Searching \ for '[PIC] RE: PIC Second Source' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/devices.htm?key=pic
Search entire site for: 'RE: PIC Second Source'.

Exact match. Not showing close matches.
PICList Thread
'[PIC] RE: PIC Second Source'
2006\04\27@002349 by Picdude

flavicon
face
Have you tried Microchip direct?  Also, check with brokers -- in my experience, they can get better prices for larger quantities, unless it is in short supply.

Cheers,
-Neil.


> ---{Original Message removed}

2006\04\27@014848 by William Chops Westfield

face picon face
>> I am looking at large quantities, amd at the moment I can source
>>  them at around AU$1, but if I can fins an manufacturer in Asia
>> or somewhere that can supply at a much better price I would be
>>  very happy. I heard a possibility of something in Taiwan, but
>>  am having trouble getting any leads on that...

So you're explicitly soliciting pirated intellectual property?
AFAIK, Microchip hasn't licensed their chips to any second sources,
so any exact equivalents not from microchip are pretty much by
definition "stolen."

I've heard that there are some chips available in pre-tested form
at some discount, aimed at competing with clone vendors...

BillW

2006\04\27@150733 by M. Adam Davis

face picon face
On 4/27/06, William Chops Westfield <spam_OUTwestfwTakeThisOuTspammac.com> wrote:
> So you're explicitly soliciting pirated intellectual property?
> AFAIK, Microchip hasn't licensed their chips to any second sources,
> so any exact equivalents not from microchip are pretty much by
> definition "stolen."

I'm interested to hear the details about this intellectual property.
What patents or copyrights are being violated if I do a clean room
reverse engineering of the PICMicro in question, then start producing
it as a second source?

Until Intel started protecting themselves with patents (specifically
the memory access patterns in the pentium chip) other chip makers were
free to reverse engineer their chips and clone them.

Of course this is different than simply analyzing the die and
producing an exact copy - that  breaks microchip's copyright.  A clean
room reverse engineering, though, could only break patents (many, if
not most, of which could be worked around).

The market isn't that big for clones in this segment.  If you want
cheap you get different 8-bit processor from China, or use a 4 bit
processor (no flash!  It's expensive!).  If you want something that is
second sourcable you use an 8051 or similar "old architecture" chip.
If you don't care about pricing or availability then there's no need
for a clone.

Clone makers are getting in trouble for producing clones and then
marking and selling them as if they were from microchip.

I imagine there are legal clones that exist, but they aren't well
documented.  Typically you'd talk to your manufacturer and they'd make
suggestions about parts they've used previously.  Or you talk to an
electronics broker based in asia.  Perhaps a careful reading of the
lawsuits microchip files.

But if you really want *cheap* cheap you don't use microchip, or you
use enough of them that the VP of sales takes you out for frequent
business lunches.

-Adam

2006\04\27@151835 by Bernd Rüter

flavicon
face
part 0 44 bytes
his is a multi-part message in MIME format.
part 1 833 bytes content-type:text/plain; charset=ISO-8859-1; format=flowed (decoded 7bit)

>>So you're explicitly soliciting pirated intellectual property?
>>AFAIK, Microchip hasn't licensed their chips to any second sources,
>>so any exact equivalents not from microchip are pretty much by
>>definition "stolen."
>
>
> I'm interested to hear the details about this intellectual property.
> What patents or copyrights are being violated if I do a clean room
> reverse engineering of the PICMicro in question, then start producing
> it as a second source?


"Second source" was born by the industry after the US-army only bought
products made of parts from several suppliers. So a lot of
chip-producers made contracts to give their plans to there partners.

Second source is not "robbery" it was well planned...


part 2 335 bytes content-type:text/x-vcard; charset=utf-8;
(decoded 7bit)

begin:vcard
fn;quoted-printable:Bernd R=C3=BCter
n;quoted-printable:R=C3=BCter;Bernd
adr;quoted-printable;dom:;;W=C3=BClferoder Stra=C3=9Fe 10;Hannover;;30539
email;internet:.....postKILLspamspam@spam@bernd-rueter.de
tel;home:+49 511 7636990
tel;cell:+49 163 2355583
url:http://www.bernd-rueter.de
version:2.1
end:vcard



part 3 35 bytes content-type:text/plain; charset="us-ascii"
(decoded 7bit)

2006\04\27@153840 by Maarten Hofman

face picon face
>
> I'm interested to hear the details about this intellectual property.
> What patents or copyrights are being violated if I do a clean room
> reverse engineering of the PICMicro in question, then start producing
> it as a second source?


Well, one of the more interesting patents that Microchip has is the "n-bit
databus in n-pin package". Which means that if your microcontroller has a
databus of 8 bits, then you can't put it into a package of only 8 pins (or
less, I believe, if I read the patent correctly... I had fun looking it up
earlier this week). This basically protects all 8-pin (and smaller) that
Microchip makes. Given the fact that they have that particular patent, I am
pretty certain that they have other patents as well.

As for second sources, I am quite certain there are a number of
manufacturers that make Microchip products under license. I believe there
are several in China and Brazil, also because these countries wouldn't allow
Microchip to sell to them unless they had such a construct in place. I'm not
certain whether these manufacturers would sell outside of their target
market, though.

Greetings,
Maarten Hofman.

2006\04\27@164222 by Bob Axtell

face picon face
I was told last year than a small Chinese company was making low-level
PICs (like
PIC16C54) for sales only inside China. I never saw samples, though.

--Bob

M. Adam Davis wrote:
{Quote hidden}

2006\04\27@193045 by Mike Hord

picon face
> Well, one of the more interesting patents that Microchip has is the "n-bit
> databus in n-pin package". Which means that if your microcontroller has a
> databus of 8 bits, then you can't put it into a package of only 8 pins (or
> less, I believe, if I read the patent correctly... I had fun looking it up
> earlier this week). This basically protects all 8-pin (and smaller) that
> Microchip makes. Given the fact that they have that particular patent, I am
> pretty certain that they have other patents as well.

Worse:  if you want to put a powerful 32-bit core into a 28-pin
package, that patent will let Microchip sue you!  Or a 16-bit into
14-pin package, etc.

<rant>
In fact, that's how that patent was brought up on the list:  they ARE
suing someone for putting a 32-bit core into a 28-pin package.
Exactly the kind of IP BS shell game that drives me nuts and allows
companies to use our political system to circumvent true competition
and subvert innovation.
</rant>

After all, if you can stick a 32-bit ARM proc in where previously only
an 8-bit PIC or AVR would fit, that gives you access to a whole new
realm of power, doesn't it?  But since Microchip owns the patent, your
choices are limited to the 8-bit or to reworking your design to fit
some monster part plus it's NVRAM and SRAM in.

Grrrrrrrrr.

Mike H.

2006\04\28@031510 by William Chops Westfield

face picon face

On Apr 27, 2006, at 12:07 PM, M. Adam Davis wrote:

>> So you're explicitly soliciting pirated intellectual property?
>> AFAIK, Microchip hasn't licensed their chips to any second sources,
>> so any exact equivalents not from microchip are pretty much by
>> definition "stolen."
>
> I'm interested to hear the details about this intellectual property.
> What patents or copyrights are being violated if I do a clean room
> reverse engineering of the PICMicro in question, then start producing
> it as a second source?

An interesting question.  Intel copyrighted the instruction set
mnemonics of the 8080, which is why Zilog used different names for
the same instructions when they did their clones.  You can do
"design patents" for a particular design of a thing, but I don't
know whether that's been applied to chips.  Industry standard
practice is to patent, copyright, and/or trademark everything you
can get away with, just to act as deterrent for this sort of
cloning.  I wouldn't expect someone to be able to produce a close
clone of any modern processor without opening themselves up to
legal action by trampling on any number of the protective measures.
(and I'm talking real infringement of real deserved IP, not just
phony patents-that-never-should-have-been like Nbit processors on
in less than N pin packages...)  (otherwise, I'd expect there to
be a lot more clones in the world.  (although, most companies capable
of doing the manufacturing seem to think they can make a better
processor anyway...))

>
> Until Intel started protecting themselves with patents (specifically
> the memory access patterns in the pentium chip) other chip makers were
> free to reverse engineer their chips and clone them.
>
Actually, Intel LICENSED the early x86 designs quite broadly.  The
early AMD, Harris, National Geode, x186s from all over, and so on
were not reverse engineered clones, they were licensed second sources
and enhancements.  I guess it was about pentium time that they
stopped the licensing and forced their former partners to either
branch off on their own or fall behind...

BillW

2006\04\28@034025 by Alan B. Pearce

face picon face
>This basically protects all 8-pin (and smaller) that Microchip
>makes. Given the fact that they have that particular patent,
>I am pretty certain that they have other patents as well.

Oh they sure have. A dose of samples that arrived yesterday has paperwork
that list a total of 231 patent numbers. Now even allowing that some of
these cover things like Keeloq and other such devices outside the
microprocessor range, there is probably still somewhere around 1/3 to 1/2 of
them related to microprocessors.

2006\04\28@034735 by Alan B. Pearce

face picon face
>Actually, Intel LICENSED the early x86 designs quite
>broadly.  The early AMD, Harris, National Geode, x186s
>from all over, and so on were not reverse engineered
>clones, they were licensed second sources and enhancements.
>I guess it was about pentium time that they stopped
>the licensing and forced their former partners to either
>branch off on their own or fall behind...

I believe it was about the time that AMD started with the 586 series chips
that would drop into 486 sockets, but had higher clock multipliers and so
on, and Intel saw that they were loosing the "lead in speed".

Somewhere in the 486 generation AMD got some engineers in and was making 486
chips that no longer needed licensing, and went on to the 586. Then Intel
found they could not trademark a number, and came up with the Pentium name
that they could trademark.

2006\04\28@035758 by Alan B. Pearce

face picon face
> I'm interested to hear the details about this intellectual property.
> What patents or copyrights are being violated if I do a clean room
> reverse engineering of the PICMicro in question, then start producing
> it as a second source?

Well, there is an open source IP for a simplified 16F84 on
http://www.opencores.org/ and I believe someone there was working on an 18F
core.


2006\04\28@175947 by Peter Todd

picon face
On Fri, Apr 28, 2006 at 12:15:09AM -0700, William Chops Westfield wrote:
> An interesting question.  Intel copyrighted the instruction set
> mnemonics of the 8080, which is why Zilog used different names for
> the same instructions when they did their clones.  You can do
> "design patents" for a particular design of a thing, but I don't
> know whether that's been applied to chips.  Industry standard
> practice is to patent, copyright, and/or trademark everything you
> can get away with, just to act as deterrent for this sort of
> cloning.  I wouldn't expect someone to be able to produce a close
> clone of any modern processor without opening themselves up to
> legal action by trampling on any number of the protective measures.
> (and I'm talking real infringement of real deserved IP, not just
> phony patents-that-never-should-have-been like Nbit processors on
> in less than N pin packages...)  (otherwise, I'd expect there to
> be a lot more clones in the world.  (although, most companies capable
> of doing the manufacturing seem to think they can make a better
> processor anyway...))

I'll add that one interesting thing about copyright law is that the
*masks* for processors and other ICs can be copyrighted as you would
expect, but the length of the term is only 10 years rather than the
usual "basically for eternity"

http://www.copyright.gov/circs/circ100.html

Of course, as you say above, lots of patents and what not to consider...
But without patents, sounds like you could just print off some 10 year
old chips and no-one could do a thing about it.

--
.....peteKILLspamspam.....petertodd.ca http://www.petertodd.ca

2006\04\29@025714 by William Chops Westfield

face picon face

On Apr 28, 2006, at 3:00 PM, Peter Todd wrote:

> But without patents, sounds like you could just print off some
> 10 year old chips and no-one could do a thing about it.
>
I think I pointed out some time ago that the Intel patents covering
the basic x86 architecture are going to expire soon.  I'm not really
sure what that means, given that there doesn't seem to be that much
of an x86 clone shortage now, and no one really wants an 8086 at
this date and time (um.  Maybe.  Let's upgrade all those 8051 chips
to have 8086 cores instead...)  But we could be headed for
interesting times...

BillW

More... (looser matching)
- Last day of these posts
- In 2006 , 2007 only
- Today
- New search...