finally! tray prices for 32nm quads!

Acanthus

Lifer
Aug 28, 2001
19,915
2
76
ostif.org
straight from the horses mouth.

E5620 (20x turbo multi)
http://ark.intel.com/Product.aspx?id=47925

E5630 (21x turbo multi)
http://ark.intel.com/Product.aspx?id=47924

E5640 (22x turbo multi)
http://ark.intel.com/Product.aspx?id=47923

E5645 (No tray price, hex core, 21x, this one fell under the radar for me until now!)
http://ark.intel.com/Product.aspx?id=48768

L5609 (14x multi, no turbo, no HT, 40w TDP)
http://ark.intel.com/Product.aspx?id=47928

L5630 (18x turbo multi, HT, 40w TDP)
http://ark.intel.com/Product.aspx?id=47927

L5640 (hex core, HT enabled, 60w TDP, 21x turbo multi)
http://ark.intel.com/Product.aspx?id=47926

X5650 (hex core, HT enabled, 95w TDP, 23x turbo multi, dual capable)
http://ark.intel.com/Product.aspx?id=47922

x5677 (quad core, HT enabled, 130w TDP, 28x turbo multi, dual capable)
http://ark.intel.com/Product.aspx?id=47929

x5680 (hex core, HT enabled, 130w TDP, 27x turbo multi, dual capable)
http://ark.intel.com/Product.aspx?id=47916

Street prices are usually 10% above tray.

Cliffs:
E5620 ~$426
E5630 ~$606
E5640 ~$861
E5645 ~?
L5609 ~$484
L5630 ~$605
L5640 ~$1100
X5650 ~$1096
X5677 ~$1830
X5680 ~$1830
 
Last edited:

aigomorla

CPU, Cases&Cooling Mod PC Gaming Mod Elite Member
Super Moderator
Sep 28, 2005
21,088
3,598
126
:O

let the crying begin.
 

cbn

Lifer
Mar 27, 2009
12,968
221
106
Well $387 doesn't seem like a really bad retail price. That is only $100 more than a Core i7 920.

Its the people that normally buy i7s at Fry's and Microcenter that will probably not like these tray prices.
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
230*20 = 4.6GHZ with HT, 12MB, other improvements, not that bad for under $500.

I was hoping for less, but I don't think it's that far off the mark for price. Especially since they have zero competition.
230? most reviews I look at for i7 cpus shows just over 200 usually being the max.
 

klocwerk

Senior member
Oct 23, 2003
680
0
76
and oh look, they're all 1366 parts.
Where are the naysayers who were preaching that 1366 was a deadend socket that wouldn't get the 32nm parts?

(Sorry, had to get that off my chest. Nyah nyah.)
 

cbn

Lifer
Mar 27, 2009
12,968
221
106
Last edited:

ilkhan

Golden Member
Jul 21, 2006
1,117
1
0
Once production ramps prices will probably drop. Strange they're not lower, since the production cost should be less than bloomfields with the smaller die sizes.
 

Ben90

Platinum Member
Jun 14, 2009
2,866
3
0
They aren't really smaller, theres more logic and a huge cache bump.

Anandtech said:
The coolest part of Gulftown is that by building it on Intel’s 32nm process it’s actually smaller than both Bloomfield and Lynnfield, despite having 50% more cores and L3 cache:


At 1.17 billion transistors, it’s a beefy chip but the monolithic die only measures 240mm^2. It’s even smaller than an AMD Phenom II X4. Not only does it have a smaller die than all quad-core Nehalem processors, but it also has the same TDP.


Gulftown 240mm2
Bloomfield 263mm2
Lynnfield 296mm2
Clarkdale 81mm2
Deneb 258mm2
 

aigomorla

CPU, Cases&Cooling Mod PC Gaming Mod Elite Member
Super Moderator
Sep 28, 2005
21,088
3,598
126
I can tell you guys who love benching, and other cpu intensive programs.

The 12MEG cache does help.

This is why i said a while ago, 32nm scales better then 45nm.
But be very careful with Vtt on these guys again, like how we were on yorkfield.

I know a lot of people who have pop'd there A0 gulftown's (me included) for unknown reasons.

But im starting to look heavily on Vtt causing it.
 

SHAQ

Senior member
Aug 5, 2002
738
0
76
It probably is vtt. From reading on various forums I saw that motherboard manufacturers have to add vtt in small increments on boot up to lessen the chances of it frying the CPU. That is one of the reasons the 1366 boot up time is so long.
 

Lemon law

Lifer
Nov 6, 2005
20,984
3
0
Grandma will be so thrilled in a few more years when the prices come down. New more expensive computers that will do basic email and light web surfing even faster than before.

Clearly what is needed is even more bloatware software, software able to slow the fastest CPU down to a crawl. Just add in a few do forever loops and terminate them five minutes later.
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
I can tell you guys who love benching, and other cpu intensive programs.

The 12MEG cache does help.

This is why i said a while ago, 32nm scales better then 45nm.
But be very careful with Vtt on these guys again, like how we were on yorkfield.

I know a lot of people who have pop'd there A0 gulftown's (me included) for unknown reasons.

But im starting to look heavily on Vtt causing it.
that cache can actually make it slower in some cases. http://anandtech.com/cpuchipsets/showdoc.aspx?i=3763&p=12

"Gaming performance is a mixed bag with the Core i7 980X. In some cases it'll be a bit faster than the 975 due to its larger L3 cache, but in some cases it'll be slower because of the L3 cache's higher latency."
 

lopri

Elite Member
Jul 27, 2002
13,315
690
126
Insane.

Edit: Didn't realize that these are Xeons.
 
Last edited:

jvroig

Platinum Member
Nov 4, 2009
2,394
1
81
No VT-d?? On a hexcore? Isn't Intel getting cheap? You would think that a hex-core customer should deserve at least all of the features that the design supports. BAAAAD Intel.
Since they did this on quad-core chips, it's not surprising at all. It must have worked well for them last round, so here they go again.