Engineering Sample Of AMD Steamroller Based APU Spotted

[Over_Lord]

Hardware news site WCCF Tech spotted an interesting entry listed in the Bionic research database. The ES (Engineering Sample) chip could be a part of AMD's next-generation APU series featuring the new and improved Steamroller core. While we don't expect performance to increase by leaps and bounds, but Steamroller builds on the Bulldozer architecture and has a target to offer as much as a 30% improvement in performance over the original core.

The ES code 2M186092H4467_23/18/12/05_1304 tells us even more. According to earlier observations (here and here), the four numbers in the middle part tell a bit about clock speeds. If the first one is not 00 (no turbo, see Kabini ES), it indicates a turbo clock of 2.3GHz. The "18? stands for 1.8GHz nominal frequency. I'm not so sure about the "12?. It could stand for 1.2Ghz North Bridge clock. Finally the "05? indicates a 500MHz GPU clock. The right part "1304? is the GPU code, which - thanks to earlier revelations - can be identified as AMD1304.1 = "KV SPECTRE MOBILE 35W (1304)".

A 2.3 GHz Turbo core is pretty low, which can be attributed to the early state of the Engineering Sample. Hopefully clock speeds hit further north of just 1.8 GHz CPU and 500 MHz graphics, especially for the 35W part. The next-generation chips will be manufactured on the new bulk 28nm manufacturing process at Global Foundries.

View at TechPowerUp Main Site

 

[erocker]

As far as the poll goes, the answer I can give is "wait and see". AMD hasn't impressed recently in single core performance. When it comes down to it, that's what people want. More cores? Sure, people like that too.

 

[Nordic]

I don't like the poll options. Its not enough of an increase to be a game changer. I do think amd can one day, probably not soon, pull off something big. I am neither a fan or a non fan of amd or intel.

The only thing I get from this is steamroller is right around the corner... I hope.

 

[Over_Lord]

I don't like the poll options. Its not enough of an increase to be a game changer. I do think amd can one day, probably not soon, pull off something big. I am neither a fan or a non fan of amd or intel.

The only thing I get from this is steamroller is right around the corner... I hope.

I think Steamroller will hit us at the end of 2013, or perhaps Q1 2014.

I can't edit the poll options, I would have made some changes otherwise.

 

[BrooksyX]

I really wish that I had bought some AMD stock about 6 months ago when it was selling for $2, could have easily doubled my money by now. I really considered it. Oh well.

Hope it does well.

 

[btarunr]

AMD humbly admitted that each [construction equipment name] will bring up to +10% performance over its predecessor. That should put Steamroller up to 21% faster than Bulldozer.

 

[TRWOV]

Steamroller should catch up with Bloomfield in terms of IPC if it maintains the 10-15% improvement from Bulldozer to Piledriver. If it goes with >4Ghz out of the box and a 125w TDP it could hang out with some i7s with the only drawback being power consumption.

Is this still a 32nm part?

edit: It's 28nm. We could see a 4Ghz 95w SKU *crosses fingers* Heck, make it a 10 core 125w SKU and I'll buy 3.

 

[RCoon]

Even if the performance increase is only 5-10%, that still makes it a better possible chip than the current Piledriver CPU's which are more than capable of running along with top end hardware. It will obviously be priced competatively in comparison to Intel's offering (I.e. pretty cheap) because of single core performance, which is a winner for those on AM3+ who want an upgrade if they're still on PII or the original Bulldozer.
Sure, AMD are lagging behind seriously, but they still produce high end chips for high end systems. Any improvement on the CPU front is a win for PC user, both from Intel and AMD.

 

[Mussels]

if its performance is 30% better yet still maintains 35W TDP, its not a game changer - but its certainly good news.

 

[Fourstaff]

Interesting they are using 28nm rather than 22nm, given that the previous one was 32nm.

 

[de.das.dude]

i dont like this news editor. he is an obvious amd hater.

 

[Ghost]

i dont like this news editor. he is an obvious amd hater.

He is an obvious AMD lover. Just like 70% of TPU, according to that poll.

 

[Over_Lord]

He is an obvious AMD lover. Just like 70% of TPU, according to that poll.

I'm neither a hater or lover

But I'd sure love some power-packed products from the company to fight Intel

 

[TheoneandonlyMrK]

Poll could do with a wait and see to add to the mega neg and double pos replies.
No Info on imc and nb damn

 

[cdawall]

Interesting they are using 28nm rather than 22nm, given that the previous one was 32nm.

AMD has been a lithography behind for some while now, but have proven that they can stuff more onto each lithography than intel can, something that was admitted by even intel in the phenom I days.

 

[Fourstaff]

AMD has been a lithography behind for some while now, but have proven that they can stuff more onto each lithography than intel can, something that was admitted by even intel in the phenom I days.

Behind or not I am just surprised that they are choosing to shrink it half-step rather than whole step at a time like how they used to. Perhaps power consumption is forcing their hand.

 

[cdawall]

Behind or not I am just surprised that they are choosing to shrink it half-step rather than whole step at a time like how they used to. Perhaps power consumption is forcing their hand.

or global foundries doesn't have a usable 22nm lithography for them.

 

[TheinsanegamerN]

Behind or not I am just surprised that they are choosing to shrink it half-step rather than whole step at a time like how they used to. Perhaps power consumption is forcing their hand.

I think this is on TSMC's 28nm process. and TSMC's 20nm isnt ready yet. that might have something to do with it.

 

[TheoneandonlyMrK]

Behind or not I am just surprised that they are choosing to shrink it half-step rather than whole step at a time like how they used to. Perhaps power consumption is forcing their hand.

28nm is stable as a node 20nm is still a bit off so they are probably going for low end high volume devices for mobile first with low clocking 28nm bulk process parts ,with a mind to transfer a tested core into 20nm high performance nodes later.

 

[TheinsanegamerN]

28nm is stable as a node 20nm is still a bit off so they are probably going for low end high volume devices for mobile first with low clocking 28nm bulk process parts ,with a mind to transfer a tested core into 20nm high performance nodes later.

"20nm high performance nodes" kaveri-hint hint

 

[brandonwh64]

With the APU's I wished they would add the extra L3 cache of the higher end desktop CPU's to compete with intel more on the CPU side.

 

[Fourstaff]

I think this is on TSMC's 28nm process. and TSMC's 20nm isnt ready yet. that might have something to do with it.

No this is on GloFo's 28nm process (last sentence), which made me wonder. TSMC has a stable 28nm though, so it wouldn't be a surprise if they used TSMC instead.

 

[TheoneandonlyMrK]

"20nm high performance nodes" kaveri-hint hint

Imho when 20nm kicks in well see steamier 8 cored parts but not the fx variety im after, initially more high end apu types

 

[HalfAHertz]

Kaveri was supposed to be here my Q2 2013. It's another delayed part in a series of delayed products. Hopefully this time the wait will be worth it. We badly need AMD to start performing better because it is quite evident that competition is lacking and progress has been slowing down.

 

[acerace]

i dont like this news editor. he is an obvious amd hater.

No, you just love AMD too much. Are you a fanboy? Only you can answer that.