| View previous topic :: View next topic |
| Author |
Message |
feipoa
Joined: 08 Mar 2011
Posts: 553
Location: Canada
|
 Posted: Mon Sep 17, 2012 1:40 am Post subject: UMC U5S and U5SX differences? |
 |
|
| Does anyone know the difference between the UMC U5S Super33 and the UMC U5SX-33? Was the 'X' added to the later chips to differentiate it with the U5SD? Thanks. |
|
| Back to top |
|
 |
feipoa
Joined: 08 Mar 2011
Posts: 553
Location: Canada
|
| Posted: Sat Sep 29, 2012 5:04 am Post subject: |
|
|
| Has nobody tested this chip? |
|
| Back to top |
|
|
lither
Joined: 04 Dec 2005
Posts: 1362
Location: Taiwan
|
| Posted: Sat Sep 29, 2012 7:04 am Post subject: |
|
|
check member : vinicio's site
http://www.cpu-museo.it/ (UMC is on bottom of the page )
U5SX U5S U5SD have the same CPUID 423
and no FPU
=>486SX
| vinicio wrote: |
U5DX2 istruzioni e piedini compatibili con l'Intel 80486DX2 nel formato CPGA (Ceramic Pin Grid Array);
U5S istruzioni e piedini compatibili con l'Intel i80486SX (CPGA);
U5SD istruzioni e piedini compatibili con l'Intel i80486SX (CPGA),
U5D istruzioni e piedini compatibili con l'Intel i80486DX (CPGA);
U5SF versione U5S con 208 pin , nel formato QFP (Quad Flat Package);
U5SP versione U5S nel formato PPGA (Plastic Pin Grid Array).
|
|
|
| Back to top |
|
|
feipoa
Joined: 08 Mar 2011
Posts: 553
Location: Canada
|
| Posted: Fri Oct 19, 2012 4:37 am Post subject: |
|
|
Ok, so it is likely that they all perform the same per-clock, with the exception of the U5D, which apparently has the FPU built-in, as with this chip, http://www.chipdb.org/img-umc-u486dx2-u5-green-3240.htm
Anyone test the performance of a U5D or at least seen one in the flesh? |
|
| Back to top |
|
|
mtx500
Joined: 21 Nov 2003
Posts: 117
Location: Nuremberg, Germany
|
| Posted: Sat Oct 20, 2012 4:40 am Post subject: |
|
|
You night want to see if the undocumented UMC instructions return the same results.
Years ago I wrote a small program to read them out. I've put it there. Source code is included. I just saw that some comments in it are in German. If you need more information on how it works, just ask. It was compiled with Borland C++. |
|
| Back to top |
|
|
feipoa
Joined: 08 Mar 2011
Posts: 553
Location: Canada
|
| Posted: Sat Oct 20, 2012 7:15 am Post subject: |
|
|
| Why not also include the 486-optimised compiled code (DOS) so I can run the program straight away? Although I think I have a computer with Borland C++ already installed on it somehwere. |
|
| Back to top |
|
|
mtx500
Joined: 21 Nov 2003
Posts: 117
Location: Nuremberg, Germany
|
| Posted: Sat Oct 20, 2012 10:52 am Post subject: |
|
|
| feipoa wrote: | | Why not also include the 486-optimised compiled code (DOS) so I can run the program straight away? |
The compilation result, umctest.exe, is included, and it should run on bare DOS (it won't run in a Console window in Windows). At least it did so on my PC with an UMC CPU years ago. If not, what's the exact error message? |
|
| Back to top |
|
|
feipoa
Joined: 08 Mar 2011
Posts: 553
Location: Canada
|
| Posted: Sat Oct 20, 2012 3:29 pm Post subject: |
|
|
| When I open the archive, I get "unexpected end of file" and do not see any file with an *.exe extension. I see umctest.r$p, umctest.~de, and umctest.asm |
|
| Back to top |
|
|
mtx500
Joined: 21 Nov 2003
Posts: 117
Location: Nuremberg, Germany
|
| Posted: Sun Oct 21, 2012 3:28 pm Post subject: |
|
|
| Seems something went wrong with the upload - the file was too short. I fixed it, please try again. Sorry for that. |
|
| Back to top |
|
|
feipoa
Joined: 08 Mar 2011
Posts: 553
Location: Canada
|
| Posted: Sun Oct 21, 2012 6:19 pm Post subject: |
|
|
| Thank you correcting the archive. When I get around to testing my U5SX chip, I'll remember to try this program. |
|
| Back to top |
|
|
|
