The chip used in the initial video has a 45W TDP, but the system was drawing upwards of 65W when it was really getting maxed out. If you throw enough at it, the CPU is going to eat up power. There's no indication of how much of that was translated into waste heat, but the AMD chip was drawing less power.
If the GPU and CPU are both being heavily stressed it'll probably get close to its TDP.
That my point its a marketing ploy and AMD is exploiting a weakness in intel drivers . This won't be the only game this happens with . Intel simply needs time . Sad part is Intel can't go pick up a llano systen to do the same . My wife and daighter been working hard the last 2 days comparing Graphics quality between NV and AMD and I not real happy with AMDs driver team at all .
The display, hard drive, and other components are going to eat up 20 watts easily, which is probably why it's drawing 65 watts, not 45.
Now I don't know what Memory was used but the Core i7-2630QM does allow ECC memory. and the cost of these laptops is sky high so I suspect it was in fact ECC memory . So this notebook isn't intended for the masses at all.
Intel doesn't officially support 1600 on any of their lower end mobile chips...
http://ark.intel.com/Product.aspx?id=53463&processor=i7-2635QM&spec-codes=SR030
(as opposed to - http://ark.intel.com/Product.aspx?id=52227&processor=i7-2820QM&spec-codes=SR00U,SR012)
I'm somewhat disappointed in myself for not making the connection on another amusing point of this demonstration earlier.
How exactly did AMD put a FCPGA998 i7-2630QM chip into the LGA1155 socket of the Asus P8H67-M that they claim to have used?
Put this into google its takes ya to same page you had . Read what its says befor ya link .
i7-2630QM ECC support
I actually got the info elsewere . I looking for that link .
Jan 8, 2011 ... ECC, Fully Buffered ... ASUS G73SW-FHD-TZ016V - 17.3" FHD LED/i7-2630QM/8GB DDR3/1 [G73SW-FHD-TZ016V] ... Cairns Tech Support.
This one is only $2550
It says that the 2630QM doesn't support ECC memory. I'd trust that Intel provides the correct information about their products unless you have some other reason to believe that Intel's website is wrong.
To play devil's advocate, I wonder what a dual-core mobile i5 v. i7 test would look like? IoW, what are the chances that the visual results would be nearly identical, while consuming less power than Llano? It would still be an AMD win, in that case, but if using significantly less power, it wouldn't look quite so bad for Intel. Creating tests like this demo, but with more varied hardware, would be a good thing to try with a first review of Llano, as a nice reality check.
What does that have to do with anything. The notebook with the Intel chip looks nothing like the one you're talking about.
OK . I may have missed it were does it say it doesn't support Ecc memory.
Did ya miss the fully buffered ECC memory . DO a price check on systems using the CPU Its insane for a bottom end system . This chip is intended for laptops that use ECC.
Its the only reason to price that high. If the OEMS are going to charge that kind of money AMD owns that market with llano . Doesn't matter about performance. But I bet llano does not support ECC.
It would only make a difference of about 10 or 15W. So maybe Intel will use 50W while AMD uses 55. On the other hand, Llano would run circles around a dual core SB(although the high-clocked, aggressively turbod i7 duallies can beat out Intels i7 720qm).
Wheres the proof of that ? Amd selected a crippled embedded system using ECC memory . To compare with. Intel does support CL you have problems reading? and its graphics on 2D are just as good as AMds