No New FX Processor From AMD in 2013

[nt300]

followed by "Richland" third-generation desktop APU, which combines "Piledriver" CPU components with "Radeon 2.0 cores"

Shouldnt this be with Steamroller Cores
What is AMD doing.

Don't quote me on this, but isn't steamroller really just piledriver on 28nm?

which would still be piledriver...

Just sayin'.

Steamroller is suppose to be the major design change and should look nothing like Bulldozer. Most of the shared coomponents in Bulldozer are suppose to be removed for Steamroller and should offer approx about 45% performance improvement.

 

[HumanSmoke]

Steamroller is suppose to be the major design change and should look nothing like Bulldozer. Most of the shared coomponents in Bulldozer are suppose to be removed for Steamroller and should offer approx about 45% performance improvement.

Suppose is the operative word. AMD go from 10-15% improvement (which is more of a tweak than an overhaul), to Mark Papermaster shooting his gob off about 45% improvement. The only AMD released info regarding projected performance has already passed into the inconsistent

 

[sergionography]

Suppose is the operative word. AMD go from 10-15% improvement (which is more of a tweak than an overhaul), to Mark Papermaster shooting his gob off about 45% improvement. The only AMD released info regarding projected performance has already passed into the inconsistent
http://img.techpowerup.org/121106/bulldozertopiledrivertoexcavator fixed.jpg

who knows maybe it was scrapped for a new architecture, or was reevaluated and turns out its not enough to face haswell. Either or hopefully the delay is to bring something more competitive later

 

[sergionography]

Shouldnt this be with Steamroller Cores
What is AMD doing.

Steamroller is suppose to be the major design change and should look nothing like Bulldozer. Most of the shared coomponents in Bulldozer are suppose to be removed for Steamroller and should offer approx about 45% performance improvement.

more like 20%better ipc and 20% better multicore scaling which gives 40% better multithread performance, pretty good but still behind hasaell, 20%ipc puts steamroller in first Gen i7 territory core by core at same clock. u Need lik 30-35% ipc increase to match sb, let alone haswell. One has to mention that the clocking capability might play a role but then who knows haswell is said to be much more dynamic in its clocks and power envelope so amd Can't depend on that

 

[TheoneandonlyMrK]

Shouldnt this be with Steamroller Cores
What is AMD doing.

Steamroller is suppose to be the major design change and should look nothing like Bulldozer. Most of the shared coomponents in Bulldozer are suppose to be removed for Steamroller and should offer approx about 45% performance improvement.

They are actually makeing sense now imho as swapping the core, uncore and gpu all at the same time sounded insane to me, from a yield perspective.

 

[Super XP]

Suppose is the operative word. AMD go from 10-15% improvement (which is more of a tweak than an overhaul), to Mark Papermaster shooting his gob off about 45% improvement. The only AMD released info regarding projected performance has already passed into the inconsistent
http://img.techpowerup.org/121106/bulldozertopiledrivertoexcavator fixed.jpg

Wrong, that 45% performance improvement clock for clock came from an insider via AMD's tech department. Papermaster stated 30% via Steamroller. I say the insider is dead on right.

 

[trickson]

AMD dragging the industry down.

 

[eidairaman1]

AMD dragging the industry down.

no they are moving forward and just ramping things up in prep for SR in 2014 since 2012 is done

 

[Super XP]

AMD dragging the industry down.

AMD's Management has failed AMD's CPU Architects, you know those very smart individuals that keep getting their funding slashed.

On a side note, this may be a very good move for AMD, give them enough time to achieve the performance they are targeting for via Steamroller.
In the meantime a little birdie is telling me that we are going to see Piledriver 2.0 sometime in 2013 based on the leaked roadmap. In this case I can see a performance improvement of up to 10% to 15% over Piledriver 1.0 based on a combination of a process improvement and clock speed.