| View previous topic :: View next topic |
| Author |
Message |
Benjagan91
Joined: 04 Sep 2018
Posts: 8
|
 Posted: Mon Sep 10, 2018 11:16 pm Post subject: Delidded Jayhawk... (destroyed) |
 |
|
Good afternoon from Australia,
I've been waiting for quite a while to no avail for someone to delid one of the Jayhawk samples that were recently circulated... So with my lack of skills and tools I did so today in my lunch break.
First I used a razor blade to remove the black glue around the edge of the IHS underside. The IHS was soldered on in the middle so I heated up the IHS with a cigarette lighter until the solder separated. I then sanded the top of the separated 'die' for lack of a better word.
I don't know much about this/what to identify so here's the photos I took. Sorry for the poor photo quality. I hope this doesn't get me banned! 
Thoughts? Discuss? |
|
| Back to top |
|
 |
Benjagan91
Joined: 04 Sep 2018
Posts: 8
|
| Posted: Mon Sep 10, 2018 11:18 pm Post subject: |
|
|
| More photos. Looks like the order of the photos is mixed up! My apologies |
|
| Back to top |
|
|
xsecret
Joined: 01 Feb 2004
Posts: 1846
Location: France
|
|
| Back to top |
|
|
xsecret
Joined: 01 Feb 2004
Posts: 1846
Location: France
|
|
| Back to top |
|
|
Thandor
Joined: 06 Oct 2008
Posts: 461
|
| Posted: Tue Sep 11, 2018 7:38 am Post subject: |
|
|
Looks a bit like the Pentium D 'Smithfield' (Or Xeon 'Paxville' as your sample is LGA771).
_________________
thandor.net - hardware
And the rest of us would be carousing the aisles, stuffing baloney. |
|
| Back to top |
|
|
xsecret
Joined: 01 Feb 2004
Posts: 1846
Location: France
|
|
| Back to top |
|
|
Thandor
Joined: 06 Oct 2008
Posts: 461
|
| Posted: Tue Sep 11, 2018 3:03 pm Post subject: |
|
|
| xsecret wrote: | | First tape-out for Smithfield is almost one year later :-/ |
The launch of the (production) Pentium D was one year later (May 2005). The chip from Benjagan91 is dated 17th week 2004. I have a fully working Pentium D 840 sample dated 45th week of 2004. That's roughly half a year difference. My Pentium D 840 sample doesn't show signs of being one of the first working samples; there must be earlier samples out there which close the gap to this 'Jayhawk' sample. I really wonder what kind of 'die' is inside this chip and I still think it can be a very early Smithfield judging by time-frames and chip-size  .
I've read that the Pentium D was built in a time-frame of just 9 months (which might be exaggerated for PR): my sample is 6 months earlier than the official launch and it's 6 months after the 'Jayhawk' sample. It's not 9 months... but it's close.
But let's wait when dimensions are measured and hopefully a die photo .
_________________
thandor.net - hardware
And the rest of us would be carousing the aisles, stuffing baloney. |
|
| Back to top |
|
|
CPUShack
Joined: 16 Jun 2003
Posts: 34259
Location: State of Jefferson, USA
|
|
| Back to top |
|
|
frag_
Joined: 17 Nov 2008
Posts: 4015
Location: Estonia
|
| Posted: Tue Sep 11, 2018 4:33 pm Post subject: |
|
|
At least it have the die.
Interesting thing to note: LGA771 was launched much later than LGA775, difference is more than one year.
Sample in question is made over two years (!) before LGA771 was released (may 2006). |
|
| Back to top |
|
|
xsecret
Joined: 01 Feb 2004
Posts: 1846
Location: France
|
| Posted: Tue Sep 11, 2018 5:13 pm Post subject: |
|
|
| Thandor wrote: | | I have a fully working Pentium D 840 sample dated 45th week of 2004. |
Do you have a picture of this CPU (or at least a QSPEC to share)?
The timeline clearly doesn't match with the information I have (except if you're talking about QDDS).
_________________
ES-Only Collector : http://www.engineering-sample.com
Universal Chip Analyzer (UCA) : https://x86.fr/uca / http://www.cpu-world.com/forum/viewtopic.php?t=34349
Last edited by xsecret on Tue Sep 11, 2018 5:25 pm; edited 1 time in total |
|
| Back to top |
|
|
xsecret
Joined: 01 Feb 2004
Posts: 1846
Location: France
|
|
| Back to top |
|
|
Benjagan91
Joined: 04 Sep 2018
Posts: 8
|
| Posted: Tue Sep 11, 2018 6:21 pm Post subject: |
|
|
No caliper sorry. This is the best I could do. Photos were taken after about 1hr of hand sanding. Not sure what to expect as I haven't done this before. I'm not sure if the 'crack' is a result of my delidding efforts or by design. It's very straight though.
I don't know what a 'Thermal Test Vehicle' should look like underneath the IHS. Do you guys think this is a functional silicon die? I have 2 more of these samples and a 771 to 775 converter coming in the mail... I've only got a 945 chipset socket 775 Shuttle XPC nowadays though. The cooling system wouldn't be up to any benchmarking but I think it'd handle a 150W TDP processor at idle... if the computer made it to POST.
Perhaps early 771 boards would have the appropriate BIOS microcode to boot one of these..?
Maybe the 'crack' on the die in my photos was put there on purpose by Intel to 'disable' the chips? Perhaps if someone else can delid one a bit more carefully and skilfully than myself and has the correct tools we can learn more... |
|
| Back to top |
|
|
xsecret
Joined: 01 Feb 2004
Posts: 1846
Location: France
|
| Posted: Tue Sep 11, 2018 6:34 pm Post subject: |
|
|
According to my source, no working silicon for Tejas/Jayhawk has ever been produced (it was canceled before tape-out).
Thermal Samples are usually built with one of more high power resistors etched inside a wafer (to mimic the same physical properties of a real die).
If you continue to sand off, some pattern will start to appear.
Anyway, the die seems way too thick to be a real die.
The size is ~200 mm˛, so very close to the Smithfield die.
_________________
ES-Only Collector : http://www.engineering-sample.com
Universal Chip Analyzer (UCA) : https://x86.fr/uca / http://www.cpu-world.com/forum/viewtopic.php?t=34349 |
|
| Back to top |
|
|
Thandor
Joined: 06 Oct 2008
Posts: 461
|
| Posted: Wed Sep 12, 2018 3:02 am Post subject: |
|
|
| xsecret wrote: | | Thandor wrote: | | I have a fully working Pentium D 840 sample dated 45th week of 2004. |
Do you have a picture of this CPU (or at least a QSPEC to share)?
The timeline clearly doesn't match with the information I have (except if you're talking about QDDS). |
Yes, indeed QDDS. I have QDDS dated 0445 and 0448. QDDZ dated 0448 and QDOM dated 0451. A link to my oldest QDDS from 45th week 2004: thandor.net - Pentium D 840.
| xsecret wrote: | | CPUShack wrote: | | I do have prescott Samples in LGA775 from late 2003 |
Same here. The first LGA775 Prescott are B1-Step QJ23, QJ29 and QJ35. (A0 & B0-Step are 478s only). Tape-out for B1-Step is WW32'03. My earliest LGA775 Prescott is WW39 or WW40. |
I have QJ32ES as well. Dated 42th week of 2003: thandor.net - Intel Pentium 4 520 (2.8 GHz).
_________________
thandor.net - hardware
And the rest of us would be carousing the aisles, stuffing baloney. |
|
| Back to top |
|
|
xsecret
Joined: 01 Feb 2004
Posts: 1846
Location: France
|
|
| Back to top |
|
|
|
![[ Hidden ]](https://cdn.cpu-world.com/forum/templates/CPUWorld/images/lang_english/icon_email.gif "[ Hidden ]"){kind=icon}