds-new-zen-engineering-samples-arrive-up-32-cores/index.html

https://arstechnica.com/gadgets/2016/12/amd-zen-performance-details-release-date/

https://wccftech.com/amd-ryzen-5-2600-pinnacle-ridge-cpu-performance-leak/

 

Edit: and I see you also found this post:

Congratulations, you've proved yourself wrong. While the retail 2600 did indeed have a base clock of 3.4GHz, its boost clock was actually 3.9GHz, 100MHz more than the engineering sample

[MadDuke]

15 minutes ago, Jurrunio said:

You can get a 2700X to push more power than that, it all comes to how aggressive factory specs are

 

It's not a problem to push above 180W.  It's a problem to push on stock with a CPU which should in theory work normally even on a B350 board-

 

8 minutes ago, sazrocks said:

Might want to take some Prevagen then.

https://www.kitguru.net/components/cpu/matthew-wilson/amd-zen-engineering-sample-benchmark-leaks/

https://www.tweaktown.com/news/53101/amds-new-zen-engineering-samples-arrive-up-32-cores/index.html

https://arstechnica.com/gadgets/2016/12/amd-zen-performance-details-release-date/

https://wccftech.com/amd-ryzen-5-2600-pinnacle-ridge-cpu-performance-leak/

 

Edit: and I see you also found this post:

Congratulations, you've proved yourself wrong. While the retail 2600 did indeed have a base clock of 3.4GHz, its boost clock was actually 3.9GHz, 100MHz more than the engineering sample

Nobody cares about boost clocks on engineering samples.  Everything important is and always was in the base clocks.  Boost clocks are tuned just before launch.

[sazrocks]

3 minutes ago, MadDuke said:

Nobody cares about boost clocks on engineering samples.  

But you do, apparently

1 hour ago, MadDuke said:

3.3 Base with 4.2 boost on 7nm performance node  compared to  2950x and 3.5 base with 4.4 boost on power saving 12nm?   

This makes absolutely no sense

 

6 minutes ago, MadDuke said:

Everything important is and always was in the base clocks.

Okay fine.

https://www.techpowerup.com/224919/amd-summit-ridge-zen-cpu-at-2-80-ghz-beats-3-40-ghz-core-i5-4670k

Quote

This sample featured clock speeds of 2.80 GHz, with 3.20 GHz boost. 

Last time I checked the base clock of the 1700 was 3.0GHz

[MadDuke]

2 minutes ago, sazrocks said:

But you do, apparently

 

Okay fine.

https://www.techpowerup.com/224919/amd-summit-ridge-zen-cpu-at-2-80-ghz-beats-3-40-ghz-core-i5-4670k

Last time I checked the base clock of the 1700 was 3.0GHz

You are comedic dude.   Before Ryzen was a thing taking a 200 MHz difference Compared to more stable of the last year with smaller differences.     This is not a leak in this topic. This is idiotic and in now way realistick to have a 200 MHz slower thant the equivalent of the older node   

 

AMD Enginering samples are almost always spot-on with the base clock.  This is then a bullshit leak. 

[S w a t s o n]

1 hour ago, valdyrgramr said:

It still means nothing because it's an ES and not a final product.  And, being popular doesn't mean you're credible.

Being right means you're credible, he's right

[sazrocks]

16 minutes ago, MadDuke said:

Before Ryzen was a thing taking a 200 MHz difference

I'd argue that it was even more than that. The ES leaked in the link I posted was probably something pushing what they had at the time. The 1800X actually launched with a base clock of 3.6GHz, pushing the gap to 800MHz. And since the 1800X and 1700 are essentially the same CPU just clocked differently, that is probably a better comparison.

16 minutes ago, MadDuke said:

Compared to more stable of the last year with smaller differences.

Yes, a minor revision in architecture and minor change in process node are not as likely to cause crazy swings in clock speeds. 

Compare that to a major revision in architecture (Zen 2), a major change in process node (14nm -> 7nm is a HUGE jump), and even an overhaul of the basic design of the chip (3 dies on one package), and I can really see why clock speeds might vary a lot from ES to ES and from ES to retail.

16 minutes ago, MadDuke said:

his is not a leak in this topic.

 

16 minutes ago, MadDuke said:

This is then a bullshit leak. 

Maybe. But the source is well-respected and has a very high accuracy on these kinds of things, so without any evidence to the contrary I'm inclined to believe this ES is legit.

[S w a t s o n]

1 minute ago, valdyrgramr said:

How is he right?  He just claimed this on twitter and had some guy say he's right.  <.<   Even if he's right about 1 ES it means nothing really compared to the final product.

 

Because this isnt his first rodeo, you clearly dont know him if you think this is the first thing he's leaked. Just scroll down his twitter lmao

[Ithanul]

5 hours ago, porina said:

Having a 5000 ram speed setting is no guarantee it'll get anywhere near that. It is way beyond DDR4 specifications and maybe only something for extreme overclockers to play with.

Plus, that with 1/2 the IF as well.

 

I rather not even 1/2 the IF, but stick to around 3200-3466MHz speeds with tighter timings and full IF speeds.

[S w a t s o n]

Just now, valdyrgramr said:

He still needs to provide evidence and not just make a claim on twitter then have some guy say he and this is credible.  And again, it doesn't mean anything about the final product it's an engineering sample.  This is just more pointless hypetrain bs.

yikes, the thread is about the es leak not the final product. the leak still matters. prototype leaks definitely matter to people

He's been right many times, how hard is it to understand

[Okjoek]

I'm probably going to try and stick with my i5 3450 until platforms start using DDR5 memory. Real Chads skip DDR4 entirely!

[S w a t s o n]

1 minute ago, valdyrgramr said:

So, because he was right before means he's auto-correct now?  I'd like to see his proof not go on blind faith.  And again, this is just adding to hype.

Alright, I hope to see you complain in every phone leak thread with a prototype and people hyping shit