AMD is claiming that they can significantly cut platform power consumption. Considering the lower thermals (at least of the 25w TDP Turions) of their CPU vs. Intel's, having a competitive platform should make AMD's mobile lineup far more formidable.
Of course, unlike Intel, they have released only a reference platform, with specs for 3rd party chipsets and controllers... that leave quite a bit in the hands of Via, Nvidia, ATi, SiS, and so on - both in terms of performance, power consumption, and power management. No doubt that AMD will have 'qualifications' that 3rd party parts will have to meet, perhaps a 5w TDP max or something of that sort, but it will still allow for wild variation within the 'platform' - A problem that Intel (vertically monopolizing their platform) does not face.
More to the point, if AMD makes it's platform requirements serious enough to give Centrino a real run for it's money, there is the possibility that players like Nvidia (known not only to make the fastest, but also the hottest AMD chipsets) either will not be able to make the spec or may not be willing to invest the capital in developing products for the spec.
However, there are definately some nice features coming: DDR-2, for example, is less power hungry than high-speed DDR, so despite having higher latencies, I think it will be a real boon for AMD's mobile platform. This Turion-DC should be interesting as well, although it might be very interesting to see how high this product scales with 25w and 35w thermal envelopes. My guess is that this is going to be a serious problem until AMD is able to make a 65nm transition.
Since Yonah is 65nm already, my intuition is that AMD will be fighting a very uphill battle. The platform initiative idea is very sound (in fact some of us have been on AMD's case to do this for well over a year now), but given that it will depend a lot on 3rd party support, I'll be surprised if it's really effecting the market at all in Q1 of this year. In Q2 would be when I'd expect it to gain traction and take off.... except that Yonah should be shipping in volume by then, and likely at higher clockspeeds.
Calculating Turion DC (X2) TDP is pretty easy. Imagine a single core Turion, and double it. AMD has single core 25w Turion/1mb chips up to 2.2 GHz. That suggests to me that current 1.8 GHz Turions are probably in the 15-20w range, and could make the 35w cut off as X2s. I think that for a 25w MT chip AMD might have to scale back to 1.2 or 1.4 GHz.
But, we have heard rumors of some possible process tweaks and refinements, so squeazing an extra 200-400 Mhz out of each power envelope by the second quarter doesn't seem terribly impossible.
Of course, that wille barely keep pace with Yonah, and early in the 3rd quarter Meron is scheduled to appear - and unless it's a step backwards from Yonah, there would seem to be little reason to suspect that AMD would be able to counter it until their 65nm shrink, which is looking to be not until 2007. So while Intel might be trying to play catch up in the Server and Desktop markets, it looks like AMD wil spend 2006 trying to keep up with Intel's mobile offerings.
What impressed me the most was that I heard all of this will still be on AMD's mobile socket, NOT Socket M2. This came straight from an AMD rep's mouth when they demoed the platform in Japan a month or so back. Not only am I surprised that they can run 2 cores and dual channel memory on such few pins (638), it also makes me wonder what else will be in the new desktop socket that has the same basic configuration (X2, dual channel), but more pins. Could it be that so much of the current sockets are for power and ground? It leads me to believe that Socket M2 will accommodate more than just 2 cores in the future, but that's not really the point of this thread.
I for one and am dying to see how AMD answers yonah. Yomato looks promising, and I think the current Turion line set the table. I said it before, by AMD capping TDPs for their mobile CPUs, they opened the door to more OEMs and more chipset developers. It made the statement that AMD was committed to a low power mobile CPU, so now it's up to ATI, nVidia, and OEMs like HP to make a good notebook around these CPUs. Maybe Turions will never match the mobility of Intel, but perhaps they will get close enough and will be a few hunder dollars cheaper. I, for one, could live with that arrangement.
I swear i saw tomato...Interesting indeed, hopefully they'll be at each others throats forever, minute one slips away they tend to jack up prices, saw some opteron DC prices today, theres literally some 330 quid 600MHz cpus out there now (165 vs 180) Big disparity with the 175 and 180, 220 quid difference for 200MHz, cpu pricing REALLY annoys me.
R.I.P Rangeral, To one of HWC's best moderators and a great guy
By the way, what does BTW stand for?
It is better to be tried by 12, than carried by 6.
Yeah, the Mobile market remains a really good opportunity for AMD - since they still have so little of it. Also, because they DO have so little of it, Intel keep the P-M pricing very high. The difference between a base system with a 1.6 GHz Dothan in it and a 2.2 GHz Dothan is about $500!!
The one think Intel DOES have is the mobile Celeron - which sells at very low clockspeeds but costs very little. In a way it's sort of like Integrated Graphics - bare minimum but cheap. So far, while AMD's Turion is a compelling main stream option, it can't really compete with 1.3 GHz Celerons in the low end, and it's less integrated platform is a continuous sticking point for getting into high end systems.
I think Yamato could help. The name Yamato is a little curious, though. Using the name of the largest, most heavily armoured battleship ever constructed is cute... except that the US blew it up a bit off Okinawa in 1945.
And yes, that mushroom cloud is bigger than those produced by many nuclear weapons, but Yamato was very heavily armed.
It only took a destroyer sqadron and a few wings of dive bombers... once Yamato's own defensive air cover and destroyer screens had been knocked out. Even then, she took a pounding for hours and hours before that dive bomber droped one down a previously breached loading shaft. When the bomb went off in her forward magazine.... well... see the above!
Anyway, it was a sad day in naval history when the last of Japan's super battleships was dispatched, but it's not like the US navy could have afforded to let the thing be just for novelty value. It's 18" main guns could sink our Iowa class battleships before they could range a shot! Since we never actually built the Montana class, no ship ever floated that could repell 18" projectiles. In fact, Iowa couldn't even repell her own 16" rounds (Yamato could). More importantly, the Yamato's armour was so think that she prooved completely immune to torpedoes (which were really the most effective capital ship killers in WW2). Impressive ship. But the Japanese were pretty good with naval innovation (they just lacked oil and steel, or they would have crushed us!). They built the first (and to this day ONLY) submersible aircraft carriers!