Richland & Kabini rumours

[ShintaiDK]

Originally Posted by Nemesis 1

Thats a laugh MS worked hand in hand with AMD on 64bit. AT least it clear it was AMD and MS that kept the X86 bagage alive , Even tho VLIW is the future.

Microsoft was forced to gamble on AMDs "hotfix" of x86. Imagine all legacy benefits gone. While still in an upper position, MS would essentially start from scratch.

Nobody gonna bet the house if they can avoid it.

x86 as such is also slowly being replaced inside out.

[Nemesis 1]

Originally Posted by NTMBK

Nah, I'm not ignoring GT3- I'm just pointing out that better integrated graphics were necessary. Intel clearly agree, based on the big improvements that they've been making. I'm looking forward to seeing how Haswell does, and I already predict that NVidia will be out of all next-gen Apple products.

There was a reason AMD bought ATI . It was because intel had already started working in this direction . Elbrus purchase was the sign of intels future direction and its almost fully implemented threw out intel compilers . Intel started work on using all transitors doing more with less . Soix is good at shouting down unused transitors , The real aim is to make those transitors useful so they don't have to shout down. Fewer transitors . Rather than more is better . Silvermont is already going with 8 core atoms for servers and that number will grow . Haswell at 7 watts will have alot of server use.
I hoping airmont fineally brings intels dream to relization

[piesquared]

Is that supposed to be 7W or 10W or 13W? Looks like intel's PR is working as planned. Was Mooly Eden hiding off screen somewhere during the demo? Surely you'll understand if people don't take intel's word for their claims.

[beginner99]

Originally Posted by NTMBK

Driving high res displays, for a start. Apparently even HD4000 graphics make for a jerky experience on the 13" Macbook Pro with Retina:



http://www.theverge.com/2012/11/1/35...display-review

And Vizio have explicitly said that they went with Hondo for their 1080p tablet because Clover Trail couldn't drive the display satisfactorily: http://www.theverge.com/2013/1/7/384...-powered-1080p

The iPad 4 already has a 2048×1536 screen. If Windows 8 tablets want to compete against that, they're going to need beefy graphics.

And this isn't even touching on gaming...

Ok good points. But I doubt we will see iPad like Resolution on Windows tablets. I mean we all know how great classic windows (desktop) deals with this...text would be unusable tiny.

[inf64]

Originally Posted by piesquared

Is that supposed to be 7W or 10W or 13W? Looks like intel's PR is working as planned. Was Mooly Eden hiding off screen somewhere during the demo? Surely you'll understand if people don't take intel's word for their claims.

He is probably driving the "dirt" formula-car behind the scenes .

[NTMBK]

Originally Posted by beginner99

Ok good points. But I doubt we will see iPad like Resolution on Windows tablets. I mean we all know how great classic windows (desktop) deals with this...text would be unusable tiny.

Actually, the DPI scaling works pretty well so long as applications are aware of it. Take a look at this lot (running Windows 8 on Retina Macbook Pro), notice how much the DPI scaling helps: http://www.anandtech.com/Gallery/Album/2080#1

[piesquared]

Originally Posted by inf64

He is probably driving the "dirt" formula-car behind the scenes .

lol they almost got away with it once, I wouldn't be surprised if they tried it again. intel probably used a new lacky this time though, the very sight of Mooly's hat would be a dead give away. Hope you traced those wires Anand!

[Nemesis 1]

Now your calling AT liar . Thats 3 honesty results from you in less than 1 hour

[Nemesis 1]

Originally Posted by piesquared

Is that supposed to be 7W or 10W or 13W? Looks like intel's PR is working as planned. Was Mooly Eden hiding off screen somewhere during the demo? Surely you'll understand if people don't take intel's word for their claims.

We have seen demo of haswell running with GT3 @8 watts but you keep spreading your lies

[piesquared]

Originally Posted by Nemesis 1

Now your calling AT liar . Thats 3 honesty results from you in less than 1 hour

Actually, no.

[Nemesis 1]

ya I seen what you did . actually yes you did . We at AT fully expect Ananda not to be decieved , You tried to undermine that . SO yes you did,

[MightyMalus]

Originally Posted by Nemesis 1

We have seen demo of haswell running with GT3 @8 watts but you keep spreading your lies

Erm, the Dirt 3 video? That is obviously not a "8watt" Haswell. That's the top of the line Haswell GT3 and eDRAM chip.

[Ventanni]

I'm with Mightymalus on this one. That Nvidia 650m is a 45w TDP part. Impressive as it is, there's no way that's an 8w Haswell chip.

[AtenRa]

Yeap, they used a desktop Haswell+eDRAM against a mobile GT650M and we havent even seen the frame numbers

[ShintaiDK]

Originally Posted by Ventanni

I'm with Mightymalus on this one. That Nvidia 650m is a 45w TDP part. Impressive as it is, there's no way that's an 8w Haswell chip.

The CPU used is most likely 37W or 47W.

[MightyMalus]

It wasn't a desktop Haswell, but it was the top end mobile haswell.

I mean, it looks nice and all. But it ain't no "8watt". The 8watt haswell shown in IDF 2012 was only comparable(honestly, it seemed alil slower) to the IVB 15watt.

[Vesku]

AMD at least has more of a fighting chance for design wins with Temash and Kabini. Intel will be reluctant to release Haswell based Pentiums like they have done against Llano and Trinity. Even so, it will be very important that they break the cycle of shipping delays they've been caught in.

I was unexpectedly impressed with the C-50/C-60 chips and it seems like these upcoming products might also impress.

[Idontcare]

Originally Posted by piesquared

Is that supposed to be 7W or 10W or 13W? Looks like intel's PR is working as planned. Was Mooly Eden hiding off screen somewhere during the demo? Surely you'll understand if people don't take intel's word for their claims.

Originally Posted by inf64

He is probably driving the "dirt" formula-car behind the scenes .

LOL, its true though. They shot their credibility in the foot on that one. I wouldn't trust a pre-packaged demo setup if I was asked to assess it. I hope Anand looked for woodscrews too

[AtenRa]

Originally Posted by MightyMalus

It wasn't a desktop Haswell, but it was the top end mobile haswell.

I mean, it looks nice and all. But it ain't no "8watt". The 8watt haswell shown in IDF 2012 was only comparable(honestly, it seemed alil slower) to the IVB 15watt.

They used a desktop tower = better thermals = higher performance.

[inf64]

Originally Posted by AtenRa

They used a desktop tower = better thermals = higher performance.

This too .
So to sum it up: solid performance(at unknown fps) on non-ULV part . Like others noticed probably 35+W model with GT3+eDRAM.

[Abwx]

Our Editor in Chief met with AMD at the show yesterday, and he was able to glean more details about the new APU—including some early performance data. A notebook running a Kabini prototype apparently scored 1019 in Musemage, an OpenCL-accelerated photo-editing application. For reference, I just got a 1770 score from an Asus UX31A ultrabook featuring Intel's Core i5-3317U processor. Geoff obtained a 1243 score from the Core i3-3217U-powered Asus VivoBook X202E

Now, Kabini will likely compete against cheaper—and lower-power—processors than the i3-3217U and i5-3317U. Kabini's 15W TDP applies to the whole system-on-a-chip, including the integrated I/O, while Intel's 17W Ivy Bridge processors require a separate platform hub. The HM67 Express PCH in our ultrabook has a 4.1W TDP, according to Intel's website, so the Intel solutions' total power requirements add up to 21.1W or so.

http://techreport.com/news/24186/new...15w-kabini-apu

[Nemesis 1]

Originally Posted by inf64

This too .
So to sum it up: solid performance(at unknown fps) on non-ULV part . Like others noticed probably 35+W model with GT3+eDRAM.

YOU don't know this at all . Its what you want to believe because it smashes your agenda , We don't know leave it at that . We don't know that AMD was showing was real it could have been anything . We don't know . But if I was a betting man I would say that was not 2core soc/soix haswell . But it really doesn't matter . Intel graphics below 10 watts will run just as well as AMDs . Given what we know intelis better at low wattage

[inf64]

Originally Posted by Abwx

http://techreport.com/news/24186/new...15w-kabini-apu

Thanks for the link. Very interesting detail about the PCH TDP there. So it's 17+4.1W total power for intel IB ULV parts.

[Nemesis 1]

Originally Posted by Abwx

http://techreport.com/news/24186/new...15w-kabini-apu

Exactly Thats why I haven't said one word about 7watt IVB . To me its a none event . But one hell of an accomplishment by intel . The real story the one being ignored because you can't do what you just did is haswell 2 core SoC /soix @7 watts . This 7 watt ivb isn't scarring anybody . Haswell most definit;ly is . or we wouldn't be talking about a short runn product intend to be a stop gap.Same as Hondo . IVB at 7 watts scares no one. hence the reason its talked about rather than haswell. From the link it didn't help your cause add in the 4 watts and its still under 15 watts not that it matters

[inf64]

BTW I've just ran that Musemage benchmark on my Radeon HD6870 and got 12300pts score . These ULV products from intel and amd still look like 486 level compared to desktop stuff(and my gfx card is getting old too).