All,

 I use to work for TI who came up with Bubble Memory, and there
 were four main factors effecting it's downfall.  I've listed
 them here.

 1. Complexity......There were several drive circuits that were
                    critical to proper operation of the memory,
                    and these circuits were complex and difficult
                    to get just right.

 2. Physical size...As memories go, Bubble memory was large in
                    comparison because of the drive circuits and
                    the coils that were required for operation.

 3. Density.........Again as memories go, the bit density of the
                    bubble memories just wasn't able to compete.
                    In the beginning, the density was equal or
                    greater than than of semiconductor memory.
                    But as memories increased in size, the bubble
                    memory just couldn't keep pace.

 4. Cost............This should come as no surprise.  This is
                    always a factor in everything. But the cost
                    of further development just couldn't be
                    recovered in time to make it a viable memory
                    contender in the "more is better" computer
                    industry.


  Hope this helps you understand.

                                         Regards,

                                           Jim





On Sun, 31 October 1999, "Paul B. Webster VK2BZC" wrote:

>
> Eric Richards wrote:
>
> > What ever happened to bubble memory that disappeared faster than it
> > came?
>
>   It seems - the bubble burst!?
>
>   I suspect the actual problem was that it required just too much
> engineering per bit - a material with a pattern on it designating the
> bits.  And I would say the domains couldn't be made small enough.  Just
> too complex.
>
>   On a hard disk, the disk is a uniform surface, the mechanics whilst
> moderately complex, define a HUGE data array on this simple surface.
> With bubble memory OTOH, bigger memories meant more complexity of the
> wafer, bigger wafers etc.
>
>   Now while this is certainly true of silicon memory (DRAM), I'd say
> there was just no way they'd ever get the cell size for bubble memory
> anywhere near the scale of silicon memory, despite what they may have
> originally imagined.
>
>   So you have a structure that has the internal complexity of silicon
> memory, *plus* the external complexity (magnets, coils, shielding) and
> more of magnetic memory.
>
>   Purely conjecture on my part of course...
> --
>   Cheers,
>         Paul B.

jim@jpes.com