[comp.sys.amiga] A3000 Upgrading 16 to 25MHz

sparks@corpane.UUCP (John Sparks) (05/02/90)

lphillips@lpami.wimsey.bc.ca (Larry Phillips) writes:
>68030 and 68882 are surface mount chips. Not exactly an easy hobbyist
>replacement.

erk! not exactly an easy professional replacement job either! So what happens
if your CPU or FPU bites the dust? I don't think my local dealer is ready
to perform such a difficult task as replacing surface mount chips. 

I wonder what happens? do they swap the mother board or ship it off to
commodore for fixing?

Sheesh, I would have at least thought all the major chips would be in sockets.

I guess CBM figures it's worth gambling the more expensive repair costs on
the units that do break down against the definite increased costs of adding
sockets for all chips in all units.


-- 
John Sparks  | D.I.S.K. 24hrs 2400bps. Accessable via Starlink (Louisville KY)
sparks@corpane.UUCP |                                     | PH: (502) 968-DISK  
I'm the person your mother warned you about.

daveh@cbmvax.commodore.com (Dave Haynie) (05/04/90)

In article <1758@corpane.UUCP> sparks@corpane.UUCP (John Sparks) writes:
>lphillips@lpami.wimsey.bc.ca (Larry Phillips) writes:
>>68030 and 68882 are surface mount chips. Not exactly an easy hobbyist
>>replacement.

>erk! not exactly an easy professional replacement job either! So what happens
>if your CPU or FPU bites the dust? 

Never happens.  Well, at least almost never.  Heck, I've plugged PGA '030s
into A2630s in all four directions, and while the machine only works when
the chip's plugging in the preferred direction, I've never killed a 68030.

>Sheesh, I would have at least thought all the major chips would be in sockets.

The criteria for socketing chips:

	[A] Do they fail?  (Example: 8520s, which can get zapped since they
	    go to external ports).
	[B] Are upgrades likely?  (Example: 512K Fat Agnus -> 1 Meg Fat Agnus).
	[C] Are they new custom chips?  (Because these change considerably
	    during development and even sometimes into beta testing).

The criteria for not socketing chips:

	[A] Noise!  Socketed parts are's as quiet as soldered in parts, and
	    surface mount can get especially quiet.
	[B] Price!  Surface mount vs. PGA can be a big savings.

>I guess CBM figures it's worth gambling the more expensive repair costs on
>the units that do break down against the definite increased costs of adding
>sockets for all chips in all units.

While the PLCC 68881/2 can be socketed, the plastic '030 we're usings can't
be socketed, it has to be surface mounted.  You can get PGA '030s, of course,
like the one on the A2630, but they're more costly.  The repair advantage is
lost if the part never fails.

>John Sparks  | D.I.S.K. 24hrs 2400bps. Accessable via Starlink (Louisville KY)
>sparks@corpane.UUCP |                                     | PH: (502) 968-DISK  
>I'm the person your mother warned you about.


-- 
Dave Haynie Commodore-Amiga (Amiga 3000) "The Crew That Never Rests"
   {uunet|pyramid|rutgers}!cbmvax!daveh      PLINK: hazy     BIX: hazy
	"I have been given the freedom to do as I see fit" -REM