well, at least they're giving us dates... and they're not that far away. If they're that close, that means all of the imminent problems were taken care of.
Ryan
Printable View
well, at least they're giving us dates... and they're not that far away. If they're that close, that means all of the imminent problems were taken care of.
Ryan
AMD Preps Partners for "Barcelona" Launch
http://www.dailytech.com/article.aspx?newsid=8185Quote:
AMD's next-generation launch date is still set for "summer 2007"
AMD is set to debut its third-generation Opteron server processor architecture later this quarter. The company publically announced its upcoming launch next month, but the company's latest processor guidance, released just this week, states otherwise.
Good they're giving us more info now :D
Ouch.
So, 1.7 to 2.0GHz DVT *samples* going out in august, with prod parts in Sept.
Then, in Q4, they hope to sample more in October, presumably launching late in Q4, along with Phenom FX:
68W 2.0GHz
95W 2.1-2.3GHz
120W 2.4-2.5GHz
No 2.6GHz until Q208 at the earliest.
Looking at the hoped-for samples for Q4, I see indications of process problems. Look how much the power bin needs to climb to get from 2.0GHz (68W) to 2.4GHz (120W).
Not good.
Harpertown will wipe the floor with these parts. Even Clovertown will. :)
What assumptions do you think I'm making?
We know clocks of both lineups, we know (from AMD) that the IPC of the K10 core is roughly that of 65nm C2D core, we know about AMD bandwidth advantages in multi-socket systems...
Of course Intel parts are going to require increased power with higher frequency. My point was to notice the very large power bin jump required for a very small frequency increase in the K10.
This assumption
It may happen, but you can't say that it will now, because we have no performance results of K10. You can't say the benchmarks from a few months ago, because they're old and may be inconclusive now.Quote:
Harpertown will wipe the floor with these parts. Even Clovertown will.
Small correction to the other guy. Q6600 has a TDP of 105W, not 120W and that's the old stepping the G0 is lower. Q67xx is 130W. AMD and Intel have always improved their steppings during each Core's life-span.
Small correction to the other guy. Q6600 has a TDP of 105W, not 120W and that's the old stepping the G0 is lower. Q67xx is 130W. AMD and Intel have always improved their steppings during each Core's life-span.
Indeed, G0 Q6600 is 95W as well as raising the "thermal specification" from 62.2C to 71C. But wrong, QX6700=130W but Q6700 is 95W http://processorfinder.intel.com/det...px?sSpec=SLACQ .
AMD said these will be the same or lower TDP as current X2 parts, so it look like faster parts simply can't be done within specs.
Somebody enlighten me please, why is it the server based cpu's are always launched first?:confused:
AMD Phenom Processor (Barcelona) Pre-Launch Update
http://www.techarp.com/showarticle.a...tno=425&pgno=0Quote:
As the official launch of the AMD quad-core Opteron (Barcelona) processors in August and the desktop Phenom processors later this year draw close, AMD is rapidly ramping up media coverage of the new Barcelona core architecture. You can count on more media coverage and updates on the new Barcelona-based processors over the coming weeks.
by 'media coverage' they mean benchmarks, right? :rolleyes:
Ryan
http://uk.theinquirer.net/?article=41256
"At the analyst day this morning, AMD demoed a 3Ghz Phenom running the new 'Tri-Fire multi-GPU' setup."
oh dear goodness I hope that's a K10, not a K8.
Ryan
Anandtech says it's quad core.
http://www.dailytech.com/article.aspx?newsid=8203
It's quad core,for all those intel fanboys :).
To AMD Fanboys, everyone doesn't blindly trust AMD.
1. Golden sample or Dual Core until proven otherwise.Quote:
An AMD engineer, speaking on terms of anonymity, claimed to DailyTech that the company took on (*1)several unplanned respins for the K10-family. "We ramped the spin aggressively. It's going to bring the clock frequencies we need for the desktop," he said.
The company would (*2)not comment on when the 3.0 GHz Phenom-based processor would debut, though the company has previously indicated (*3)Phenom will ship for volume by the 2007 holiday season.
2. Proof of number one. They must have tested tons to find a Quad one:)
3. People were flamed for saying AMD was having problems debugging.
That goes for lying folks at Intel as well. How can we forget the 4GHz Northwood demo, this is the same kind of schlock! I honestly hope it was Quad Core, but how can ANYONE prove it? :rofl: I say again, AMD could STOP most of this very easily. As it stands, AMD could have said it was Octo (8) Core some here would believe them.
Bashing strongly i see.What would we do without you guys,i wonder ...
thats enough..closed