| View previous topic :: View next topic |
| Author |
Message |
Robev
Joined: 10 Mar 2006
Posts: 3693
Location: New Zealand
|
 Posted: Sat Feb 25, 2012 5:28 am Post subject: IBM BGA Chips |
 |
|
I haven't a clue what these are and the part #'s don't line up with anything in the database.
Anyone got any Ideas? Early Power PC Maybe?
Cheers
_________________
The Older they are the Better they are. |
|
| Back to top |
|
 |
Neon_WA
Joined: 08 Nov 2008
Posts: 7146
Location: Margaret River, West Australia
|
|
| Back to top |
|
|
bccwchan
Joined: 20 Sep 2004
Posts: 2585
Location: Hong Kong
|
| Posted: Sat Feb 25, 2012 11:56 pm Post subject: |
|
|
is it possible that they are power3 CPU?
They have the same pin count. |
|
| Back to top |
|
|
Robev
Joined: 10 Mar 2006
Posts: 3693
Location: New Zealand
|
| Posted: Sat Feb 25, 2012 11:58 pm Post subject: |
|
|
| bccwchan wrote: | is it possible that they are power3 CPU?
They have the same pin count. |
But how can I find out 
_________________
The Older they are the Better they are. |
|
| Back to top |
|
|
bccwchan
Joined: 20 Sep 2004
Posts: 2585
Location: Hong Kong
|
| Posted: Sun Feb 26, 2012 12:29 am Post subject: |
|
|
| Robev wrote: | | bccwchan wrote: | is it possible that they are power3 CPU?
They have the same pin count. |
But how can I find out |
Searching in the Internet by "IBM power3 CPU" and compare the bottom with yours. You will find it has the same bottom gold dot. |
|
| Back to top |
|
|
doccybrown
Joined: 03 Oct 2005
Posts: 1736
Location: Germany
|
| Posted: Sun Feb 26, 2012 12:35 am Post subject: |
|
|
| Robev wrote: | | bccwchan wrote: | is it possible that they are power3 CPU?
They have the same pin count. |
But how can I find out |
Alas this part-numbers are absolutely unknown I think.
The count of lands on the bottom does not match to
a Power3-II (41x41-3).
So demounting the heatspreader and checking the size, alignment
of the chipdie and position and counting of capacitors on
the top will allow to say very certain what they are.
Opening with a vise should work although it is a very
risky job because the heatspreaders are edged and you may
need to get some experiences with normal heatspreaders before.
Here you can see a picture how to remove heatsinks, heatspreaders
can be removed the same way (with due care).
http://www.cpu-world.com/forum/viewtopic.php?t=6849&highlight=vise
There will be still interested collectors for the chips
if they are opened and turn out to be CPUs, I personally
prefer opened ones.
_________________
Ordem e Progresso |
|
| Back to top |
|
|
doccybrown
Joined: 03 Oct 2005
Posts: 1736
Location: Germany
|
| Posted: Sun Feb 26, 2012 12:38 am Post subject: |
|
|
There is a chance these are RS64-I processors, I will check the landcount...
_________________
Ordem e Progresso |
|
| Back to top |
|
|
Robev
Joined: 10 Mar 2006
Posts: 3693
Location: New Zealand
|
| Posted: Sun Feb 26, 2012 12:53 am Post subject: |
|
|
| doccybrown wrote: | Alas this part-numbers are absolutely unknown I think.
The count of lands on the bottom does not match to
a Power3-II (41x41-3).
So demounting the heatspreader and checking the size, alignment
of the chipdie and position and counting of capacitors on
the top will allow to say very certain what they are.
Opening with a vise should work although it is a very
risky job because the heatspreaders are edged and you may
need to get some experiences with normal heatspreaders before.
Here you can see a picture how to remove heatsinks, heatspreaders
can be removed the same way (with due care).
http://www.cpu-world.com/forum/viewtopic.php?t=6849&highlight=vise
There will be still interested collectors for the chips
if they are opened and turn out to be CPUs, I personally
prefer opened ones. |
I used to remove heatsinks off 486's and Pentiums this way but I doubt that these would be as easy to remove that way.
Maybe I will sell to someone who wants to try it 
_________________
The Older they are the Better they are. |
|
| Back to top |
|
|
Robev
Joined: 10 Mar 2006
Posts: 3693
Location: New Zealand
|
| Posted: Sun Feb 26, 2012 1:29 am Post subject: |
|
|
| bccwchan wrote: | | searching in the Internet by "IBM power3 CPU" and compare the bottom with yours. You will find it has the same bottom gold dot. |
not that it helps me but I did find this
Following is an example of the processor card information for a 333 MHz Model
170:
Processor Card:
Part Number.................11K0857
EC Level....................D72830
Serial Number...............L200016008
FRU Number..................00P2180
Manufacture ID..............1980
Version.....................RS6K
Product Specific.(ZC).......PS=0013D92D40,LB=0009EC96A0,
SB=0005ABC3C9,NP=01,L2=01024,
PF=711,SV=3,VR=2,ER=0000
Product Specific.(ZB).......BC=30602,SG=
Physical Location: P1-C1
PS shows the processor speed in hex-number digit (0x0013D92D40 =
333000000 Hz = 333 MHz).
Following is an example of the processor card information for a 400 MHz Model
170:
Processor Card:
Part Number.................11K0864
EC Level....................D72830
Serial Number...............L200016010
FRU Number..................00P2181
Manufacture ID..............1980
Version.....................RS6K
Product Specific.(ZC).......PS=0017D78400,LB=000BEBC200,
SB=0005F5E100,NP=01,L2=04096,
PF=7D4,SV=3,VR=2,ER=0000
Product Specific.(ZB).......BC=30602,SG=
Physical Location: P1-C1
PS shows the processor speed in hex-number digit (0x0017D78400 = 400000000
Hz = 400 MHz).
Level 1 Cache
The Model 170 uses a 64 KB data and 32 KB instruction 128-way set associative
L1 cache. The size of both data and instruction cache reduces the number of
cache misses, results in more cache hits, and maximizes performance. Both data
and instruction cache are parity protected.
Level 2 Cache
The 44P Model 170 processor card has either 1 MB (333 MHz) or 4 MB
(400 MHz) of L2 cache located on the processor card. L2 cache is used to lower
the time spent accessing memory data and increase performance. The L2 cache
extends L1 cache benefits by adding more cache to the memory pipeline.
The speed of the L2 cache is dependent upon the processor speed. The L2
cache speed for the 333 MHz processor is 167 MHz (2:1 ratio) and for the 400
MHz processor is 200 MHz (2:1 ratio).
The L2 cache uses a direct mapped cache methodology. There is a dedicated
external interface to the L2 cache not shared with the 6XX bus. This allows
concurrent access to both the L2 cache and the 6XX bus.
_________________
The Older they are the Better they are. |
|
| Back to top |
|
|
doccybrown
Joined: 03 Oct 2005
Posts: 1736
Location: Germany
|
| Posted: Sun Feb 26, 2012 1:41 am Post subject: |
|
|
I'm sorry Robev, without the p/n's of the boards these
chips come from there are as good as no chances to reveal it.
_________________
Ordem e Progresso |
|
| Back to top |
|
|
bccwchan
Joined: 20 Sep 2004
Posts: 2585
Location: Hong Kong
|
| Posted: Tue Feb 28, 2012 6:50 am Post subject: |
|
|
| I have two purple Power3 cpu that have the same pin count. Please refer to the photo below. |
|
| Back to top |
|
|
Robev
Joined: 10 Mar 2006
Posts: 3693
Location: New Zealand
|
| Posted: Tue Feb 28, 2012 6:55 am Post subject: |
|
|
Did the Chip have that yellow label on it when you took the top off
_________________
The Older they are the Better they are. |
|
| Back to top |
|
|
bccwchan
Joined: 20 Sep 2004
Posts: 2585
Location: Hong Kong
|
| Posted: Tue Feb 28, 2012 7:02 am Post subject: |
|
|
| Robev wrote: | | Did the Chip have that yellow label on it when you took the top off |
The seller made it for me.
BTW, I have the white ceramic one with the same pin count.
Please have a look. |
|
| Back to top |
|
|
bccwchan
Joined: 20 Sep 2004
Posts: 2585
Location: Hong Kong
|
| Posted: Tue Feb 28, 2012 7:09 am Post subject: |
|
|
If you want to know exactly what they're. You need to take off the silver cover.
Please refer to the professional gudeline below.
http://www.cpu-galaxy.at/Article/IBM_Power4.htm
Or you can sell one to me, I'll open it to find its identity. |
|
| Back to top |
|
|
Robev
Joined: 10 Mar 2006
Posts: 3693
Location: New Zealand
|
| Posted: Tue Feb 28, 2012 7:10 am Post subject: |
|
|
 Now that method makes sense, I will try it on one, Thankyou 
_________________
The Older they are the Better they are. |
|
| Back to top |
|
|
|
![[ Hidden ]](https://cdn.cpu-world.com/forum/templates/CPUWorld/images/lang_english/icon_email.gif "[ Hidden ]"){kind=icon}
