68060 testing on Replay board

[mikej]

So, I've picked up 15 x 68060 E41J mask set devices from two suppliers in China for testing of the daughter boards.

I'll use the Replay board+daughter board and a ZIF socket to do screening at the supplier later.

Running up workbench is not a bad test, but does anybody have ideas on how to do a more thorough test? Is there any CPU test suites I can use to test all aspects of the device?

I also need to read the id register to verify the mask revision.
Any help appreciated!
Cheers,
MikeJ

[Cosmos Amiga]

With Asm-one :

 move.l 4.w,a6
 lea readPCR(PC),a5
 jsr -$1E(a6)
 nop

readPCR:
 movec PCR,d0
 rte

Revision number in d0 is the X : $xxxxxXxx


Et voilà !

[Crumb]

So, I've picked up 15 x 68060 E41J mask set devices from two suppliers in China for testing of the daughter boards.

I'll use the Replay board+daughter board and a ZIF socket to do screening at the supplier later.

Running up workbench is not a bad test, but does anybody have ideas on how to do a more thorough test? Is there any CPU test suites I can use to test all aspects of the device?

I also need to read the id register to verify the mask revision.
Any help appreciated!
Cheers,
MikeJ

Try out some cpu hungry Amiga demos like (in no particular order):
We come in peace:
http://ada.untergrund.net/?p=demo&i=713
Massive:
http://ada.untergrund.net/?p=demo&i=737
Human traffic:
http://ada.untergrund.net/?p=demo&i=734
Kioea:
http://ada.untergrund.net/?p=demo&i=758
Starstruck
http://ada.untergrund.net/?p=demo&i=540

ClickBoom's Quake demo may be a good idea too :-)

Piru's WhichAmiga will also give you information about current cpu mask IIRC.

[Foul]

http://www.sintonen.fi/sw/WhichAmiga_beta.lha

WhichAmiga can display 68060 rev

[freqmax]

Too bad SETI@Home or such won't run on m68k

[wawrzon]

if of any help, seems to be some sort of 040 testing suite:

https://dl.dropbox.com/u/2886976/AMIGA/m68k-tester.src.tar.bz2
https://dl.dropbox.com/u/2886976/AMIGA/m68k-tests.tar.bz2

[mikej]

Good advice.
I'll look into the test suite.
Ideally I would have something fairly automated that says go/no-go and does not take too long. May have a lot of devices to go through.

/MikeJ

[Cosmos Amiga]

Any pics of your 060 ?


This one for example is a fake :

[mikej]

Any pics of your 060 ?


This one for example is a fake :

I'll take some. Can you tell from the date code etc?

[wawrzon]

Good advice.
I'll look into the test suite.
Ideally I would have something fairly automated that says go/no-go and does not take too long. May have a lot of devices to go through.

/MikeJ

seems someone on natami forum have written some testcases, i might ask around if you dont want to do that yourself.

[Cosmos Amiga]

Can you tell from the date code etc?

Written on the 060 : 0124 = year 2001, week 24

[mikej]

seems someone on natami forum have written some testcases, i might ask around if you dont want to do that yourself.

Would be great if you could.

[hooverphonique]

I recently purchased an MC68040 L88M from chinese supplier utsource, which I believe to be fake...

[mikej]

Written on the 060 : 0124 = year 2001, week 24

Why is that a problem? Were they not manufacturing them at that point?

[mikej]

I recently purchased an MC68040 L88M from chinese supplier utsource, which I believe to be fake...

It is a risk. I have visited two suppliers and have an agreement to swap chips I don't like - so it is in their interest to give me good parts. However, I do need to come up with a decent test suite. Reading the PCN is easy and will validate the mask, however I still need to make sure the chip is fully functional.
/Mike
Edit, they are not fake but may be reprinted - for example mask set changed, or just broken.