[fehu]

i remember that
rambus saved amd that time :S

[ronvalencia]

Originally Posted by Paran

He said 50% more compute capability at almost half the power. Think twice about it. AMD announced on CES 50% more compute power for the 35W Trinity. 17W and +50% power doesn't match AMDs claim. You can be sure it is a marketing trick.

Something more serious:
http://www.youtube.com/watch?v=agJxehoSBmY

He clearly stated the demo system used a Mainstream APU. Ultra Thin APU was separately mentioned.

Refer to http://www.youtube.com/watch?feature...&v=lsmTDb-Mlws

[Albuquerque]

Originally Posted by ronvalencia

Refer to http://www.youtube.com/watch?feature...&v=lsmTDb-Mlws

Yes, we have seen this video. He never says anything about 17 watts, he only says it's coming from a laptop. A separate video was presented, at the same booth, where the AMD person specifically mentions the laptop is using the mainstream Trinity part, which is not 17W.

There have been a lot of people claiming 17W because of the video you posted, but that video never mentions 17W. I assume that various sites have latched onto the 17W "slide" that was presented, and made the conclusion that they were the same parts. Unfortunately, while a lot of sites claimed 17W, that doesn't make it correct.

Again: AMD"s own spokesperson called it a mainstream part, and mainstream parts aren't 17W -- those are ULV parts. Case closed.

[ronvalencia]

Originally Posted by Albuquerque

Yes, we have seen this video. He never says anything about 17 watts, he only says it's coming from a laptop. A separate video was presented, at the same booth, where the AMD person specifically mentions the laptop is using the mainstream Trinity part, which is not 17W.

There have been a lot of people claiming 17W because of the video you posted, but that video never mentions 17W. I assume that various sites have latched onto the 17W "slide" that was presented, and made the conclusion that they were the same parts. Unfortunately, while a lot of sites claimed 17W, that doesn't make it correct.

Again: AMD"s own spokesperson called it a mainstream part, and mainstream parts aren't 17W -- those are ULV parts. Case closed.

In time stamp: 1:06, He talks about "thin" notebooks and He opens the desktop PC case and shows the notebook i.e. "what we are showing you" section. He then claims claims it fits into "ultra thin designs".

[ronvalencia]

Originally Posted by Albuquerque

Yes, we have seen this video. He never says anything about 17 watts, he only says it's coming from a laptop. A separate video was presented, at the same booth, where the AMD person specifically mentions the laptop is using the mainstream Trinity part, which is not 17W.

There have been a lot of people claiming 17W because of the video you posted, but that video never mentions 17W. I assume that various sites have latched onto the 17W "slide" that was presented, and made the conclusion that they were the same parts. Unfortunately, while a lot of sites claimed 17W, that doesn't make it correct.

Again: AMD"s own spokesperson called it a mainstream part, and mainstream parts aren't 17W -- those are ULV parts. Case closed.

Notice "mainstream ultrathin segment" point for Trinity ULV.

[Alexko]

Guys, AMD's Analyst Day is in 3 days, just drop this for about ~80 hours and I'm sure you'll have much more precise information to bicker about.

[Albuquerque]

Originally Posted by ronvalencia

In time stamp: 1:06, He talks about "thin" notebooks and He opens the desktop PC case and shows the notebook i.e. "what we are showing you" section. He then claims claims it fits into "ultra thin designs".

To which, later, it is referred to as a mainstream part and not an ultrathin. And as evidenced by the chassis itself, isn't an ultrathin. The Trinity ULV parts are not the Trinity Mainstream parts; you are confusing Trinity naming schemes (mainstream higher wattage vs ULV lower wattage) with laptop chassis form factor naming schemes (mainstream chassis sizes versus ultrathin chassis sizes.)

I'm fine with waiting If I'm wrong, so be it, but I doubt it...

[ronvalencia]

Originally Posted by Albuquerque

To which, later, it is referred to as a mainstream part and not an ultrathin. And as evidenced by the chassis itself, isn't an ultrathin. The Trinity ULV parts are not the Trinity Mainstream parts; you are confusing Trinity naming schemes (mainstream higher wattage vs ULV lower wattage) with laptop chassis form factor naming schemes (mainstream chassis sizes versus ultrathin chassis sizes.)

I'm fine with waiting If I'm wrong, so be it, but I doubt it...

AMD assigned "mainstream ultrathin segment" on thier Trinity ULV slide.

I'm also fine with waiting.

[ToTTenTranz]

Originally Posted by Albuquerque

To which, later, it is referred to as a mainstream part and not an ultrathin.

Wow.. I mean, really?
Even though there's a sentence saying "mainstream ultrathin" in the official slide, you still insist it's either one or the other?

Originally Posted by Albuquerque

And as evidenced by the chassis itself, isn't an ultrathin.

And what size is the chassis, actually?
Looking at the keyboard size, I know it's not more than 13", but nonetheless I'm eager to watch say it's a 17" laptop and very thick, lol.

Originally Posted by Albuquerque

The Trinity ULV parts are not the Trinity Mainstream parts; you are confusing Trinity naming schemes (mainstream higher wattage vs ULV lower wattage) with laptop chassis form factor naming schemes (mainstream chassis sizes versus ultrathin chassis sizes.)

Oh, now you think "mainstream" is a name that defines power consumption.
That's wrong.


Mainstream is a name that defines target audience. In this case, mainstream means it'll reach the majority of the possible audience, so it's actually a pricing definition, and not power consumption or size or color or gender or whatever you come up with next...
As anyone (except you?) could see in pretty much every presentation/interview/press release/report/etc. from AMD since CES, Trinity is going to be an APU for mainstream systems.
There's no "high-end" Trinity down the line (that's for Piledriver FX CPUs), and there's no "low-end" either (that's for Brazos 2.0). There's only Trinity, a mainstream APU that can be put into desktops, laptops and ultrabooks/ultrathins (in its ULV BGA form).

Originally Posted by Albuquerque

I'm fine with waiting If I'm wrong, so be it (...)

That'll be fun.

[rpg.314]

Originally Posted by Alexko

Guys, AMD's Analyst Day is in 3 days, just drop this for about ~80 hours and I'm sure you'll have much more precise information to bicker about.

Where is the fun in bickering when you have precise information?

[mboeller]

Originally Posted by Alexko

Guys, AMD's Analyst Day is in 3 days, just drop this for about ~80 hours and I'm sure you'll have much more precise information to bicker about.

So you promise that in 80h this kindergarden is closed?

[Albuquerque]

Ok, I have posted several times the irrifutable proof that the CES demo was not on an "ultrathin" CPU, ie NOT the ULV BGA package, which absolutely means it wasn't 17W. And it's direct from AMD's mouth. And it's on video, so there's no room for error.

Since multiple people in this thread seemingly missed the video (or just want to pretend it doesn't exist), I'm going to link directly to the moment at which it is absolutely defined: http://www.youtube.com/watch?v=agJxehoSBmY&t=2m14s

Originally Posted by John Taylor, AMD

This is, of course, the Trinity APU for a mainstream notebook like you see here (gesturing at the laptop inside the desktop case.) We'll also be introducing this technology with all those capabilities for premium ultrathins. So you see here, this is uh, what's called a BGA package (holding up a separate chip example) for allow for very thin, z-high very thin form factors yet all of these incredible capabilities come along for the ride.

He directly indicates the laptop on display is not an ultrathin, he directly indicates it is not using the BGA packaged CPU, he directly indicates that all t he power of the (standardly sized) notebook will be coming to ultrathins with later Trinity devices.

I will not reply ever again to this thread, as I have utterly proven my point beyond any question or shadow of a doubt.

[DarthShader]

Originally Posted by Albuquerque

Ok, I have posted several times the irrifutable proof that the CES demo was not on an "ultrathin" CPU, ie NOT the ULV BGA package, which absolutely means it wasn't 17W.

But why is it such a big deal to fight over it for two pages? :S

Llanos can play Dirt3 on low settings and 768p at close to 60fps (source: notebookcheck). So if the ULV Trinity has similar performance levels and seperate circuits for decode and encode, then it isn't impossible that 17W chip could indeed perform such demo, even if it didn't on said video. So what's the big deal? :S

[ToTTenTranz]

Originally Posted by DarthShader

So if the ULV Trinity has similar performance levels and seperate circuits for decode and encode, then it isn't impossible that 17W chip could indeed perform such demo, even if it didn't on said video. So what's the big deal? :S

Well, there's this score..

[jimbo75]

Perhaps we should look at something a bit more relevant.

http://techreport.com/articles.x/20401/5

The e350 is an 18W cpu according to AMD and the d525 is a 13W cpu according to intel. One of them is either undervaluing or overvaluing their actual TDP. By a lot.

[Kaarlisk]

The newest leaked Intel IGP driver (on station-drivers) only supports Ivy Bridge and Sandy Bridge. Basically, for some time already, it seems that Intel supports only the two newest generations of IGPs. The older IGPs may get updates, but only as bugfixes for older branches, so forget about any new features or performance increases if you do not have a new Intel IGP. This is quite sad considering that large parts of the basic architecture have been the same since GMA X3000 (G965) - and Linux support for older IGPs does continue.

[DavidC]

Originally Posted by jimbo75

Perhaps we should look at something a bit more relevant.

http://techreport.com/articles.x/20401/5

The e350 is an 18W cpu according to AMD and the d525 is a 13W cpu according to intel. One of them is either undervaluing or overvaluing their actual TDP. By a lot.

Neither is overvaluing the TDP. Look at the difference between idle and peak. D525 needs extra 4W at load and the E350 requires about 9W, both significantly under TDP values. Actually, the ratio of undervaluing is about the same for both.

[jimbo75]

Originally Posted by DavidC

Neither is overvaluing the TDP. Look at the difference between idle and peak. D525 needs extra 4W at load and the E350 requires about 9W, both significantly under TDP values. Actually, the ratio of undervaluing is about the same for both.

Yes but it's total system draw.

If the rest of the system was 5W, the e350 would be 9W idle and 18W load while the Atom would be 17W idle and 21W load.

If the rest of the system was 10W, the e350 would be 4W idle and 13W load, while the Atom would be 12W idle and 16W load.

Either the e350 draws a lot less than 18W, or the Atom draws a lot more than 13W. It could just have been a particularly bad motherboard on the Atom I guess, but I don't think so - even xbitlabs often shows Brazos having lower draw than the Atom. I think intel is being somewhat generous to themselves on this one.

[ToTTenTranz]

Well, Arctic just "announced" the Trinity A8 and A10 APUs:

http://vr-zone.com/articles/arctic-a...pus/14712.html

[Paran]

Originally Posted by ronvalencia

Refer to http://www.youtube.com/watch?feature...&v=lsmTDb-Mlws

I don't trust a person who lied to the press. He lied on the part 50% more compute capability at almost half the power. If you trust a liar you are fine.

[DavidC]

Either the e350 draws a lot less than 18W, or the Atom draws a lot more than 13W. It could just have been a particularly bad motherboard on the Atom I guess,

You can spin it anyway you want, comparing load vs idle is way more accurate than anything else as CPUs usually dominate power usage at load.

It's not the motherboard, but due to lack of power management on the D-series Atoms. While the N-series chips go to C4 state, D-series stop at C1. The E-350 is the same E-350 in Netbooks, that feature full C states and power/clock gating. The Atom probably uses few W at idle , but E-350 has higher TDP, balancing it out at higher load usages when looking at in terms of system power.

[AnarchX]

Lets announce the 2013 APU fight:

Haswell
20? EUs GMA Gen7.5 @ >1,3GHz

VS

Kaveri
512SPs GCN @ >900MHz

Originally Posted by APPENDIX A

Testing performed by AMD Performance Labs. Calculated compute performance or Theoretical Maximum GFLOPS score for 2013 Kaveri (4C, 8CU) 100w APU, use standard formula of (CPU Cores x freq x 8 FLOPS) + (GPU Cores x freq x 2 FLOPS). The calculated GFLOPS for the 2013 Kaveri (4C, 8CU) 100w APU was 1050. GFLOPs scores for 2011 A-Series “Llano” was 580 and the 2013 A-Series “Trinity” was 819. Scores rounded to the nearest whole number.

http://phx.corporate-ir.net/External...F8VHlwZT0z&t=1

[tunafish]

Originally Posted by AnarchX

Lets announce the 2013 APU fight:

Haswell
20? EUs GMA Gen7.5 @ >1,3GHz

VS

Kaveri
512SPs GCN @ >900MHz

Weren't there some rumors around that Intel would use TSV:s to provide a high-bandwidth frame buffer for Haswell IGP? That would matter more than the GFlops I think.

[mboeller]

what? No discussion about 17W Trinity?



look at productivity!

[Tim]

Originally Posted by jimbo75

Yes but it's total system draw.

If the rest of the system was 5W, the e350 would be 9W idle and 18W load while the Atom would be 17W idle and 21W load.

If the rest of the system was 10W, the e350 would be 4W idle and 13W load, while the Atom would be 12W idle and 16W load.

Either the e350 draws a lot less than 18W, or the Atom draws a lot more than 13W. It could just have been a particularly bad motherboard on the Atom I guess, but I don't think so - even xbitlabs often shows Brazos having lower draw than the Atom. I think intel is being somewhat generous to themselves on this one.

The GPU-part is idle in Cinebench and the E-350 has no turbo mode to take advantage of this, the E-350 is running way under the TDP.

The D525 "GPU" is so slow that it won't use much more power in a game than just sitting in Windows, thus the D525 should be running near TDP in Cinebench.