Upshot: motarola planning to ship $1 flash-based part soon, AND:
"All of our new designs will be flash-based instead of OTP
because we can offer this memory storage at about the same
price as the one-time programmable solutions," said Kilbane.
I don't see how microchip can compete against that sort of thing unless
they get a lot more agressive with new parts, lower prices, and better
availability. (Motarola isn't alone, of course. Atmel and Scenix are
also essentially flash-only suppliers.)
On Tue, Oct 26, 1999 at 01:24:11PM -0700, William Chops Westfield wrote:
>
> Upshot: motarola planning to ship $1 flash-based part soon, AND:
Motorola's record on actually delivering parts to everyone who
wants them is pretty poor. Ask any 68HC11 user how many times they've
been told a particular part is on allocation.
Microchip's success has more to do with savvy marketing, a willingness
to deal with customers of any size, and the ability to actually deliver
parts most of the time than with price. Don't write them off just yet.
I think 'til Microchip will be available thru DigiKey for a few thousand
pieces, and have a friendly salesperson who willing to talk You even if You
do not order 10 million pieces they are not going to have any problem.
PeterSSubject: coming death of microchip?
> http://www.eet.com/story/industry/semiconductor_news/OEG19991018S0042
>
> Upshot: motarola planning to ship $1 flash-based part soon, AND:
>
> "All of our new designs will be flash-based instead of OTP
> because we can offer this memory storage at about the same
> price as the one-time programmable solutions," said Kilbane.
>
> I don't see how microchip can compete against that sort of thing unless
> they get a lot more agressive with new parts, lower prices, and better
> availability. (Motarola isn't alone, of course. Atmel and Scenix are
> also essentially flash-only suppliers.)
>
> BillW
> http://www.eet.com/story/industry/semiconductor_news/OEG19991018S0042
>
> Upshot: motarola planning to ship $1 flash-based part soon, AND:
>
> "All of our new designs will be flash-based instead of OTP
> because we can offer this memory storage at about the same
> price as the one-time programmable solutions," said Kilbane.
>
> I don't see how microchip can compete against that sort of thing unless
> they get a lot more agressive with new parts, lower prices, and better
> availability. (Motarola isn't alone, of course. Atmel and Scenix are
> also essentially flash-only suppliers.)
>
> BillW
Content-Type: text/x-vcard; charset=us-ascii;
name="briones.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Francisco Armenta
Content-Disposition: attachment;
filename="briones.vcf"
Consider also that flash is little or no advantage for many applications (such
as low-end mass produced trackballs and mice), so those applications aren't
going to rush away from otp & mask rom parts even if flash could be had for
the same price. But for the many applications that do benefit from flash,
microchip may be forced to become more competitive, either by dropping
prices or increasing features. I'm sure they are already feeling some
competitive pressure coming from parts that have performance and features
that PIC's still lack. The fast Scenix chips and the chips with built-in USB
from Philips and others come to mind.
I really hope mchip can turn their USB parts in production soon, and more
importantly, get USB functionality into flash parts with better than 8 bit a/d.
Their upcoming 16C7xx USB parts that are mentioned on their web site are
cool, but a 16F8xx part with USB would be a better bet. After all, USB is
primarily for PC peripherals, and the ability to reprogram such peripherals
in the field is an extremely desirable feature. Of course, flash won't mean
much for these applications unless there is some mechanism to flash them
over the USB connection itself. Maybe have both OTP and flash on the
same part so that code in the OTP can handle re-writing the flash from data
sent over USB. Or maybe something like two microcontrollers on a single
chip, one bare-bones one with OTP just to handle communications and flash
upgrading and a more feature-rich flashable one that is free to do other
things with minimal communications overhead. Major design cost I'm sure.
But I can dream can't I? And if Microchip could produce such a thing at a
reasonable cost, I don't think they'd have any shortage of customers for it.
And isn't this kind of co-processor function just the kind of thing that PICs
were created for in the first place?
> Microchip's success has more to do with savvy marketing, a willingness
> to deal with customers of any size, and the ability to actually deliver
> parts most of the time than with price. Don't write them off just yet.
Andy Kunz wrote:
>
> >Motorola's record on actually delivering parts to everyone who
> >wants them is pretty poor. Ask any 68HC11 user how many times they've
>
> Only those with "Motors" as part of the name seem to get them.
>
> Andy
I must agree with Clyde and Andy on the Motorola subject.
The reason I got into PICs was Motorola.
Robert Nansel has the Amateur Robotics column in Nuts & Volts magazine,
and when he first wrote about SimmSticks way back in Aug-98, he said:
There are currently no Motorola MCU chips such as the 68HC05 or 68HC11
supported by SimmSticks. This is a shame since these seem to be naturals
for SimmSticks, and they are generally more popular with us gearheads.
If Antti, or Don, or Ben don't design an HC11 SimmStick, I may have to
do one myself. (are you listening guys?)
I told Robert what I thought, and the problems I have had with Motorola
in the past, as I found prices and lead times became science fiction. If
you don't build phones or cars, forget them.
I haven't completely wiped the idea, but some one will have to do a good
job of convincing me otherwise.
At 07:48 27/10/99 +1000, you wrote:
>Andy Kunz wrote:
>>
>> >Motorola's record on actually delivering parts to everyone who
>> >wants them is pretty poor. Ask any 68HC11 user how many times they've
>>
>> Only those with "Motors" as part of the name seem to get them.
>>
>> Andy
>
>I must agree with Clyde and Andy on the Motorola subject.
>The reason I got into PICs was Motorola.
>
Also there is the "Vaporware" issue, we must be dilligent in choice and use
as often manufacturers test the water by anoucing some item, give an
intended delivery date, then wait for cusomers to ask about it.
If no questions, idem dies, if questions = big orders the silicon G-D gives
life to the creation = some happy customers
William Chops Westfield wrote:
>
> Just out of curiosity, has anyone had IMPROVED luck with motorola since their
> "No Excuses, Just delivered" OTP marketing campaign?
Mostly yes.
I haven't had trouble getting HC05P6As. Oddball parts like windowed
HC11P2s took a little more time (and money), but even that took less
than a week.
However, when I tried ordering parts over their web site, it turned
into a big fiasco. The local rep called to see what was going on,
even though the order was supposed to be going through a different
distributor. The local rep got me parts in short order. The on-line
order took a couple weeks. This was early in the roll-out of the
program, so it may have improved since then.
On Tue, Oct 26, 1999 at 04:23:25PM -0500, William J. Kitchen wrote:
> Consider also that flash is little or no advantage for many applications (such
> as low-end mass produced trackballs and mice), so those applications aren't
> going to rush away from otp & mask rom parts even if flash could be had for
> the same price.
The mask parts do not need programming, and the code memory does not deteriorate
with time so are better then flash and OTP parts for such applications.
But why OTP should be better suited for such apps than flash?
--
Greetings
Wojciech Zabolotny http://www.ise.pw.edu.pl/~wzab
> On Tue, Oct 26, 1999 at 04:23:25PM -0500, William J. Kitchen wrote:
> > Consider also that flash is little or no advantage for many applications
> (such
> > as low-end mass produced trackballs and mice), so those applications
> aren't
> > going to rush away from otp & mask rom parts even if flash could be had
> for
> > the same price.
>
Wojciech Zabolotny replied:
> The mask parts do not need programming, and the code memory does not
> deteriorate
> with time so are better then flash and OTP parts for such applications.
> But why OTP should be better suited for such apps than flash?
> --
>
I think the point William is making is that generally Flash parts have a
cost penalty, and if the application your PIC sits in would not benefit form
the advantages of flash, i.r. re-programability. When was the last time you
upgraded the firmware in a $3 Taiwanese mouse?
> I think the point William is making is that generally Flash parts have a
> cost penalty, and if the application your PIC sits in would not benefit form
> the advantages of flash, i.r. re-programability. When was the last time you
> upgraded the firmware in a $3 Taiwanese mouse?
That's it exactly. There are applications in which flash is of little or no
practical advantage, so cost would be the deciding factor. The flash parts
would have to become cheaper than OTP to be worthwhile in applications
such as this. If they were the same cost, then they would be an equally
good choice for totally new development, but there would still be no incentive
to convert existing designs or for a developer with existing PIC experience to
start designing around a different architecture.
htsoft.com | Phone Fax
