Fudzilla: Bulldozer performance figures are in

[bridito]

I take it that no real world performance indicators can yet be gleaned from this latest "leak"? Two more months of waiting in limbo...

 

[piesquared]

Who gets to make the call when hardware details are released, the hardware designer of the part or a bunch of demanding anonymous forum posters.

 

[LOL_Wut_Axel]

With power gating, the GPU's contribution to power usage when running purely CPU loads should be minimal. Even then, the power usage of a 2.9GHz Llano quad-core when running a purely CPU load is comparable, if not higher than a 2600K despite the 2600K having roughly 2x the throughput.

Yeah, and? It's using an architecture that's two years older. Again, this is comparing apples to oranges.

 

[RussianSensation]

AMD better release these by September, my upgrade itch is getting unbearable.

Core i5 @ 3.7ghz slowing you down? Outta boy! Keep the enthusiast bug alive! :thumbsup:

If that chart linked is correct, the performance difference between the 8100 and the the Q1 2012 8170 is going to be huge (2.8ghz --> 3.9ghz!!!). Could it be that lower 8xxx parts will be monster overclockers then? I have a feeling AMD is being overly conservative with the 2.8ghz clock speed. *Fingers crossed*. That might be the "sleeper" chip of the entire lineup.

 

[Idontcare]

That's not really an accurate assessment when it comes to comparing Llano CPU clock speed to Sandy Bridge CPU clock speed. AMD invested most of the die area and transistors into the GPU portion while Intel did the opposite. AMD was also able to get competitive power consumption. This is comparing apples to oranges.

I should certainly hope so considering the post of mine which you quoted has absolutely nothing to do with Sandy Bridge in the first place.

My point regarding Llano's power-consumption is that it is surprisingly high compared to the performance and the core clockspeeds involved, even when due consideration is given for the dGPU.

Idle results for the A8-3850 compared to the Core i3-2100 are pretty good with the AMD APU using less energy at this point. The load conditions aren't as good with the APU pulling down an additional 49 watts in CB10. Considering the APU is supposed to be a 100 watt part and the Core i3-2100 is a 65 watt CPU, those are larger than expected gaps.
And if you take a look at the AMD A8-3850 compared to the Sandy Bridge based Core i7-2600K, it uses nearly as much power and it does not have a discrete GPU installed while the 2600K does. Despite the poor x86 performance of the A8-3850 APU it seems to be using a lot of power while falling behind.

http://www.pcper.com/reviews/Graphi...r-Review-Can-AMD-compete-Sandy-Bridge/CPU-and-

Once all four of the A8's cores are occupied, though, we're reminded of its 100W TDP rating. Power draw on our Core i3-2100 test rig is 38W lower under load. Also, even though it's based on a 32-nm CPU at a lower clock speed, our A8-3850 system draws nearly as much power under load as our Phenom II X4 840 system. That suggests GlobalFoundries' 32-nm process may not be paying the dividends one would expect, at least not at these relatively high clock speeds.

http://techreport.com/articles.x/21208/14

 

[badb0y]

Core i5 @ 3.7ghz slowing you down? Outta boy! Keep the enthusiast bug alive! :thumbsup:

If that chart linked is correct, the performance difference between the 8100 and the the Q1 2012 8170 is going to be huge (2.8ghz --> 3.9ghz!!!). Could it be that lower 8xxx parts will be monster overclockers then? I have a feeling AMD is being overly conservative with the 2.8ghz clock speed. *Fingers crossed*. That might be the "sleeper" chip of the entire lineup.

Hehe, you know how it gets, there's nothing I can't play right now but still something within me wants me to throw money at my computer.

I am waiting for AMD because I am not in a rush to upgrade but if they fail I will be disappoint.

 

[bridito]

Who gets to make the call when hardware details are released, the hardware designer of the part or a bunch of demanding anonymous forum posters.

I'm pretty sure that the shareholders are getting fairly antsy too.

 

[BLaber]

Some Speculation as to September 19th is the launch date

http://wccftech.com/amd-bulldozer-launch-september-official-8-core-fx-processors-arrive-q1-2012/

 

[Mr Vain]

Some new stuff on AMD 2012 LINEUP

Quote
AMD's 2012 CPU lineup detailed

http://mb.zol.com.cn/240/2405453.html

10-core Komodo next year

http://www.fudzilla.com/processors/item/23525-amds-2012-cpu-lineup-detailed

Zol.com.cn scored a roadmap that reveals more details on AMD's 2012 desktop lineup and although some of the details have already shown up earlier, the new roadmap slides shows some new details regarding enthusiast Komodo CPU, mainstream Trinity APU and the low end Wichita APU.

The newest suprise comes from enthusiast desktop Corona platform with Komodo CPU and Hudson D4 chipset. According to the details in the slide, Komodo CPU will feature up to ten Piledriver cores (most likely updated Bulldozer design), will feature Turbo Core 3.0 tech, support for DDR3 memory and will be based on FM2 infrastructure. As far as the Hudson D4 FCH is concerned, this one comes with support for four USB 3.0 ports and eight SATA 6.0Gbps ports. The Corona platform will be rounded up with next gen. AMD discrete graphics.

The next in line is Virgo mainstream desktop platform based on the Trinity APU. The Trinity APU features up to four Piledriver CPU cores paired up with DirectX 11 GPU core. It is also based on FM2 infrastructure and has DDR3 support and support for AMD's Turbo Core 3.0. This one is paired up with AMD's A75 and A55 FCH with four USB 3.0 ports and six SATA 6Gbps ports and will also be rounded up with AMD's next gen. discrete graphics, probably with support for some sort of Dual Graphics that we've seen on Llano.

The last but not least, the Deccan platform based on the Wichita APU with up to four Bobcat CPU cores and DirectX 11 GPU core. It is based on FT2 BGA infrastructure and has support for DDR3 memory. It will be paired up with integrated Yuba FCH.

The entire lineup is scheduled for 2012. As a part of mainstream desktop lineup, Trinity will replace the current desktop Llano lineup and will manufactured on 32nm process, while Wichita will be based on TSMC's 28nm process.

 

[JimKiler]

Some new stuff on AMD 2012 LINEUP

Quote
AMD's 2012 CPU lineup detailed

http://mb.zol.com.cn/240/2405453.html

10-core Komodo next year

http://www.fudzilla.com/processors/item/23525-amds-2012-cpu-lineup-detailed

Seeing how this years chips will not be seen till almost Q4 or later next years will not be released for at least a year.

 

[Mr Vain]

Seeing how this years chips will not be seen till almost Q4 or later next years will not be released for at least a year.

Once the BD bugs are ironed out by September, SKU's should come on tap as per maps, that’s usually how it goes, Fingers +.

 

[podspi]

So, does this mean AM3+ is dead?! ALREADY?!

That can't be right...

 

[Ares1214]

So, does this mean AM3+ is dead?! ALREADY?!

That can't be right...

A new socket is almost always required for IGP, if any of these have IGP's then they would likely need a new socket. Same if there were large architectural changes.

 

[sm625]

I cant be certain but I am confident that amd will offer AM3+ parts based on the latest core design even after FM2 releases. I would be more confident if they actually released a 45nm socket 939 upgrade special. Of course there's no money in that is there? If they did though I'd buy at least 4 of those chips.

 

[bryanW1995]

Every day of additional delay makes this feel more and more like the barcelona release. Only difference is that back then everybody assumed that AMD would have an answer for core2, while nowadays expectations are so low that any kind of release at all will probably be greeted with trumpets/streamers/cats and dogs dancing together/etc.

 

[RussianSensation]

I take it that no real world performance indicators can yet be gleaned from this latest "leak"? Two more months of waiting in limbo...

Well AMD just released this Ruby / FX ninjas video pawning some dudes in blue suits. That has got to count for something.

 

[Cogman]

A new socket is almost always required for IGP, if any of these have IGP's then they would likely need a new socket. Same if there were large architectural changes.

Sad but true. Long gone are the days of Socket A when we went for something like 5 or 6 generations of CPUs all using the same socket. Those were good days. (though, in hindsight you usually had to upgrade your motherboard anyways because most MB manufactures wouldn't include new BIOS firmware for the next generation. And even if they did, they usually wouldn't continue the trend a second generation.).

 

[BLaber]

So, does this mean AM3+ is dead?! ALREADY?!

That can't be right...

Its dead for good , so most of the future processors can share same sockets as was the case till AM3 socket was released but changed with FMx sockets for zacate & Liano

 

[badb0y]

Well AMD just released this Ruby / FX ninjas video pawning some dudes in blue suits. That has got to count for something.

LOL 0:15 on the bottom left:

ember 19th

September 19
November 19
December 19
???
Your cartoons cannot fool me AMD gimme the benchies NAO!

 

[bridito]

Every day of additional delay makes this feel more and more like the barcelona release. Only difference is that back then everybody assumed that AMD would have an answer for core2, while nowadays expectations are so low that any kind of release at all will probably be greeted with trumpets/streamers/cats and dogs dancing together/etc.

Whoowheeeeeeee! We're Number Two! We're Number Two! Let's Party!

I may be one of the very few who still hold out a glimmer of hope for an SB-shredding BD but I have to admit this total squashing of anything that even remotely resembles a reliable benchmark is really causing me to swiftly abandon expectations that there will be a BD in my next system.

 

[drizek]

Hey, they're kicking VIAs ass.

 

[drizek]

BTW, if you think the fanboys on this site are bad,

F*cking intel loser F*ck off if u do not know anything. This video is an hint to the date of release of the new chips losers. F*ck your eyes open and see 0:15. Losers. AMD FOR LIFE.

JakeTeddywithR34 3 hours ago

 

[LOL_Wut_Axel]

I should certainly hope so considering the post of mine which you quoted has absolutely nothing to do with Sandy Bridge in the first place.

My point regarding Llano's power-consumption is that it is surprisingly high compared to the performance and the core clockspeeds involved, even when due consideration is given for the dGPU.

http://www.pcper.com/reviews/Graphi...r-Review-Can-AMD-compete-Sandy-Bridge/CPU-and-

http://techreport.com/articles.x/21208/14

Again, your point?

It's using a two-year-old architecture.

 

[piesquared]

I'm pretty sure that the shareholders are getting fairly antsy too.

I'm pretty sure that the stock explosion of %19.5 on Friday indicates that investors are quite satisfied with AMD's execution and forecast, mainly due to Fusion's accelerated ramp. The future is fusion tagline is there for a reason, that's the direction AMD is heading and CPU cores (ie Bulldozer) is one part of the Fusion process. With the added bonus of them being server CPU's of course. Fusion is AMD's biggest threat to intel in many years.

Now if you meant "AMD shorts are getting nervous too", then I agree.

 

[AtenRa]

Again, your point?

It's using a two-year-old architecture.

As i have said before, i believe it has to do more about the smaller (Lgate) transistors AMD have used in Llano than the architecture.