Official Desktop Penryn Discussion Thread

[Shintai]

Theinq is seriously being more retarded than fudzilla now, and thats some of a record by itself.

Theo cant be taken serious...ever.

Anyway, let me asure people there will be atleast a 3Ghz Wolfdale and higher. And there will be 2.66 and 3Ghz Yorkfields too.

[ediflorianus]

i hate it .... Xtreme only 8 multiplyer... I want an UNLOKED CPU... i'l pay well.

[Donnie27]

i hate it .... Xtreme only 8 multiplyer... I want an UNLOKED CPU... i'l pay well.

Those are test processors, geesh! E6600 was sent out last time. Intel showed a 6700 before the 6800, so whoopty do "All four CPUs are listed as samples and MSI hasn't posted any information with regards to which boards these CPUs works with, as it only states "under testing". Now surely folks don't think speeds are this limited, do they?

[safan80]

http://theinquirer.net/?article=41468

here's the text

Motherboard makers fiddle with Penryn parts

By Theo Valich: Friday 03 August 2007, 17:59
AS THE TIME of launch for new processors approaches, more and more interesting information starts to leak from Taiwan. Thanks to our friends at Au-ja.de, we hear that motherboard vendors are now testing 45nm Penryn processors.

These tests are regularly done to prepare BIOS updates for motherboards that are already present, and to discover possible incompatabities - something that makers of Socket 775 motherboards are very well aware of.

We hear that motherboard makers are in possession of four different processors of the Penryn generation: two Wolfdales and two Yorkfields.

Wolfdale is a dual-core, single-die processor core with integrated 6MB of L2 cache, and the clock-speeds in play are 2.33 and 2.66GHz. Since these rely on the 1.33GHz FSB,the multiplier is set at 7x333 for the 2.33GHz part and 8x333 for the 2.66GHz one.

Yorkfield is simply waht Conroe was to Kentsfield: two Wolfdale dies are placed onto single Socket 775 organic package - so the amount of L2 cache is doubled to 12MB. And, right now, two things are being tested, the 2.33GHz and 3.33GHz models, with respective multipliers. The 2.33GHz part uses 7x333, while the 3.33GHz Extreme one works at 10x333.

As it stands right now, it looks like the difference between the regular version and the Extremes will be more pronounced. Unlike the original Conroe and Kentsfield with a difference of a mediocre 266MHz, Yorkfield EE is clocked at 3.33GHz with 12MB of L2 cache, while the fastest regular Yorkfield is set at 2.33GHz, a clear 1GHz difference in clock.

That should make up for the difference in price. µ

[Shadowmage]

Those are test processors, geesh! E6600 was sent out last time. Intel showed a 6700 before the 6800, so whoopty do "All four CPUs are listed as samples and MSI hasn't posted any information with regards to which boards these CPUs works with, as it only states "under testing". Now surely folks don't think speeds are this limited, do they?

It's definitely a good indication of current CPU health. Let's hope upcoming steppings solve the problem.

This is also proof that the 3.33ghz demos were cherry picked. If they weren't, then the 3.33ghz ES samples would be much more prevalent.

[Donnie27]

It's definitely a good indication of current CPU health. Let's hope upcoming steppings solve the problem.

This is also proof that the 3.33ghz demos were cherry picked. If they weren't, then the 3.33ghz ES samples would be much more prevalent.

TOTALLY Disagree. Like with the hidden Barcelona tests, let's just wait and see Now don't be surprised if Intel launches higher end models for more money. Then ship slower ones later. Again, these are test processors just like the 2.2 and 2.4GHz Conroes ES that they sent out.

[Obscured]

There is something more interersting than maximum Penryn clock. I'm talking about minium clock of the cheapest Penryn. After all, prices on Penryns will depend on it in particular.

[savantu]

It's definitely a good indication of current CPU health. Let's hope upcoming steppings solve the problem.

This is also proof that the 3.33ghz demos were cherry picked. If they weren't, then the 3.33ghz ES samples would be much more prevalent.

Please do tell us how you managed to come by this conclusion.

Can I remind you that current C2Ds on a 65nm process manage to reach 3GHz.Based solely on the shrink to 45nm , that brings you 20% more headroom. ( in other words Intel can make a 3.6GHz QC with as much effort as it takes to make a QX6850 ).Add high-k and metal to the mix and you can bring another few %.

[dexman]

Can I remind you that current C2Ds on a 65nm process manage to reach 3GHz.Based solely on the shrink to 45nm, that brings you 20% more headroom.

And please tell us how you came to that conclusion. It's not just math, there are more factors that play role. they might have problems with new process. Process shrink doesn't automatically stands for better clocks, at least not right away.

[Obscured]

Intel can make a 3.6GHz QC with as much effort as it takes to make a QX6850

And what about stock TDP?

[Shintai]

There is something more interersting than maximum Penryn clock. I'm talking about minium clock of the cheapest Penryn. After all, prices on Penryns will depend on it in particular.

Lowest change right now. I think the 1.6Ghz Pentium 2140 is on the way out for a 1.8Ghz. Or simply dumped as an yet more cheap dualcore. The 4000 series got EOL on the 1.8Ghz..2Ghz is lowest there. So I would say 2Ghz and up. Most likely around 2.33 or 2.4Ghz, because I will expect 65nm to cover the lowend all that time. For quads its 2-2.33Ghz i think, but 45nm is also a bigger economic issue there due to total diesizes. The 50W quad xeons is 2.33 and 2.66Ghz. Today its 1.86 and 2Ghz.

[Obscured]

Shintai Same thoughts here.) And by the way notice that the minimum multiplier for Conroe-generation CPUs is x7 (with the exception of EIST, of course). So... may be... the lowest mutiplier for Penryn-generation CPUs'll be the same. Then we can expect even 2.33GHz as the lowest clock for 333MHz bus speed CPUs.

[Donnie27]

There is something more interersting than maximum Penryn clock. I'm talking about minium clock of the cheapest Penryn. After all, prices on Penryns will depend on it in particular.

I know, hell I can't afford any of the Top models anyway. Mid-range at best.

[alpha0ne]

All I want for Christmas is .................................................. Native Quad 45nm

[Obscured]

But... if the lowest clock for Penryn-generation CPU would be 2.33GHz then the lowest price will be correspondingly high. (i guess in a range of 224$)

[savantu]

And what about stock TDP?

Same 130w.

[savantu]

And please tell us how you came to that conclusion. It's not just math, there are more factors that play role. they might have problems with new process. Process shrink doesn't automatically stands for better clocks, at least not right away.

Maybe in AMD's world...

[Obscured]

savantu

Same 130w.

Comparing 3GHz quad-core and 3.6GHz quad-core?

[red]

http://www.extremetech.com/article2/...2142515,00.asp Seems it surfaced before but no one noticed but here it is again, Penryn's partner in performance

[Donnie27]

http://www.extremetech.com/article2/...2142515,00.asp Seems it surfaced before but no one noticed but here it is again, Penryn's partner in performance

X35 is enough for me I've not gotten it yet.

[hollo]

http://www.extremetech.com/article2/...2142515,00.asp ....

Demo: next-gen desktop processor, Yorkfield quad core processor, two ATI 2900XT cards in Crossfire mode with x16 bandwidth, as well as X38 chipset. Demo of Ghost Recon Advanced Warfighter 2 (which has yet to be released). A0 silicon on X38. One of the silicons of the Yorkfield, whatever that means.

reminds me of something i read in a barcelona thread

poor neglected nvidia

[Shintai]

reminds me of something i read in a barcelona thread

poor neglected nvidia

http://www.youtube.com/watch?v=DqT_5Nv8gs0
http://www.youtube.com/watch?v=-CD-mJMB7Lw

[informal]

As firts results for Penryn show up,http://forum.coolaler.com/showthread...=158813&page=4,
we can put to an end the talk and FUD about Penryn pefromance.
As for gaming benches OCed Penryn and Kentsfield score almost the same in CPU test in mark06
http://img.coolaler.com.tw/images/wd...jylnvnqtzn.png
http://img.coolaler.com.tw/images/ih...j3inlymtug.png

Difference at OCed speed is mere 2%.(where Kentsfield nullifies the FSB advantage of Yorkfield since it is reaching high FSB)
So the scores are practicly the same.
SPI benches show also a small advantage for Penryn(1.5s faster) that equally clocked Conroe at 2.33Ghz.
Where is the fast radix and other improvements intel was bombing the press?The only thing responsible for this small improvements is 6MB of L2 and higher FSB of Penryn.

Next look at Tmpengc perfomance in that topic.NO IMPROVEMENT AT ALL,and that test uses 128bit SEE! Where is so much desired improved super duper shuffle engine?Funny stuff indeed.

So Penryn is almost the same as Conroe,only featuring larger L2 and higehr FSB(later gets absolete since most of overclockers reach the same FSB with their Core2Duo & Quad systems).
Advantages are the SSE4 which COULD bring improvements once there is SSE4 aware software,possible higher OC(this hasn't been confirmed yet) and lower wattage at same clock as Conroe.Everything else is just a PR.Generally it is 5-6% faster or the same as Conroe and Core2Quad.

[red]

http://www.hkepc.com/bbs/hwdb.php?ti...iew&rid=837360
"In addition, Intel former generation 65-nanometer products at the highest vein about 3GHz, clock more than 3GHz, power there will be significant growth, and 45-nm products up to a maximum of about 4GHz."
===
Conroe 65nm 2.33
Power - 61W
Temp - 41C

Wolfdale 45nm 2.33
Power - 43W
Temp - 31C
===
Conroe
Power - 83W
Temp - 49C

Wolfdale
Power - 59W
Temp - 37C
===
Divx 6.6 Alpha w/SSE4 +115.63%

Mainconcept H.264 Encoder (no SSE4?) +12.49%

Half-Life 2 +31.12%

CineBench 9.5

Single-Core/CPU +10.26%
Multi-Core/CPU +7.55%
C4D Shading +8.87%
OpenGL Software +19.20%
C4D Shading +9.24%

ScienceMark 2 +6.22%
===
Doesn't seem like an apples to apples comparison with two mobos specified, but still, Wolfdale on load cooler than Conroe idles plus decent boosts all around for a tweaked Conroe Wonder how results would be paired with X38

[informal]

It won't be enough clock/clock against K10 derivatives.Especially for quad core ones.