| View previous topic :: View next topic |
| Author |
Message |
Neon_WA
Joined: 08 Nov 2008
Posts: 7146
Location: Margaret River, West Australia
|
 Posted: Sat Jul 17, 2010 9:27 pm Post subject: Opteron sample Id |
 |
|
just got a Opteron sample delivered to Avicc yesterday. AMDs part numbering of Opteron samples I find isn't the easiest to follow at times 
The part number is 2S190805L4BGC and this is how I presume the part number works for samples?
2 = 2-Way
S = Server
190 = 1900MHz
805 = ???
L = 1207-land LGA (socket Fr2)
4 = 4 Cores
B = 2Mb L3 cache
GC = Stepping B1
so it is likely the production part for this would be OS2347WAL4BGC ?
http://www.cpu-world.com/CPUs/K10/AMD-Third%20Generation%20Opteron%202347%20-%20OS2347WAL4BGC.html
_________________
There are 10 types of people in this world:
those who understand binary and those who don't. ~Author Unknown
http://www.x86-guide.net/Neon-WA/en/collection.html |
|
| Back to top |
|
 |
Neon
Joined: 04 Feb 2008
Posts: 1512
Location: Dallas, Texas, USA
|
| Posted: Sat Jul 17, 2010 10:16 pm Post subject: |
|
|
I think that is mostly correct. Not sure if the 2 refers to 2-way server.
In fact, I am not sure if AMD limited their Opteron samples to 2-way operation. It may be that all samples were 8-way capable, but it would require a 4P or 8P motherboard to test.
I can say that my 2S200805L4BGC is a Barcelona B1. If you look here, check the CPUID tab, notice that the chip identifies as "Quad-Core AMD Opteron(tm) Processor 8300":
http://www.cpu-world.com/CPUs/K10/AMD-Third%20Generation%20Opteron%202S200805L4BGC.html |
|
| Back to top |
|
|
Neon_WA
Joined: 08 Nov 2008
Posts: 7146
Location: Margaret River, West Australia
|
| Posted: Sat Jul 17, 2010 10:25 pm Post subject: |
|
|
| Neon wrote: | I think that is mostly correct. Not sure if the 2 refers to 2-way server.
In fact, I am not sure if AMD limited their Opteron samples to 2-way operation. It may be that all samples were 8-way capable, but it would require a 4P or 8P motherboard to test.
I can say that my 2S200805L4BGC is a Barcelona B1. If you look here, check the CPUID tab, notice that the chip identifies as "Quad-Core AMD Opteron(tm) Processor 8300":
|
Thx for the info mate 
Yes was wondering about the 2S.. but i see samples are marked with 1D, 2S & ZS
1D = 1-way Desktop?
2S = 2-way Server?
ZS = up to 8-way Server?
_________________
There are 10 types of people in this world:
those who understand binary and those who don't. ~Author Unknown
http://www.x86-guide.net/Neon-WA/en/collection.html |
|
| Back to top |
|
|
Neon
Joined: 04 Feb 2008
Posts: 1512
Location: Dallas, Texas, USA
|
| Posted: Sat Jul 17, 2010 10:41 pm Post subject: |
|
|
That seems like a logical schema, but it would mean that CPUID is incorrect.
I have
1D samples for Opteron 1300 and Phenom X3;
2S samples for Opteron 8300
ZS samples for Opteron 8300 |
|
| Back to top |
|
|
Neon_WA
Joined: 08 Nov 2008
Posts: 7146
Location: Margaret River, West Australia
|
| Posted: Sat Jul 17, 2010 11:34 pm Post subject: |
|
|
| Neon wrote: | That seems like a logical schema, but it would mean that CPUID is incorrect.
|
well we know about CPUID info
with early samples.. maybe the Processor name (CWID) isn't necessarily correctly set in the processor.
I see your 1D160110J4BGA comes up as a 130000 Quad-Core not as a 1300 
_________________
There are 10 types of people in this world:
those who understand binary and those who don't. ~Author Unknown
http://www.x86-guide.net/Neon-WA/en/collection.html |
|
| Back to top |
|
|
gshv
Joined: 01 Feb 2003
Posts: 7898
Location: Fairfax, VA USA
|
| Posted: Tue Jul 20, 2010 7:52 pm Post subject: |
|
|
| Neon wrote: | That seems like a logical schema, but it would mean that CPUID is incorrect.
I have
1D samples for Opteron 1300 and Phenom X3;
2S samples for Opteron 8300
ZS samples for Opteron 8300 |
I'm inclined to think I think that 2S are samples of 2-way processors because I'm not convinced that generated CPU name string is correct. I use algorithm that was published by AMD, but it may not work with engineering samples. AMD in their CPUID paper recommends to show "AMD engineering sample" if certain condition is met. My code ignores this condition and generates the CPU name. Sorry, I don't remember what this condition is, it's some kind of check for one or more bits returned by the CPUID instruction.
Gennadiy |
|
| Back to top |
|
|
Neon
Joined: 04 Feb 2008
Posts: 1512
Location: Dallas, Texas, USA
|
| Posted: Thu Sep 02, 2010 2:26 pm Post subject: |
|
|
I have found a 3rd generation Opteron that seems to vary from the proposed scheme. It is a socket AM3 Opteron, likely a Suzuka core, with part number:
2S200102K4B01
2S = 2-way server ??
200 = 2000 MHz
1 = 1-way ??
02
K = socket AM3
4 = 4 core
B = 2048 KB L2
01 = revision ??
It seems unlikely that this AM3 can be used in a 2-way server.
Maybe the 6th digit indicates maximum # of processor per box, 1 in the part number above. The 3rd generation Opteron ES part numbers so far have either 1 or 8 in the 6th digit.
We need more data. |
|
| Back to top |
|
|
gshv
Joined: 01 Feb 2003
Posts: 7898
Location: Fairfax, VA USA
|
| Posted: Thu Sep 02, 2010 3:06 pm Post subject: |
|
|
Here are a few ZS* part numbers that may help (or confuse even more):
ZS180805L4BGA
ZS190800TCE25
ZS2008WAL4BGH
ZS2108WAL4BGH
ZS2208WAL4BGH
ZS230805L4D12
ZS2308WAL4BGH
ZS251805L4D12
It seems that fifth digit has some meaning, and it can be 0 or 1. Digits 7-8 are probably relate to power consumption and the socket. I agree with Neon that the last two letters or digits could be a core stepping or revision. I guess "01", "12", "25" are some kind of early or intermediate steppings.
Gennadiy |
|
| Back to top |
|
|
gshv
Joined: 01 Feb 2003
Posts: 7898
Location: Fairfax, VA USA
|
| Posted: Sat Sep 04, 2010 3:04 pm Post subject: |
|
|
More Opteron sample part numbers from my collection:
1SW2600GK0603
2S190805L4BGC
2S200805L4BGC
ZS160805L4BGA
ZS180805L4BGA
ZS200805L4BGC
ZSA2200GAA6CR
ZSW2200GAA631
ZSW2200GAA635
ZSW2400GAA635
ZSW2600GAA631
Neon, if you want to test any of these, please drop me a list with the part numbers, and I'll ship them to you.
Gennadiy |
|
| Back to top |
|
|
Neon
Joined: 04 Feb 2008
Posts: 1512
Location: Dallas, Texas, USA
|
| Posted: Sat Sep 04, 2010 7:19 pm Post subject: |
|
|
Gennadiy,
From your list of socket F Opterons, the following are parts that I have not yet tested:
ZS190800TCE25
ZS2008WAL4BGH
ZS2108WAL4BGH
ZS2208WAL4BGH
ZS230805L4D12
ZS2308WAL4BGH
ZS251805L4D12
1SW2600GK0603
ZSA2200GAA6CR
ZSW2200GAA635
ZSW2400GAA635
ZSW2600GAA631
Thus, all of those would be good candidates for testing, and I will run your CWID program with each. However, note that the first part on that list is likely socket G34, so I cannot test it. I have already tested the remaining parts listed below, but if you would like me to determine if yours are functional, then I would be glad to do that.
ZS180805L4BGA
2S190805L4BGC
2S200805L4BGC
ZS160805L4BGA
ZS180805L4BGA
ZS200805L4BGC
ZSW2200GAA631 |
|
| Back to top |
|
|
Neon_WA
Joined: 08 Nov 2008
Posts: 7146
Location: Margaret River, West Australia
|
| Posted: Tue Feb 22, 2011 2:41 am Post subject: |
|
|
i just purchased a pair of ES 2nd Gen Opterons
the part is for a 2216 and currently these ES parts are listed
ZSW2400GAA631
ZSW2400GAA635
I will need to check when i receive them, but the part number is listed as ZSW2400GAA618
Is the variation between these parts likely only be minor stepping differences or the last 2 numbers refer to something else ?
_________________
There are 10 types of people in this world:
those who understand binary and those who don't. ~Author Unknown
http://www.x86-guide.net/Neon-WA/en/collection.html |
|
| Back to top |
|
|
lither
Joined: 04 Dec 2005
Posts: 1362
Location: Taiwan
|
| Posted: Tue Feb 22, 2011 2:52 am Post subject: |
|
|
Hi
do you know what ZPN164xxx, and ZPN166 mean?
ZPN 164
opteron ZPN166
best regards |
|
| Back to top |
|
|
Neon_WA
Joined: 08 Nov 2008
Posts: 7146
Location: Margaret River, West Australia
|
|
| Back to top |
|
|
lither
Joined: 04 Dec 2005
Posts: 1362
Location: Taiwan
|
| Posted: Tue Feb 22, 2011 10:25 am Post subject: |
|
|
| thank you ! |
|
| Back to top |
|
|
gshv
Joined: 01 Feb 2003
Posts: 7898
Location: Fairfax, VA USA
|
| Posted: Tue Feb 22, 2011 10:57 pm Post subject: |
|
|
| Neon_WA wrote: | i just purchased a pair of ES 2nd Gen Opterons
the part is for a 2216 and currently these ES parts are listed
ZSW2400GAA631
ZSW2400GAA635
I will need to check when i receive them, but the part number is listed as ZSW2400GAA618
Is the variation between these parts likely only be minor stepping differences or the last 2 numbers refer to something else ? |
I think you're right, and the difference is only in core stepping.
Gennadiy |
|
| Back to top |
|
|
|
![[ Hidden ]](https://cdn.cpu-world.com/forum/templates/CPUWorld/images/lang_english/icon_email.gif "[ Hidden ]"){kind=icon}
