Penryn...

All about them.

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
CoolColJ
Posts: 344
Joined: Thu Jul 10, 2003 11:58 pm
Location: Australia

Penryn...

Post by CoolColJ » Wed Sep 19, 2007 3:54 pm

I originally wanted to get a Q6600 after the price drop....
but good things come to those who wait - I want a Quad with SEE4 8)

although the desktop Penryns will be early next year, much more later than I expected! :?

So I think I may have to get a stop gap E2140 and overclock it or something to tide me over, because the p4 3.2ghz cpu is seriously lagging...
even that would give me a nice 3x speed boost

AMD may be planning tri-cores but until those will be released, arch-rival Intel will have its new, 45nm processor line-up already in stores. With 45nm production set to begin soon the desktop parts, that will follow the new Xeons is closing in fast and will be ready to give AMD a big slap even before the end of the year.

The first wave of 45nm desktop CPUs is set to include no less than 8 (eight) releases, all depicted below. First to arrive will be the QX9650, an Extreme Edition part featuring four cores clocked at 3 Ghz, 12MB of L2 cache (6MB per die) and a 130W TDP. This one will be launched in November and will hit the stores soon after. With 2007 gone and 2008 young and naive, Intel will have seven more CPUs, four dual-cores and three quad-cores ready. The highest-end non-Extreme quad-core is the Q9550, with 12MB of L2 cache, a frequency of 2.83GHz and a TDP of 95W.

The cheapest 45nm quad, the Q9300 will be clocked at 2.5GHz and will have only 6MB of L2 cache (3MB per die). As for the dual-cores, Intel will release the E8500, E8400, E8300 and E8200, clocked from 3.16Ghz to 2.66GHz and all with a 65W TDP. All processors (including the QX9650) will have a 1333 MHz FSB. Intel is probably saving the 1600FSB for later.

http://www.tcmagazine.com/comments.php? ... 94&catid=2

Image



Clock per clock, Penryns are faster, Intel said. It is unclear how much of this comes from the 50% larger L2 cache and how much comes from new architectures. Otellini, however, did indicate Penryns deliver 20% greater performance due to the 45nm hi-k/metal gate solution.

The big surprise of the day? Two live demonstrations of Nehalem in a physical machine. In the morning, we had a simple Nehalem system setup running Windows XP and a few applications. Otellini told us that Nehalems had been taped out for about a month, and what we saw was three week old "A0" silicon. He then held up a wafer showing wide, native quad-core Nehalems that were in use. Intel representatives made sure that they weren't using the AMD term “native quad-coreâ€

CoolColJ
Posts: 344
Joined: Thu Jul 10, 2003 11:58 pm
Location: Australia

Post by CoolColJ » Wed Sep 19, 2007 3:57 pm

quadcore on a single die means 8 core, 2x4 cores bolted together on a CPU is not far off 8)

smilingcrow
*Lifetime Patron*
Posts: 1809
Joined: Sat Apr 24, 2004 1:45 am
Location: At Home

Post by smilingcrow » Thu Sep 20, 2007 1:58 pm

CoolColJ wrote:quadcore on a single die means 8 core, 2x4 cores bolted together on a CPU is not far off 8)
Penryn is only a dual-core chip so the quads are already two chips on one die. AMD have the only quad-core chip and it’s possible that they will release an octo core (2 quads on one die) chip before Intel but not until they transition to 45nm. Intel’s quad core chip (Nehalem) isn’t due for about a year and probably even longer for the desktop.

The TDP for the Penryn desktop quads in the data you quoted is noticeably higher than data I’ve seen for the Xeon chips. The Xeon’s are supposedly rated at 80W up to 3GHz and the desktop chips are showing as being 95W up to 2.83GHz. Considering how similar the chips are architecturally this suggest that either the data is incorrect or that the designs or manufacturing processes differ significantly.

Post Reply