2400G at 4.5ghz Spotted in 3d mark

[Space Reptile]

well there is not much to quote , so i hope this still conforms 

but according to this result https://www.3dmark.com/compare/fs/14879086/fs/14755451 (scroll down and open the dropdowns for the cpu )
the tested Ryzen ran at an astonishing 4556mhz boost , its very likely to be heavily overclocked and watercooled

 

but impressive non the least 

this was also accomplished on an MSI B350I ITX motherboard wich is not a fancy overclocking series motherboard , likely included w/ a review kit 

[Coaxialgamer]

What I'm wondering is if AMD used an updated version of their 14nm process or if they're using the exact same manufacturing tech as ryzen. 

[DocSwag]

Could be LN2, just saying.

[FloRolf]

S I C C

[Humbug]

The Zen+ refresh of the desktop lineup which AMD has taped out and is expected to launch in April will be built on the Global Foundries 12 nano meter process, and AMD has stated that this brings higher clocks. So 4.5Ghz overclocked for those would be expected.

 

But these Ryzen APUs?
As far as I know these are the same first gen Zen cores, just with Vega graphics. Don't see how it could clock like this. Unless I missed something and it's on 12nm.

 

[MaktimS]

11 minutes ago, Coaxialgamer said:

What I'm wondering is if AMD used an updated version of their 14nm process or if they're using the exact same manufacturing tech as ryzen

The APUs run on the 14nm and Ryzen 2000 on 12nm process I think

[Space Reptile]

oh and for those who care , here is a stock 2400G vs a i5 7500 & GT 1030
https://www.3dmark.com/compare/fs/14878503/fs/14842180

[Bananasplit_00]

well we can at least hope this was on a shitty air cooler, that would be sweet. like, its really unlikely and probably really not, but i want to hope at least  

[Trixanity]

12 minutes ago, Humbug said:

As far as I know these are the same first gen Zen cores, just with Vega graphics. Don't see how it could clock like this. Unless I missed something and it's on 12nm.

It's 14nm but architecturally it's more similar to Pinnacle Ridge which launches in April.

[Eduard the weeb]

unless someone stole it from AMD thats a tech reviewer or new publication and they have the supplies to push it to the MAX LEVEL

[Nicnac]

damn son

[cj09beira]

34 minutes ago, Coaxialgamer said:

What I'm wondering is if AMD used an updated version of their 14nm process or if they're using the exact same manufacturing tech as ryzen. 

 

22 minutes ago, Humbug said:

But these Ryzen APUs?
As far as I know these are the same first gen Zen cores, just with Vega graphics. Don't see how it could clock like this. Unless I missed something and it's on 12nm.

 

amd for ryzen 1 used the more energy efficiency focused library of 14nm, i guess mostly so that they can make epyc have good power consumption and also so that it would kill the amd is hot meme, they could have switched to the more frequency/ performance library and in doing so they could have lifted the 4ghz wall,

if thats the case these chips will be a very very good choice for entry level builds, fingers crossed for next weeks reviews

[Taf the Ghost]

It's thought to be a reporting bug, I believe.

[cj09beira]

7 minutes ago, Taf the Ghost said:

It's thought to be a reporting bug, I believe.

dont say something like that  

[Humbug]

6 minutes ago, Taf the Ghost said:

It's thought to be a reporting bug, I believe.

But click through to the 3d mark site and look at the performance figures compared to the Ryzen 5 2500U.

The graphics, physics and overall performance scores are all 60-80% faster. Which suggests it was really running at those clocks.

[Brooksie359]

Just now, cj09beira said:

dont say something like that  

I mean any time there are results like this it seems like it could be a reporting bug. 

[Taf the Ghost]

12 minutes ago, Humbug said:

But click through to the 3d mark site and look at the performance figures compared to the Ryzen 5 2500U.

The graphics, physics and overall performance scores are all 60-80% faster. Which suggests it was really running at those clocks.

There was a timing bug with the early Ryzen BIOS in some configurations. If its misreporting time, it's misreporting scores.

 

I expect them to be good, but I don't expect 4.5 Ghz without at least Beyond normal water cooling.

[samcool55]

2 hours ago, Coaxialgamer said:

What I'm wondering is if AMD used an updated version of their 14nm process or if they're using the exact same manufacturing tech as ryzen. 

Well, bit of both, it still uses normal Zen but with a few optimizations, but not Zen+. At least that's what i heard...

[Sakkura]

1 hour ago, Eduard the weeb said:

unless someone stole it from AMD thats a tech reviewer or new publication and they have the supplies to push it to the MAX LEVEL

Reviewers aren't usually the ones who push it to the max with LN2. They will want representative performance numbers first and foremost.

[Doobeedoo]

Very nice.

[Space Reptile]

and heres a video on the matter 

[Taf the Ghost]

@Space Reptile

 

That guy's approach could be actually really interesting to find testing of new products beyond the G-series parts.

 

As to the results, they look pretty good. It'll be interesting to see the scaling effects in games with GPU overclocks. Should be some really fun testing for a lot of the review space. For AMD, their Fusion design finally is coming together. 

[asus killer]

if that is true, ryzen+ seems all the more sweeter 

 

having just 4 cores should help no? i mean in a cpu with more cores should be more difficult, more heat?

[Space Reptile]

1 minute ago, asus killer said:

if that is true, ryzen+ seems all the more sweeter 

 

having just 4 cores should help no? i mean in a cpu with more cores should be more difficult, more heat?

yea , extreme overclockers usually disable a chip down to like one or 2 cores to get the max out of it 

[SpaceGhostC2C]

9 hours ago, DocSwag said:

Could be LN2, just saying.

That would be news still, since Ryzen's inability to overclock higher than 4.2GHz (4.0 most of the times) wasn't due to thermals.