Gratuitous Naked Penryn Die Shots - bit-tech.net Forums

Tim S · 17th Apr 2007, 08:19

http://www.bit-tech.net/news/2007/04...ryn_die_shots/

DougEdey · 17th Apr 2007, 08:24

So that's a good few thousand dollars dead right ther.

L2wis · 17th Apr 2007, 09:01

lol, better if there was a shot of it smashed on the floor.

mmorgue · 17th Apr 2007, 10:12

It's neat and all, but uh.. nothing I can do with that pretty coloured wafer of silicon. I'll get moist when it's in a nicely boxed package with my delivery address on it!

However I am sure there are those who will be replying to the article with one hand only..

_DTM2000_ · 17th Apr 2007, 10:23

Quote:

Originally Posted by mmorgue

However I am sure there are those who will be replying to the article with one hand only..

One handed people?

Bindibadgi · 17th Apr 2007, 12:25

Quote:

Originally Posted by mmorgue

It's neat and all, but uh.. nothing I can do with that pretty coloured wafer of silicon. I'll get moist when it's in a nicely boxed package with my delivery address on it!

However I am sure there are those who will be replying to the article with one hand only..

Well I get to play with one tomorrow.

/note to self, don't wear tight underwear.

mmorgue · 17th Apr 2007, 12:41

^^^ doh!

/curse you lucky people!

Hamish · 17th Apr 2007, 12:43

why do they use round wafers? wouldnt it be less wasteful to use square ones?

Munterofamodder · 17th Apr 2007, 12:49

Quote:

Originally Posted by Hamish

why do they use round wafers? wouldnt it be less wasteful to use square ones?

Its a little bit tricky to grow cuboid ingots...

DougEdey · 17th Apr 2007, 13:04

Quote:

Originally Posted by Bindibadgi

Well I get to play with one tomorrow.

/note to self, don't wear tight underwear.

4-storey MEGA WEDGIE!

Hamish · 17th Apr 2007, 13:05

Quote:

Originally Posted by Munterofamodder

Its a little bit tricky to grow cuboid ingots...

care to elaborate? i dont really know anything about cpu manufacturing other than small nm = good

kenco_uk · 17th Apr 2007, 13:06

Quote:

Originally Posted by Munterofamodder

Its a little bit tricky to grow cuboid ingots...

Say what now?

You mean cpu's are grown to order?

DougEdey · 17th Apr 2007, 13:08

Silicon is a crystal, it's grown to a wafer.

Hamish · 17th Apr 2007, 13:13

http://en.wikipedia.org/wiki/Czochralski_process

k so its basicaly cos they have to rotate the thingy to get a solid crystal, squares are a bit hard to rotate properly like that

Tim S · 17th Apr 2007, 14:56

yes, it's to do with the way single crystal Silicon is "grown". They use the same technique for the materials used in aircraft engines, as impurities mean a potentially lower melting point in that particular scenario.

kenco_uk · 17th Apr 2007, 16:24

Well, blimey. Every day is an education. Is there a theory about that a particular place on a wafer always produces the best dies?

devdevil85 · 17th Apr 2007, 16:50

Quote:

Originally Posted by kenco_uk

Is there a theory about that a particular place on a wafer always produces the best dies?

That's kind of like the glass needed for Optical Cable. The more expensive the cable, the closer the glass fibers were to the center of the sheet, or so you'd expect.

Firehed · 18th Apr 2007, 00:36

Well towards the edges tend to be somewhat worse. Back when my dad was engineering stuff along those lines, I actually tossed in a solution that helped fix their problematic data quite a bit.

There's a little alignment notch at one point along the edge. Err, not point as it's a circle... but you know what I mean. Anyways, as it turns out, the tiny amount of extra surface area you get along the edge because of that can really screw with heat uniformity data. Toss that point (and, more likely than not, the chip that would be built there), and the data gets a surprising amount better.

It's true for the whole edge area (it adds something like 900mm^2), so the edge will cool slightly faster, and as a result might not always have as good of a yield. As dies get smaller though, we can fit more chips per wafer and help counteract this.

17th Apr 2007, 08:19	#1
Tim S Pewlius Caesar bit-tech Staff Join Date: Nov 2001 Location: Ascot, Berks Posts: 18,021	Gratuitous Naked Penryn Die Shots http://www.bit-tech.net/news/2007/04...ryn_die_shots/ __________________ Tim on Twitter \| Tim's Blog \| Tim's Flickr The bit-tech Blog \| bit-tech on Twitter \| Join the bit-tech & Custom PC folding team (#35947)

17th Apr 2007, 08:24	#2
DougEdey I pwn all your storage Join Date: Jul 2005 Location: Southampton Posts: 13,933	So that's a good few thousand dollars dead right ther. __________________ Burnout: Paradise Stats!XBL: DougEdey Bindi PSN ID: DougEdey Twitter

17th Apr 2007, 09:01	#3
L2wis Hypermodder Join Date: Jun 2004 Location: UK Posts: 659	lol, better if there was a shot of it smashed on the floor. __________________ Lewis Wells Get Firefox! \| Get Thunderbird! \| Join Bit-tech folding@home

17th Apr 2007, 13:08	#13
DougEdey I pwn all your storage Join Date: Jul 2005 Location: Southampton Posts: 13,933	Silicon is a crystal, it's grown to a wafer. __________________ Burnout: Paradise Stats!XBL: DougEdey Bindi PSN ID: DougEdey Twitter

17th Apr 2007, 14:56	#15
Tim S Pewlius Caesar bit-tech Staff Join Date: Nov 2001 Location: Ascot, Berks Posts: 18,021	yes, it's to do with the way single crystal Silicon is "grown". They use the same technique for the materials used in aircraft engines, as impurities mean a potentially lower melting point in that particular scenario. __________________ Tim on Twitter \| Tim's Blog \| Tim's Flickr The bit-tech Blog \| bit-tech on Twitter \| Join the bit-tech & Custom PC folding team (#35947)

17th Apr 2007, 10:12	#4
mmorgue Supermodder Join Date: Feb 2005 Location: Mos Eisley, in the bar... Posts: 378	It's neat and all, but uh.. nothing I can do with that pretty coloured wafer of silicon. I'll get moist when it's in a nicely boxed package with my delivery address on it! However I am sure there are those who will be replying to the article with one hand only..

17th Apr 2007, 12:41	#7
mmorgue Supermodder Join Date: Feb 2005 Location: Mos Eisley, in the bar... Posts: 378	^^^ doh! /curse you lucky people!

17th Apr 2007, 12:43	#8
Hamish I Mod, Therefore I Own Join Date: Nov 2002 Location: London Posts: 3,587	why do they use round wafers? wouldnt it be less wasteful to use square ones?

17th Apr 2007, 13:13	#14
Hamish I Mod, Therefore I Own Join Date: Nov 2002 Location: London Posts: 3,587	http://en.wikipedia.org/wiki/Czochralski_process k so its basicaly cos they have to rotate the thingy to get a solid crystal, squares are a bit hard to rotate properly like that

17th Apr 2007, 16:24	#16
kenco_uk glourious bacon? Join Date: Nov 2003 Location: Bourne, Lincolnshire, UK Posts: 6,292	Well, blimey. Every day is an education. Is there a theory about that a particular place on a wafer always produces the best dies? __________________ MSI X58M - i7 920 4Ghz - 6GB Corsair - GTX260 648/2520 ::: Vote for bit-tech T - theevilelephant - "it sounded like bouncing around on a pogo stick, screaming in chinese while occasionally throwing a cat at a door."

18th Apr 2007, 00:36	#18
Firehed Why not? I own a domain to match. Join Date: Feb 2004 Location: An hour north of Boston Posts: 12,576	Well towards the edges tend to be somewhat worse. Back when my dad was engineering stuff along those lines, I actually tossed in a solution that helped fix their problematic data quite a bit. There's a little alignment notch at one point along the edge. Err, not point as it's a circle... but you know what I mean. Anyways, as it turns out, the tiny amount of extra surface area you get along the edge because of that can really screw with heat uniformity data. Toss that point (and, more likely than not, the chip that would be built there), and the data gets a surprising amount better. It's true for the whole edge area (it adds something like 900mm^2), so the edge will cool slightly faster, and as a result might not always have as good of a yield. As dies get smaller though, we can fit more chips per wafer and help counteract this. __________________ hire me @ eric-stern.com - web developer and php ninja pics @ my smugmug :: Twitter @firehed :: blog @ firehed.net 40D\|580EXII\|285HV\|AB800\|70-200f/4LIS\|17-50f/2.8\|150f/2.8Macro\|50f/1.8 MacPro @ 8x2.8GHz, 10GB FBDDR2, 3TB HD :: MBP @ 2x2.2GHz, 4GB DDR2, 320GB HD