Its hard to believe in their roadmap now. Anyway mass production of Kuma and Phenom must be in Q1 2008. By that time Intel will mass production of Penryn that might oc higher and they will be hard time for AMD to recover.

guess wat, the cheapest K10 quad core will be around RM1100 and this one is opteron, agena lowest end will be a lower than this.

click here

Those TDP figures don't look very good at all.

We people that using Core architecture have no reason to migrate for AMD if the performance is just on par. Wait until next battle. Nahelem and AMD's 45nm processor.

Penryn QC mobile will come into existence soon.

aiks more battle to come ahh? adeyy, so sad laa ..

I don't mind any as long they're relatively affordable for everyone

we are the 1's who benefit if they "battle" more often
but our wallet is the 1's who got the pierce and gunshot's from the "battles"

It is best to use our wallet is the 1 who's hurt the most

Still, can't wait for its official release. Wanna know its performance

is..this made me to wait more b4 change to c2q or phenom

*must resist the urge for selling of my current cpu and mobo...*

must resist the urge to buy it....and OC the crap out of it ........

but the tdp figures dont look so good, but it sure beats intels (intel also nid more power for mobo wut...so power wise..amd might bu using less coz no memory controller to pump watts into mar..)

and also, Intel measures TDP by measuring the average TDP, while AMD is measuring the max TDP. so, a 95W TDP intel C2D isnt same as a 95W TDP AMD k8.

and by the way, AMD got memory controller lar , it is integrated with the processor itself. its TDP also contributed from that memory controller.

ehehe, sorry, what i meant was an separate controller/chip and not on cpu, all intel boards got 1 north and 1 south bridge mar...both are chips, chips use power..so figure the power usage =b

but why amd want to measure the[b]max[/] wor...if they measure average, they will definitely beat intel =/

im nt sure about that, but it is not "why AMD want to measure max", u shuld ask, "why intel dun measure max". IMO, intel dun wan measure max coz that figure is nt very "marketable".

Anyway, most review sites usually using overall system power consumption to measure power usage. For now performance/watt goes to Intel.

Intel's reasoning is your processor won't be running at 100% all the time. If it is then the program you're running is either over-taxing your proc or is very poorly written. So they measure TDP as around 80% of the max (need to verify this).

TDP is not exactly a good way to measure heat generated by a proc. TDP is more of a yardstick for engineers to know how much heat can be safely dissipated by a cooling solution.

i.e. target TDP is 20W + other variables (junction temp etc)
Based on some formula they need to use certain heatsink material (alu or cu) and certain dimension.

This whole TDP as defined by Intel/AMD is just a PR gimmick.

That used to be the way they measured. I'm pretty sure that AMD switched to intel's method a while ago.

Yup, right..... "marketable". It's all really marketing stuff this "TDP". Its better to get a 3rd party view and results on this TDP like.... AMD X2 6000+ "125W TDP" vs Intel Core 2 Quad QX6700 "130W" at AMD HotHardware: Athlon 64 X2 6000+ - Power consumption... or AMD FX-62 "125W TDP" vs Intel QX6700 "130W TDP" at The Tech Report - Page 16: Intel's Core 2 Extreme QX6700 processor.

I think the TDP issue is best judged by comparing 3rd party reviewes. The choice of motherboard, chipset or RAM also affects overall "TDP". Anyway Barcelona looks late, and I'm not surprised either (from the Computex "fiasco"). It may go head to head with faster Penryn by that time.

This post has been edited by lex: Jun 28 2007, 11:22 AM

guess what, it was confirm, Barcelona to kick a$$ in august benchmark will appear, and hopefully within this few weeks.

Barcelona performance number may not be assumed as same as Phenom coz its using HT 1.0. Also phenom will receive new stepping.

it is HT2.0.