Site Home   Archive Home   FAQ Home   How to search the Archive   How to Navigate the Archive   
Compare FPGA features and resources   

Threads starting:
1994JulAugSepOctNovDec1994
1995JanFebMarAprMayJunJulAugSepOctNovDec1995
1996JanFebMarAprMayJunJulAugSepOctNovDec1996
1997JanFebMarAprMayJunJulAugSepOctNovDec1997
1998JanFebMarAprMayJunJulAugSepOctNovDec1998
1999JanFebMarAprMayJunJulAugSepOctNovDec1999
2000JanFebMarAprMayJunJulAugSepOctNovDec2000
2001JanFebMarAprMayJunJulAugSepOctNovDec2001
2002JanFebMarAprMayJunJulAugSepOctNovDec2002
2003JanFebMarAprMayJunJulAugSepOctNovDec2003
2004JanFebMarAprMayJunJulAugSepOctNovDec2004
2005JanFebMarAprMayJunJulAugSepOctNovDec2005
2006JanFebMarAprMayJunJulAugSepOctNovDec2006
2007JanFebMarAprMayJunJulAugSepOctNovDec2007
2008JanFebMarAprMayJunJulAugSepOctNovDec2008
2009JanFebMarAprMayJunJulAugSepOctNovDec2009
2010JanFebMarAprMayJunJulAugSepOctNovDec2010
2011JanFebMarAprMayJunJulAugSepOctNovDec2011
2012JanFebMarAprMayJunJulAugSepOctNovDec2012
2013JanFebMarAprMayJunJulAugSepOctNovDec2013
2014JanFebMarAprMayJunJulAugSepOctNovDec2014
2015JanFebMarAprMayJunJulAugSepOctNovDec2015
2016JanFebMarAprMayJunJulAugSepOctNovDec2016
2017JanFebMarAprMayJunJulAugSepOctNovDec2017
2018JanFebMarAprMayJunJulAugSepOctNovDec2018
2019JanFebMarAprMayJunJulAugSepOctNovDec2019
2020JanFebMarAprMay2020

Authors:A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Custom Search

Messages from 161525

Article: 161525
Subject: Re: New coding method for a state machine in groups in HDL
From: Andy Bennet <andyb@andy.com>
Date: Tue, 26 Nov 2019 07:26:55 +0000
Links: << >>  << T >>  << A >>
On 25/11/2019 21:27, Weng Tianxiang wrote:
> Hi everyone,
> 
> Welcome all critics from ones who are interested in coding state machine that seems to many as matured, but  can be further improved.
> 
> Here is a coding snippet for new method you can immediately understand what will happening for coding a state machine.
> 
> type State_Machine_t is (
>    First_group : (s1, s2, s3),
>    Second_Group : (s4, s5, s6),
>    Third_Group ; (s7, s8, s9)
> );
> signal State_Machine, State_Machine_Next : State_Machine_t ;
> 
> Here is only one sentence that tells you what you can do with new method:
> Any state machine's states can be grouped together in coding in HDL at one's discretion and each group acts like a small state machine sharing a same initial state and all act together as coordinated as a full state machine.
> 
> Here is a famous paper about the method with 244 cites, using probability theory:
>   http://www.scarpaz.com/2100-papers/Low%20Power/00503933.pdf
> 
> But my patent describes a new simpler and perfect method using group theory:
> 
> Patent:10482208, "Coding and synthesizing a state machine in state groups"
> 
> https://patents.justia.com/search?q=tianxiang+weng
> 
> Thank KJ for helping me to improve the patent text.
> 
> Thank you.
> 
> Weng
> 

So presumably as you have it patented, you will be requiring money for 
anyone to use it?


Article: 161526
Subject: Re: New coding method for a state machine in groups in HDL
From: rob dickinson <rjd4567@gmail.com>
Date: Tue, 26 Nov 2019 04:39:59 -0800 (PST)
Links: << >>  << T >>  << A >>
On Monday, November 25, 2019 at 9:27:41 PM UTC, Weng Tianxiang wrote:
> Hi everyone,
> 
> Welcome all critics from ones who are interested in coding state machine that seems to many as matured, but  can be further improved.
> 
> Here is a coding snippet for new method you can immediately understand what will happening for coding a state machine.
> 
> type State_Machine_t is (
>   First_group : (s1, s2, s3),
>   Second_Group : (s4, s5, s6),
>   Third_Group ; (s7, s8, s9)
> );
> signal State_Machine, State_Machine_Next : State_Machine_t ; 
> 
> Here is only one sentence that tells you what you can do with new method:
> Any state machine's states can be grouped together in coding in HDL at one's discretion and each group acts like a small state machine sharing a same initial state and all act together as coordinated as a full state machine.
> 
> Here is a famous paper about the method with 244 cites, using probability theory:
>  http://www.scarpaz.com/2100-papers/Low%20Power/00503933.pdf
> 
> But my patent describes a new simpler and perfect method using group theory:
> 
> Patent:10482208, "Coding and synthesizing a state machine in state groups"
> 
> https://patents.justia.com/search?q=tianxiang+weng 
> 
> Thank KJ for helping me to improve the patent text.
> 
> Thank you. 
> 
> Weng

Good luck proving that I have never done this before.

Article: 161527
Subject: Re: New coding method for a state machine in groups in HDL
From: Michael Kellett <mk@mkesc.co.uk>
Date: Tue, 26 Nov 2019 12:59:34 +0000
Links: << >>  << T >>  << A >>
On 26/11/2019 12:39, rob dickinson wrote:
> On Monday, November 25, 2019 at 9:27:41 PM UTC, Weng Tianxiang wrote:
>> Hi everyone,
>>
>> Welcome all critics from ones who are interested in coding state machine that seems to many as matured, but  can be further improved.
>>
>> Here is a coding snippet for new method you can immediately understand what will happening for coding a state machine.
>>
>> type State_Machine_t is (
>>    First_group : (s1, s2, s3),
>>    Second_Group : (s4, s5, s6),
>>    Third_Group ; (s7, s8, s9)
>> );
>> signal State_Machine, State_Machine_Next : State_Machine_t ;
>>
>> Here is only one sentence that tells you what you can do with new method:
>> Any state machine's states can be grouped together in coding in HDL at one's discretion and each group acts like a small state machine sharing a same initial state and all act together as coordinated as a full state machine.
>>
>> Here is a famous paper about the method with 244 cites, using probability theory:
>>   http://www.scarpaz.com/2100-papers/Low%20Power/00503933.pdf
>>
>> But my patent describes a new simpler and perfect method using group theory:
>>
>> Patent:10482208, "Coding and synthesizing a state machine in state groups"
>>
>> https://patents.justia.com/search?q=tianxiang+weng
>>
>> Thank KJ for helping me to improve the patent text.
>>
>> Thank you.
>>
>> Weng
> 
> Good luck proving that I have never done this before.
> 

The patent has been granted -  so I think it may be up to you to prove 
that you have :-)

(Would you have wanted to ?)

MK

Article: 161528
Subject: Re: New coding method for a state machine in groups in HDL
From: Weng Tianxiang <wtxwtx@gmail.com>
Date: Tue, 26 Nov 2019 05:50:18 -0800 (PST)
Links: << >>  << T >>  << A >>
On Tuesday, November 26, 2019 at 4:59:38 AM UTC-8, Michael Kellett wrote:
> On 26/11/2019 12:39, rob dickinson wrote:
> > On Monday, November 25, 2019 at 9:27:41 PM UTC, Weng Tianxiang wrote:
> >> Hi everyone,
> >>
> >> Welcome all critics from ones who are interested in coding state machine that seems to many as matured, but  can be further improved.
> >>
> >> Here is a coding snippet for new method you can immediately understand what will happening for coding a state machine.
> >>
> >> type State_Machine_t is (
> >>    First_group : (s1, s2, s3),
> >>    Second_Group : (s4, s5, s6),
> >>    Third_Group ; (s7, s8, s9)
> >> );
> >> signal State_Machine, State_Machine_Next : State_Machine_t ;
> >>
> >> Here is only one sentence that tells you what you can do with new method:
> >> Any state machine's states can be grouped together in coding in HDL at one's discretion and each group acts like a small state machine sharing a same initial state and all act together as coordinated as a full state machine.
> >>
> >> Here is a famous paper about the method with 244 cites, using probability theory:
> >>   http://www.scarpaz.com/2100-papers/Low%20Power/00503933.pdf
> >>
> >> But my patent describes a new simpler and perfect method using group theory:
> >>
> >> Patent:10482208, "Coding and synthesizing a state machine in state groups"
> >>
> >> https://patents.justia.com/search?q=tianxiang+weng
> >>
> >> Thank KJ for helping me to improve the patent text.
> >>
> >> Thank you.
> >>
> >> Weng
> > 
> > Good luck proving that I have never done this before.
> > 
> 
> The patent has been granted -  so I think it may be up to you to prove 
> that you have :-)
> 
> (Would you have wanted to ?)
> 
> MK

Hi MK,
I don't understand your saying "I think it may be up to you to prove 
> that you have".

It's USPTO official publication, what do I have to prove?

I have patented not the coding method, but on how to generate the specific circuit that can be grouped with easiest coding method on mind.

Andy,
As an independent inventor, I want to sell the patent to a company to get financial support for my next tens of new exciting inventions.

Weng

Article: 161529
Subject: Re: New coding method for a state machine in groups in HDL
From: Rick C <gnuarm.deletethisbit@gmail.com>
Date: Tue, 26 Nov 2019 07:44:20 -0800 (PST)
Links: << >>  << T >>  << A >>
On Tuesday, November 26, 2019 at 8:50:23 AM UTC-5, Weng Tianxiang wrote:
> On Tuesday, November 26, 2019 at 4:59:38 AM UTC-8, Michael Kellett wrote:
> > On 26/11/2019 12:39, rob dickinson wrote:
> > > On Monday, November 25, 2019 at 9:27:41 PM UTC, Weng Tianxiang wrote:
> > >> Hi everyone,
> > >>
> > >> Welcome all critics from ones who are interested in coding state machine that seems to many as matured, but  can be further improved.
> > >>
> > >> Here is a coding snippet for new method you can immediately understand what will happening for coding a state machine.
> > >>
> > >> type State_Machine_t is (
> > >>    First_group : (s1, s2, s3),
> > >>    Second_Group : (s4, s5, s6),
> > >>    Third_Group ; (s7, s8, s9)
> > >> );
> > >> signal State_Machine, State_Machine_Next : State_Machine_t ;
> > >>
> > >> Here is only one sentence that tells you what you can do with new method:
> > >> Any state machine's states can be grouped together in coding in HDL at one's discretion and each group acts like a small state machine sharing a same initial state and all act together as coordinated as a full state machine.
> > >>
> > >> Here is a famous paper about the method with 244 cites, using probability theory:
> > >>   http://www.scarpaz.com/2100-papers/Low%20Power/00503933.pdf
> > >>
> > >> But my patent describes a new simpler and perfect method using group theory:
> > >>
> > >> Patent:10482208, "Coding and synthesizing a state machine in state groups"
> > >>
> > >> https://patents.justia.com/search?q=tianxiang+weng
> > >>
> > >> Thank KJ for helping me to improve the patent text.
> > >>
> > >> Thank you.
> > >>
> > >> Weng
> > > 
> > > Good luck proving that I have never done this before.
> > > 
> > 
> > The patent has been granted -  so I think it may be up to you to prove 
> > that you have :-)
> > 
> > (Would you have wanted to ?)
> > 
> > MK
> 
> Hi MK,
> I don't understand your saying "I think it may be up to you to prove 
> > that you have".
> 
> It's USPTO official publication, what do I have to prove?
> 
> I have patented not the coding method, but on how to generate the specific circuit that can be grouped with easiest coding method on mind.
> 
> Andy,
> As an independent inventor, I want to sell the patent to a company to get financial support for my next tens of new exciting inventions.
> 
> Weng

Please let us know when you have success in selling the patent.  I assume you would go for a royalty rather than a single payment, or perhaps a combination.  

How would you know if a company is using your patented idea in their tools?  Enforcement might be the hard part.  Heck, how do you know companies aren't using the idea already? 

-- 

  Rick C.

  - Get 1,000 miles of free Supercharging
  - Tesla referral code - https://ts.la/richard11209

Article: 161530
Subject: Re: New coding method for a state machine in groups in HDL
From: Weng Tianxiang <wtxwtx@gmail.com>
Date: Tue, 26 Nov 2019 10:15:49 -0800 (PST)
Links: << >>  << T >>  << A >>
On Tuesday, November 26, 2019 at 7:44:24 AM UTC-8, Rick C wrote:
> On Tuesday, November 26, 2019 at 8:50:23 AM UTC-5, Weng Tianxiang wrote:
> > On Tuesday, November 26, 2019 at 4:59:38 AM UTC-8, Michael Kellett wrot=
e:
> > > On 26/11/2019 12:39, rob dickinson wrote:
> > > > On Monday, November 25, 2019 at 9:27:41 PM UTC, Weng Tianxiang wrot=
e:
> > > >> Hi everyone,
> > > >>
> > > >> Welcome all critics from ones who are interested in coding state m=
achine that seems to many as matured, but  can be further improved.
> > > >>
> > > >> Here is a coding snippet for new method you can immediately unders=
tand what will happening for coding a state machine.
> > > >>
> > > >> type State_Machine_t is (
> > > >>    First_group : (s1, s2, s3),
> > > >>    Second_Group : (s4, s5, s6),
> > > >>    Third_Group ; (s7, s8, s9)
> > > >> );
> > > >> signal State_Machine, State_Machine_Next : State_Machine_t ;
> > > >>
> > > >> Here is only one sentence that tells you what you can do with new =
method:
> > > >> Any state machine's states can be grouped together in coding in HD=
L at one's discretion and each group acts like a small state machine sharin=
g a same initial state and all act together as coordinated as a full state =
machine.
> > > >>
> > > >> Here is a famous paper about the method with 244 cites, using prob=
ability theory:
> > > >>   http://www.scarpaz.com/2100-papers/Low%20Power/00503933.pdf
> > > >>
> > > >> But my patent describes a new simpler and perfect method using gro=
up theory:
> > > >>
> > > >> Patent:10482208, "Coding and synthesizing a state machine in state=
 groups"
> > > >>
> > > >> https://patents.justia.com/search?q=3Dtianxiang+weng
> > > >>
> > > >> Thank KJ for helping me to improve the patent text.
> > > >>
> > > >> Thank you.
> > > >>
> > > >> Weng
> > > >=20
> > > > Good luck proving that I have never done this before.
> > > >=20
> > >=20
> > > The patent has been granted -  so I think it may be up to you to prov=
e=20
> > > that you have :-)
> > >=20
> > > (Would you have wanted to ?)
> > >=20
> > > MK
> >=20
> > Hi MK,
> > I don't understand your saying "I think it may be up to you to prove=20
> > > that you have".
> >=20
> > It's USPTO official publication, what do I have to prove?
> >=20
> > I have patented not the coding method, but on how to generate the speci=
fic circuit that can be grouped with easiest coding method on mind.
> >=20
> > Andy,
> > As an independent inventor, I want to sell the patent to a company to g=
et financial support for my next tens of new exciting inventions.
> >=20
> > Weng
>=20
> Please let us know when you have success in selling the patent.  I assume=
 you would go for a royalty rather than a single payment, or perhaps a comb=
ination. =20
>=20
> How would you know if a company is using your patented idea in their tool=
s?  Enforcement might be the hard part.  Heck, how do you know companies ar=
en't using the idea already?=20
>=20
> --=20
>=20
>   Rick C.
>=20
>   - Get 1,000 miles of free Supercharging
>   - Tesla referral code - https://ts.la/richard11209

Rick,

What I have done is trivial and not important, and I have patented them for=
 living. What is important is what I am now developing and the final techno=
logy will be used by everyone, cellphones, computers and supercomputers, bu=
t not for industrial product controllers.

"Please let us know when you have success in selling the patent. " OK.


As an individual inventor, I have no resources to detect any situations as =
you described: How would you know if a company is using your patented idea =
in their tools?

What I want to do is to sell the patent once and get money to support my ne=
xt project that will benefit everyone in the world.

Article: 161531
Subject: Re: New coding method for a state machine in groups in HDL
From: Rick C <gnuarm.deletethisbit@gmail.com>
Date: Tue, 26 Nov 2019 11:03:18 -0800 (PST)
Links: << >>  << T >>  << A >>
On Tuesday, November 26, 2019 at 1:15:54 PM UTC-5, Weng Tianxiang wrote:
> On Tuesday, November 26, 2019 at 7:44:24 AM UTC-8, Rick C wrote:
> > On Tuesday, November 26, 2019 at 8:50:23 AM UTC-5, Weng Tianxiang wrote=
:
> > > On Tuesday, November 26, 2019 at 4:59:38 AM UTC-8, Michael Kellett wr=
ote:
> > > > On 26/11/2019 12:39, rob dickinson wrote:
> > > > > On Monday, November 25, 2019 at 9:27:41 PM UTC, Weng Tianxiang wr=
ote:
> > > > >> Hi everyone,
> > > > >>
> > > > >> Welcome all critics from ones who are interested in coding state=
 machine that seems to many as matured, but  can be further improved.
> > > > >>
> > > > >> Here is a coding snippet for new method you can immediately unde=
rstand what will happening for coding a state machine.
> > > > >>
> > > > >> type State_Machine_t is (
> > > > >>    First_group : (s1, s2, s3),
> > > > >>    Second_Group : (s4, s5, s6),
> > > > >>    Third_Group ; (s7, s8, s9)
> > > > >> );
> > > > >> signal State_Machine, State_Machine_Next : State_Machine_t ;
> > > > >>
> > > > >> Here is only one sentence that tells you what you can do with ne=
w method:
> > > > >> Any state machine's states can be grouped together in coding in =
HDL at one's discretion and each group acts like a small state machine shar=
ing a same initial state and all act together as coordinated as a full stat=
e machine.
> > > > >>
> > > > >> Here is a famous paper about the method with 244 cites, using pr=
obability theory:
> > > > >>   http://www.scarpaz.com/2100-papers/Low%20Power/00503933.pdf
> > > > >>
> > > > >> But my patent describes a new simpler and perfect method using g=
roup theory:
> > > > >>
> > > > >> Patent:10482208, "Coding and synthesizing a state machine in sta=
te groups"
> > > > >>
> > > > >> https://patents.justia.com/search?q=3Dtianxiang+weng
> > > > >>
> > > > >> Thank KJ for helping me to improve the patent text.
> > > > >>
> > > > >> Thank you.
> > > > >>
> > > > >> Weng
> > > > >=20
> > > > > Good luck proving that I have never done this before.
> > > > >=20
> > > >=20
> > > > The patent has been granted -  so I think it may be up to you to pr=
ove=20
> > > > that you have :-)
> > > >=20
> > > > (Would you have wanted to ?)
> > > >=20
> > > > MK
> > >=20
> > > Hi MK,
> > > I don't understand your saying "I think it may be up to you to prove=
=20
> > > > that you have".
> > >=20
> > > It's USPTO official publication, what do I have to prove?
> > >=20
> > > I have patented not the coding method, but on how to generate the spe=
cific circuit that can be grouped with easiest coding method on mind.
> > >=20
> > > Andy,
> > > As an independent inventor, I want to sell the patent to a company to=
 get financial support for my next tens of new exciting inventions.
> > >=20
> > > Weng
> >=20
> > Please let us know when you have success in selling the patent.  I assu=
me you would go for a royalty rather than a single payment, or perhaps a co=
mbination. =20
> >=20
> > How would you know if a company is using your patented idea in their to=
ols?  Enforcement might be the hard part.  Heck, how do you know companies =
aren't using the idea already?=20
> >=20
> > --=20
> >=20
> >   Rick C.
> >=20
> >   - Get 1,000 miles of free Supercharging
> >   - Tesla referral code - https://ts.la/richard11209
>=20
> Rick,
>=20
> What I have done is trivial and not important, and I have patented them f=
or living. What is important is what I am now developing and the final tech=
nology will be used by everyone, cellphones, computers and supercomputers, =
but not for industrial product controllers.
>=20
> "Please let us know when you have success in selling the patent. " OK.
>=20
>=20
> As an individual inventor, I have no resources to detect any situations a=
s you described: How would you know if a company is using your patented ide=
a in their tools?
>=20
> What I want to do is to sell the patent once and get money to support my =
next project that will benefit everyone in the world.

If you don't have a way of knowing whether a company is using your patent, =
why would anyone pay you for it rather than just use it and wait for you to=
 come after them for not paying royalties?  That happens often even between=
 large companies.  Eventually a company will reverse design a product and f=
ind out their patent is infringed.  But there has to be a way of knowing if=
 the patent is infringed or not.  Otherwise no one will pay for it.=20

--=20

  Rick C.

  + Get 1,000 miles of free Supercharging
  + Tesla referral code - https://ts.la/richard11209

Article: 161532
Subject: Re: New coding method for a state machine in groups in HDL
From: KJ <kkjennings@sbcglobal.net>
Date: Tue, 26 Nov 2019 12:15:53 -0800 (PST)
Links: << >>  << T >>  << A >>
On Monday, November 25, 2019 at 4:27:41 PM UTC-5, Weng Tianxiang wrote:
> Here is a coding snippet for new method you can immediately understand what will happening for coding a state machine.
> 
> type State_Machine_t is (
>   First_group : (s1, s2, s3),
>   Second_Group : (s4, s5, s6),
>   Third_Group ; (s7, s8, s9)
> );
> signal State_Machine, State_Machine_Next : State_Machine_t ; 
> 
> Here is only one sentence that tells you what you can do with new method:
> Any state machine's states can be grouped together in coding in HDL at one's discretion and each group acts like a small state machine sharing a same initial state and all act together as coordinated as a full state machine.

KJ: And I suppose you think that this is new or novel or something (Hint:  I do not)

> Andy,
> As an independent inventor, I want to sell the patent to a company to get financial support for my next tens of new exciting inventions.

KJ:  Current patents that have your name are all assigned only to you.  Does that mean you have not sold any patents yet?  When somebody does sell a patent, does the assignee have to get updated at USPTO as part of that sale?

Kevin Jennings

Article: 161533
Subject: Re: New coding method for a state machine in groups in HDL
From: Michael Kellett <mk@mkesc.co.uk>
Date: Wed, 27 Nov 2019 09:42:30 +0000
Links: << >>  << T >>  << A >>
On 26/11/2019 13:50, Weng Tianxiang wrote:
> On Tuesday, November 26, 2019 at 4:59:38 AM UTC-8, Michael Kellett wrote:
>> On 26/11/2019 12:39, rob dickinson wrote:
>>> On Monday, November 25, 2019 at 9:27:41 PM UTC, Weng Tianxiang wrote:
>>>> Hi everyone,
>>>>
>>>> Welcome all critics from ones who are interested in coding state machine that seems to many as matured, but  can be further improved.
>>>>
>>>> Here is a coding snippet for new method you can immediately understand what will happening for coding a state machine.
>>>>
>>>> type State_Machine_t is (
>>>>     First_group : (s1, s2, s3),
>>>>     Second_Group : (s4, s5, s6),
>>>>     Third_Group ; (s7, s8, s9)
>>>> );
>>>> signal State_Machine, State_Machine_Next : State_Machine_t ;
>>>>
>>>> Here is only one sentence that tells you what you can do with new method:
>>>> Any state machine's states can be grouped together in coding in HDL at one's discretion and each group acts like a small state machine sharing a same initial state and all act together as coordinated as a full state machine.
>>>>
>>>> Here is a famous paper about the method with 244 cites, using probability theory:
>>>>    http://www.scarpaz.com/2100-papers/Low%20Power/00503933.pdf
>>>>
>>>> But my patent describes a new simpler and perfect method using group theory:
>>>>
>>>> Patent:10482208, "Coding and synthesizing a state machine in state groups"
>>>>
>>>> https://patents.justia.com/search?q=tianxiang+weng
>>>>
>>>> Thank KJ for helping me to improve the patent text.
>>>>
>>>> Thank you.
>>>>
>>>> Weng
>>>
>>> Good luck proving that I have never done this before.
>>>
>>
>> The patent has been granted -  so I think it may be up to you to prove
>> that you have :-)
>>
>> (Would you have wanted to ?)
>>
>> MK
> 
> Hi MK,
> I don't understand your saying "I think it may be up to you to prove
>> that you have".
> 
> It's USPTO official publication, what do I have to prove?
> 
> I have patented not the coding method, but on how to generate the specific circuit that can be grouped with easiest coding method on mind.
> 
> Andy,
> As an independent inventor, I want to sell the patent to a company to get financial support for my next tens of new exciting inventions.
> 
> Weng
> 

Strewth Weng !

Here you are patenting state machines and you don't seem to get news net 
threads.
My comment was to Rob Dickinson, not to you.

MK

Article: 161534
Subject: Re: New coding method for a state machine in groups in HDL
From: Michael Kellett <mk@mkesc.co.uk>
Date: Wed, 27 Nov 2019 09:47:09 +0000
Links: << >>  << T >>  << A >>
On 26/11/2019 15:44, Rick C wrote:
> On Tuesday, November 26, 2019 at 8:50:23 AM UTC-5, Weng Tianxiang wrote:
>> On Tuesday, November 26, 2019 at 4:59:38 AM UTC-8, Michael Kellett wrote:
>>> On 26/11/2019 12:39, rob dickinson wrote:
>>>> On Monday, November 25, 2019 at 9:27:41 PM UTC, Weng Tianxiang wrote:
>>>>> Hi everyone,
>>>>>
>>>>> Welcome all critics from ones who are interested in coding state machine that seems to many as matured, but  can be further improved.
>>>>>
>>>>> Here is a coding snippet for new method you can immediately understand what will happening for coding a state machine.
>>>>>
>>>>> type State_Machine_t is (
>>>>>     First_group : (s1, s2, s3),
>>>>>     Second_Group : (s4, s5, s6),
>>>>>     Third_Group ; (s7, s8, s9)
>>>>> );
>>>>> signal State_Machine, State_Machine_Next : State_Machine_t ;
>>>>>
>>>>> Here is only one sentence that tells you what you can do with new method:
>>>>> Any state machine's states can be grouped together in coding in HDL at one's discretion and each group acts like a small state machine sharing a same initial state and all act together as coordinated as a full state machine.
>>>>>
>>>>> Here is a famous paper about the method with 244 cites, using probability theory:
>>>>>    http://www.scarpaz.com/2100-papers/Low%20Power/00503933.pdf
>>>>>
>>>>> But my patent describes a new simpler and perfect method using group theory:
>>>>>
>>>>> Patent:10482208, "Coding and synthesizing a state machine in state groups"
>>>>>
>>>>> https://patents.justia.com/search?q=tianxiang+weng
>>>>>
>>>>> Thank KJ for helping me to improve the patent text.
>>>>>
>>>>> Thank you.
>>>>>
>>>>> Weng
>>>>
>>>> Good luck proving that I have never done this before.
>>>>
>>>
>>> The patent has been granted -  so I think it may be up to you to prove
>>> that you have :-)
>>>
>>> (Would you have wanted to ?)
>>>
>>> MK
>>
>> Hi MK,
>> I don't understand your saying "I think it may be up to you to prove
>>> that you have".
>>
>> It's USPTO official publication, what do I have to prove?
>>
>> I have patented not the coding method, but on how to generate the specific circuit that can be grouped with easiest coding method on mind.
>>
>> Andy,
>> As an independent inventor, I want to sell the patent to a company to get financial support for my next tens of new exciting inventions.
>>
>> Weng
> 
> Please let us know when you have success in selling the patent.  I assume you would go for a royalty rather than a single payment, or perhaps a combination.
> 
> How would you know if a company is using your patented idea in their tools?  Enforcement might be the hard part.  Heck, how do you know companies aren't using the idea already?
> 

He may be able to sell it to an NPE, although they prefer to buy in bulk.
The way they operate doesn't rely on precise identification of infringement.

Here's a link:

https://www.investopedia.com/articles/investing/071515/who-are-patent-trolls-how-do-they-work.asp

Not that good an article but will give you pointers to follow if you are 
really interested.

MK

Article: 161535
Subject: Re: New coding method for a state machine in groups in HDL
From: Weng Tianxiang <wtxwtx@gmail.com>
Date: Wed, 27 Nov 2019 08:40:59 -0800 (PST)
Links: << >>  << T >>  << A >>

> If you don't have a way of knowing whether a company is using your patent=
, why would anyone pay you for it rather than just use it and wait for you =
to come after them for not paying royalties?  That happens often even betwe=
en large companies.  Eventually a company will reverse design a product and=
 find out their patent is infringed.  But there has to be a way of knowing =
if the patent is infringed or not.  Otherwise no one will pay for it.=20
>=20
> --=20
>=20
>   Rick C.

Rick,
I think I will have many many more patents in the future.

My strategy dealing with your infringing cases is very simple:
To any engineer who has solid evidence to prove in court that your current =
or previous employer has infringed my patents, please cooperate with my com=
pany as witness to suit the employer and he/she will get 20% of damaging pe=
nalty of court or one time compensation the employer will pay for previous =
infringement if he agrees to resolve the issue before law suit.

Weng





 =20


Article: 161536
Subject: Re: New coding method for a state machine in groups in HDL
From: Rick C <gnuarm.deletethisbit@gmail.com>
Date: Wed, 27 Nov 2019 09:32:52 -0800 (PST)
Links: << >>  << T >>  << A >>
On Wednesday, November 27, 2019 at 11:41:03 AM UTC-5, Weng Tianxiang wrote:
> > If you don't have a way of knowing whether a company is using your pate=
nt, why would anyone pay you for it rather than just use it and wait for yo=
u to come after them for not paying royalties?  That happens often even bet=
ween large companies.  Eventually a company will reverse design a product a=
nd find out their patent is infringed.  But there has to be a way of knowin=
g if the patent is infringed or not.  Otherwise no one will pay for it.=20
> >=20
> > --=20
> >=20
> >   Rick C.
>=20
> Rick,
> I think I will have many many more patents in the future.
>=20
> My strategy dealing with your infringing cases is very simple:
> To any engineer who has solid evidence to prove in court that your curren=
t or previous employer has infringed my patents, please cooperate with my c=
ompany as witness to suit the employer and he/she will get 20% of damaging =
penalty of court or one time compensation the employer will pay for previou=
s infringement if he agrees to resolve the issue before law suit.

Virtually all employees have to sign non-disclosure agreements these days w=
hich would prevent them from coming to you with any information about what =
the company was working on.  If they come to you, they will be held liable =
for ALL costs the infringing company has to pay you... and all legal fees. =
=20

That might put a damper on their willingness to get 20% of what you make.=
=20

I think your main problem is what appears to be a total lack of exposure to=
 the engineering work environment.  You come up with lots of ideas, but don=
't seem to understand their utility or lack thereof.  It appears that you d=
on't understand the real issues that designers encounter.  Designers are no=
t mathematicians or logicians.  They don't typically write research papers.=
  They just get their work done.=20

--=20

  Rick C.

  -- Get 1,000 miles of free Supercharging
  -- Tesla referral code - https://ts.la/richard11209

Article: 161537
Subject: Re: New coding method for a state machine in groups in HDL
From: Weng Tianxiang <wtxwtx@gmail.com>
Date: Wed, 27 Nov 2019 10:03:10 -0800 (PST)
Links: << >>  << T >>  << A >>
On Wednesday, November 27, 2019 at 9:32:56 AM UTC-8, Rick C wrote:
> On Wednesday, November 27, 2019 at 11:41:03 AM UTC-5, Weng Tianxiang wrot=
e:
> > > If you don't have a way of knowing whether a company is using your pa=
tent, why would anyone pay you for it rather than just use it and wait for =
you to come after them for not paying royalties?  That happens often even b=
etween large companies.  Eventually a company will reverse design a product=
 and find out their patent is infringed.  But there has to be a way of know=
ing if the patent is infringed or not.  Otherwise no one will pay for it.=
=20
> > >=20
> > > --=20
> > >=20
> > >   Rick C.
> >=20
> > Rick,
> > I think I will have many many more patents in the future.
> >=20
> > My strategy dealing with your infringing cases is very simple:
> > To any engineer who has solid evidence to prove in court that your curr=
ent or previous employer has infringed my patents, please cooperate with my=
 company as witness to suit the employer and he/she will get 20% of damagin=
g penalty of court or one time compensation the employer will pay for previ=
ous infringement if he agrees to resolve the issue before law suit.
>=20
> Virtually all employees have to sign non-disclosure agreements these days=
 which would prevent them from coming to you with any information about wha=
t the company was working on.  If they come to you, they will be held liabl=
e for ALL costs the infringing company has to pay you... and all legal fees=
. =20
>=20
> That might put a damper on their willingness to get 20% of what you make.=
=20
>=20
> I think your main problem is what appears to be a total lack of exposure =
to the engineering work environment.  You come up with lots of ideas, but d=
on't seem to understand their utility or lack thereof.  It appears that you=
 don't understand the real issues that designers encounter.  Designers are =
not mathematicians or logicians.  They don't typically write research paper=
s.  They just get their work done.=20
>=20
> --=20
>=20
>   Rick C.
>=20
>   -- Get 1,000 miles of free Supercharging
>   -- Tesla referral code - https://ts.la/richard11209

Rick,

You are right in some points, but not all.

Court system has precedence on how it may work.

Chinese Huawei was suited by Cisco company for coping their manuals in thei=
r laboratory research department. The proof was provided by a white America=
n engineer previously employed by Huawei. Huawei resolved the suit by payin=
g $500 million to Cosco and what the white American engineer got?=20

Did not he sign non-disclosure agreement before he joined the company?

Non-disclosure agreement does not shut your employees' mouth up totally, an=
d in law scope it does not cover any illegal activities.=20

Of cause the employee who would cooperate with me, in most cases, would hav=
e left his company before any lawsuit.

Weng

Article: 161538
Subject: Re: New coding method for a state machine in groups in HDL
From: Rick C <gnuarm.deletethisbit@gmail.com>
Date: Wed, 27 Nov 2019 13:03:23 -0800 (PST)
Links: << >>  << T >>  << A >>
On Wednesday, November 27, 2019 at 1:03:15 PM UTC-5, Weng Tianxiang wrote:
> On Wednesday, November 27, 2019 at 9:32:56 AM UTC-8, Rick C wrote:
> > On Wednesday, November 27, 2019 at 11:41:03 AM UTC-5, Weng Tianxiang wr=
ote:
> > > > If you don't have a way of knowing whether a company is using your =
patent, why would anyone pay you for it rather than just use it and wait fo=
r you to come after them for not paying royalties?  That happens often even=
 between large companies.  Eventually a company will reverse design a produ=
ct and find out their patent is infringed.  But there has to be a way of kn=
owing if the patent is infringed or not.  Otherwise no one will pay for it.=
=20
> > > >=20
> > > > --=20
> > > >=20
> > > >   Rick C.
> > >=20
> > > Rick,
> > > I think I will have many many more patents in the future.
> > >=20
> > > My strategy dealing with your infringing cases is very simple:
> > > To any engineer who has solid evidence to prove in court that your cu=
rrent or previous employer has infringed my patents, please cooperate with =
my company as witness to suit the employer and he/she will get 20% of damag=
ing penalty of court or one time compensation the employer will pay for pre=
vious infringement if he agrees to resolve the issue before law suit.
> >=20
> > Virtually all employees have to sign non-disclosure agreements these da=
ys which would prevent them from coming to you with any information about w=
hat the company was working on.  If they come to you, they will be held lia=
ble for ALL costs the infringing company has to pay you... and all legal fe=
es. =20
> >=20
> > That might put a damper on their willingness to get 20% of what you mak=
e.=20
> >=20
> > I think your main problem is what appears to be a total lack of exposur=
e to the engineering work environment.  You come up with lots of ideas, but=
 don't seem to understand their utility or lack thereof.  It appears that y=
ou don't understand the real issues that designers encounter.  Designers ar=
e not mathematicians or logicians.  They don't typically write research pap=
ers.  They just get their work done.=20
> >=20
> > --=20
> >=20
> >   Rick C.
> >=20
> >   -- Get 1,000 miles of free Supercharging
> >   -- Tesla referral code - https://ts.la/richard11209
>=20
> Rick,
>=20
> You are right in some points, but not all.
>=20
> Court system has precedence on how it may work.
>=20
> Chinese Huawei was suited by Cisco company for coping their manuals in th=
eir laboratory research department. The proof was provided by a white Ameri=
can engineer previously employed by Huawei. Huawei resolved the suit by pay=
ing $500 million to Cosco and what the white American engineer got?=20
>=20
> Did not he sign non-disclosure agreement before he joined the company?
>=20
> Non-disclosure agreement does not shut your employees' mouth up totally, =
and in law scope it does not cover any illegal activities.=20
>=20
> Of cause the employee who would cooperate with me, in most cases, would h=
ave left his company before any lawsuit.
>=20
> Weng

Leaving a company does not relieve a person of the responsibility of adheri=
ng to the non disclosure agreement.  I can't attest to any particular case =
unless I have the details and there may be some cases where a company won't=
 pursue the matter.  But it does tend to be a huge impediment to employees =
and ex-employees from talking about work.=20

I stand my my previous statements.=20
=20
--=20

  Rick C.

  -+ Get 1,000 miles of free Supercharging
  -+ Tesla referral code - https://ts.la/richard11209

Article: 161539
Subject: Re: New coding method for a state machine in groups in HDL
From: dlheliski@gmail.com
Date: Wed, 27 Nov 2019 13:41:37 -0800 (PST)
Links: << >>  << T >>  << A >>
On Monday, November 25, 2019 at 4:27:41 PM UTC-5, Weng Tianxiang wrote:
> Hi everyone,
>=20
> Welcome all critics from ones who are interested in coding state machine =
that seems to many as matured, but  can be further improved.
>=20
> Here is a coding snippet for new method you can immediately understand wh=
at will happening for coding a state machine.
>=20
> type State_Machine_t is (
>   First_group : (s1, s2, s3),
>   Second_Group : (s4, s5, s6),
>   Third_Group ; (s7, s8, s9)
> );
> signal State_Machine, State_Machine_Next : State_Machine_t ;=20
>=20
> Here is only one sentence that tells you what you can do with new method:
> Any state machine's states can be grouped together in coding in HDL at on=
e's discretion and each group acts like a small state machine sharing a sam=
e initial state and all act together as coordinated as a full state machine=
.
>=20
> Here is a famous paper about the method with 244 cites, using probability=
 theory:
>  http://www.scarpaz.com/2100-papers/Low%20Power/00503933.pdf
>=20
> But my patent describes a new simpler and perfect method using group theo=
ry:
>=20
> Patent:10482208, "Coding and synthesizing a state machine in state groups=
"
>=20
> https://patents.justia.com/search?q=3Dtianxiang+weng=20
>=20
> Thank KJ for helping me to improve the patent text.
>=20
> Thank you.=20
>=20
> Weng

Can you compare to this paper which divides states into groups and only act=
ivates clocks of active FSMs:

Asynchronous control of low-power gated-clock finite-state-machines

: ICECS'99. Proceedings of ICECS '99. 6th IEEE International Conference on =
Electronics, Circuits and Systems=20

An efficient approach to reduce power consumption in a synchronous Finite-S=
tate Machine (FSM) is to de-compose it, according to a partitioning algorit=
hm, to a number of sub-FSMs that interact through some communication signal=
s. Only one sub-FSM is clocked at a time and low power operation is obtaine=
d by only clocking the active sub-FSM. In this paper we introduce a new asy=
nchronous communication control for the interacting sub-FSMs, which reduces=
 the total capacitance switched by the system clock. Experimental results s=
how that this leads to significant power savings when the FSM is partitione=
d into many sub-FSMs.

Article: 161540
Subject: Re: New coding method for a state machine in groups in HDL
From: Weng Tianxiang <wtxwtx@gmail.com>
Date: Wed, 27 Nov 2019 14:26:12 -0800 (PST)
Links: << >>  << T >>  << A >>

> Can you compare to this paper which divides states into groups and only a=
ctivates clocks of active FSMs:
>=20
> Asynchronous control of low-power gated-clock finite-state-machines
>=20
> : ICECS'99. Proceedings of ICECS '99. 6th IEEE International Conference o=
n Electronics, Circuits and Systems=20
>=20
> An efficient approach to reduce power consumption in a synchronous Finite=
-State Machine (FSM) is to de-compose it, according to a partitioning algor=
ithm, to a number of sub-FSMs that interact through some communication sign=
als. Only one sub-FSM is clocked at a time and low power operation is obtai=
ned by only clocking the active sub-FSM. In this paper we introduce a new a=
synchronous communication control for the interacting sub-FSMs, which reduc=
es the total capacitance switched by the system clock. Experimental results=
 show that this leads to significant power savings when the FSM is partitio=
ned into many sub-FSMs.


Hi dhe,

Thank you for your post. I like your post very much because until now nobod=
y cares about my working principles that are unique and inventive and many =
experienced designers can learn something from them.

I will return to your post within a few hours after I carefully read the pa=
per.

From your post, I know at least 2 important differences between your post a=
nd my method so two methods work for the same function, but use two  differ=
ent principles.

1) Your post: a partitioning algorithm.
My method: partitioning method is at your discretion so you can  partition =
a state machine freely.

2) Your post: a new asynchronous communication control for the interacting =
sub-FSMs.
My method: No new asynchronous communication control for the interacting su=
b-FSMs is generated except that only a few "or" and "and" operators are inv=
olved so that the logic is simpler than others.

I did not read and know the paper before until now.

Weng

Article: 161541
Subject: Re: New coding method for a state machine in groups in HDL
From: Weng Tianxiang <wtxwtx@gmail.com>
Date: Wed, 27 Nov 2019 15:56:16 -0800 (PST)
Links: << >>  << T >>  << A >>
On Wednesday, November 27, 2019 at 2:26:15 PM UTC-8, Weng Tianxiang wrote:
> > Can you compare to this paper which divides states into groups and only=
 activates clocks of active FSMs:
> >=20
> > Asynchronous control of low-power gated-clock finite-state-machines
> >=20
> > : ICECS'99. Proceedings of ICECS '99. 6th IEEE International Conference=
 on Electronics, Circuits and Systems=20
> >=20
> > An efficient approach to reduce power consumption in a synchronous Fini=
te-State Machine (FSM) is to de-compose it, according to a partitioning alg=
orithm, to a number of sub-FSMs that interact through some communication si=
gnals. Only one sub-FSM is clocked at a time and low power operation is obt=
ained by only clocking the active sub-FSM. In this paper we introduce a new=
 asynchronous communication control for the interacting sub-FSMs, which red=
uces the total capacitance switched by the system clock. Experimental resul=
ts show that this leads to significant power savings when the FSM is partit=
ioned into many sub-FSMs.
>=20
>=20
> Hi dhe,
>=20
> Thank you for your post. I like your post very much because until now nob=
ody cares about my working principles that are unique and inventive and man=
y experienced designers can learn something from them.
>=20
> I will return to your post within a few hours after I carefully read the =
paper.
>=20
> From your post, I know at least 2 important differences between your post=
 and my method so two methods work for the same function, but use two  diff=
erent principles.
>=20
> 1) Your post: a partitioning algorithm.
> My method: partitioning method is at your discretion so you can  partitio=
n a state machine freely.
>=20
> 2) Your post: a new asynchronous communication control for the interactin=
g sub-FSMs.
> My method: No new asynchronous communication control for the interacting =
sub-FSMs is generated except that only a few "or" and "and" operators are i=
nvolved so that the logic is simpler than others.
>=20
> I did not read and know the paper before until now.
>=20
> Weng

Hi dlhe,

Sorry for misspelling your name in my last response to you.

The paper uses another paper's probability theory to generate group scheme.
=20
Their method uses an asynchronous transition state machine to generate rela=
ted clock gating signals and my method uses combination of state jumping si=
gnals to generate related clock gating signals.

Working principles of the present invention:
1) A piece of combinational logic is defined as a jumping signal for a stat=
e machine, and each jumping signal has the following characteristics:
 a) A jumping signal has a current state.
 b) A jumping signal has a target state.
 c) A jumping signal has a transfer function that doesn=E2=80=99t include i=
nformation of the state machine.=20
 d) A jumping signal is deasserted if input signal SINI is asserted.=20
 e) One and only one jumping signal is asserted on any cycle after the stat=
e machine is initialized.
 f) If a jumping signal is asserted on the current cycle, the state machine=
 will jump from the jumping signal=E2=80=99s current state to its target st=
ate on the next cycle.=20
2) All states are grouped into one or more State Groups (SGs) by a designer=
 or a synthesizer, and an SG may have from one state to all states in the s=
tate machine.=20
3) Attach each of SGs with a clock gating device.
4) Define a jumping signal as a true jumping signal for an SG if the jumpin=
g signal=E2=80=99s current state and target state are different, and the ta=
rget state belongs to the SG.=20
5) Feed each state of an SG with all true jumping signals whose target stat=
e is that state.
6) An SG will change states on the next cycle if the SG has at least one of=
 the currently asserted jumping signal=E2=80=99s current state or target st=
ate, and the current state and the target state are 2 different states.=20
7) Generate a clock pulse to each of SGs on the next cycle when either inpu=
t signal SINI is asserted on the current cycle or the SG will change states=
 on the next cycle.

I copy this paragraph from my patent's specification.

Weng

Article: 161542
Subject: Re: New coding method for a state machine in groups in HDL
From: Weng Tianxiang <wtxwtx@gmail.com>
Date: Wed, 27 Nov 2019 20:45:46 -0800 (PST)
Links: << >>  << T >>  << A >>
> >=20
> > Hi dhe,
> >=20
> > Thank you for your post. I like your post very much because until now n=
obody cares about my working principles that are unique and inventive and m=
any experienced designers can learn something from them.
> >=20
> > I will return to your post within a few hours after I carefully read th=
e paper.
> >=20
> > From your post, I know at least 2 important differences between your po=
st and my method so two methods work for the same function, but use two  di=
fferent principles.
> >=20
> > 1) Your post: a partitioning algorithm.
> > My method: partitioning method is at your discretion so you can  partit=
ion a state machine freely.
> >=20
> > 2) Your post: a new asynchronous communication control for the interact=
ing sub-FSMs.
> > My method: No new asynchronous communication control for the interactin=
g sub-FSMs is generated except that only a few "or" and "and" operators are=
 involved so that the logic is simpler than others.
> >=20
> > I did not read and know the paper before until now.
> >=20
> > Weng
>=20
> Hi dlhe,
>=20
> Sorry for misspelling your name in my last response to you.
>=20
> The paper uses another paper's probability theory to generate group schem=
e.
> =20
> Their method uses an asynchronous transition state machine to generate re=
lated clock gating signals and my method uses combination of state jumping =
signals to generate related clock gating signals.
>=20
> Working principles of the present invention:
> 1) A piece of combinational logic is defined as a jumping signal for a st=
ate machine, and each jumping signal has the following characteristics:
>  a) A jumping signal has a current state.
>  b) A jumping signal has a target state.
>  c) A jumping signal has a transfer function that doesn=E2=80=99t include=
 information of the state machine.=20
>  d) A jumping signal is deasserted if input signal SINI is asserted.=20
>  e) One and only one jumping signal is asserted on any cycle after the st=
ate machine is initialized.
>  f) If a jumping signal is asserted on the current cycle, the state machi=
ne will jump from the jumping signal=E2=80=99s current state to its target =
state on the next cycle.=20
> 2) All states are grouped into one or more State Groups (SGs) by a design=
er or a synthesizer, and an SG may have from one state to all states in the=
 state machine.=20
> 3) Attach each of SGs with a clock gating device.
> 4) Define a jumping signal as a true jumping signal for an SG if the jump=
ing signal=E2=80=99s current state and target state are different, and the =
target state belongs to the SG.=20
> 5) Feed each state of an SG with all true jumping signals whose target st=
ate is that state.
> 6) An SG will change states on the next cycle if the SG has at least one =
of the currently asserted jumping signal=E2=80=99s current state or target =
state, and the current state and the target state are 2 different states.=
=20
> 7) Generate a clock pulse to each of SGs on the next cycle when either in=
put signal SINI is asserted on the current cycle or the SG will change stat=
es on the next cycle.
>=20
> I copy this paragraph from my patent's specification.
>=20
> Weng

Hi dlhe,=20

You may be confused by above description, here are more definitions which a=
re bases for developing my method fully and systematically.

12) A jumping signal is defined as a true jumping signal for an SG if the j=
umping signal=E2=80=99s current state and target state are different, and i=
ts target state belongs to the SG.=20

13) A true jumping signal is called an entry jumping signal for an SG if th=
e SG does not have the true jumping signal=E2=80=99s current state. =20

14) A true jumping signal is called a local jumping signal for an SG if the=
 SG has the true jumping signal=E2=80=99s current state. =20

15) A jumping signal is called a leaving jumping signal for an SG if the SG=
 has the jumping signal=E2=80=99s current state and does not have its targe=
t state. =20

16) A jumping signal is called a crossing jumping signal for a state machin=
e if the jumping signal=E2=80=99s current state and target state belong to =
2 different SGs.=20

17) A crossing jumping signal belongs to two SGs: a) the SG has the crossin=
g jumping signal=E2=80=99s current state and the crossing jumping signal is=
 used as a leaving jumping signal for the SG; and b) if the SG has the cros=
sing jumping signal=E2=80=99s target state and the crossing jumping signal =
is used as an entry jumping signal for the SG.

18) A jumping signal is called a holding jumping signal for an SG if the SG=
 has the jumping signal=E2=80=99s current state and target state, and both =
are the same. =20

19) An SG is said to have a currently asserted jumping signal on the curren=
t cycle if the currently asserted jumping signal belongs to the SG.

Using different combinations of all jumping signals belonging to a SG, one =
can generate different simplest circuits to finish the job.

Weng

Article: 161543
Subject: Re: New coding method for a state machine in groups in HDL
From: KJ <kkjennings@sbcglobal.net>
Date: Thu, 28 Nov 2019 12:44:44 -0800 (PST)
Links: << >>  << T >>  << A >>

> Using different combinations of all jumping signals belonging to a SG, on=
e can generate different simplest circuits to finish the job.=C2=A0

For years now you've made these same claims about your state machines witho=
ut providing any evidence to back the claim. During those years I've provid=
ed evidence showing your claims to be false.  I doubt there will be anythin=
g different this time since it appears to be more of the same.

Kevin Jennings

Article: 161544
Subject: Re: New coding method for a state machine in groups in HDL
From: Weng Tianxiang <wtxwtx@gmail.com>
Date: Thu, 28 Nov 2019 13:33:56 -0800 (PST)
Links: << >>  << T >>  << A >>
On Thursday, November 28, 2019 at 12:44:48 PM UTC-8, KJ wrote:
> > Using different combinations of all jumping signals belonging to a SG, =
one can generate different simplest circuits to finish the job.=C2=A0
>=20
> For years now you've made these same claims about your state machines wit=
hout providing any evidence to back the claim. During those years I've prov=
ided evidence showing your claims to be false.  I doubt there will be anyth=
ing different this time since it appears to be more of the same.
>=20
> Kevin Jennings

KJ,

Please read my diagrams carefully in my patent and you are familiar with no=
ne of what I asked you to check my application for specification and diagra=
ms 9 years ago.

If you can find any point, please show it and don't talk nothing.=20

I remember you were talking about some jumping signal generations last time=
 and I didn't see any point of your argument.=20

Let HT-Lab or Rick as arbitrator to determine whether your standpoint is me=
aningless or not. Last time Rick was on my side.

Actually this patent, US 10482208, has nothing to do with how to generate j=
umping signals for a state machine!!!

I define the jumping signals for a state machine in the patent, but never m=
ention how to generate them.

Weng

Article: 161545
Subject: Efinix and their Trion FPGAs
From: Brane2 <brankob@avtomatika.com>
Date: Thu, 28 Nov 2019 23:42:58 -0800 (PST)
Links: << >>  << T >>  << A >>
They are relatively new:
https://www.efinixinc.com

Their Trion program has few notable peculiarities:

- unified roruting/LUT tiles. One tile can be used for both roles.
- 5 bit BLOCK RAM instaed of traditional 9-bit

They claim that this and process innovation (only 7 metalic  layers) etc that ahould bringg 4x lower prices.

But at least from theri initial offering, as much as it can

Article: 161546
Subject: Efinix and their new Trion FPGAs -
From: Brane2 <brankob@avtomatika.com>
Date: Thu, 28 Nov 2019 23:51:47 -0800 (PST)
Links: << >>  << T >>  << A >>
I'm talking about these guys:

https://www.efinixinc.com

Their Trion program seems interesting:

- it stretches from area that is occupied by Lattice's MachXO3 on the low end and ECP5 on hight end
- no onboard FLASH.Just OTP on few small models and nothing on high end
- universal tile that can do routing as well as LUT/MEM/logic
- 5 bit BLOCK RAM instead of traditional 9-bit
- additional simpplifications on the process end claim 4x chip price reduction ( only 7 metalisation layers instead of 14 etc)

Trion on first impression looks nice, but:

- a bit slower than ECP5
- based on digikey prices, not cheaper than EPC5, or even pricier at some points

has anyone used them and has some data to share on the matter ?






Article: 161547
Subject: Lattice's ECP5 - half of the program went MIA - WTF ?
From: Brane2 <brankob@avtomatika.com>
Date: Thu, 28 Nov 2019 23:56:16 -0800 (PST)
Links: << >>  << T >>  << A >>
I've noticed that literally over night ordinary ( non-SERDES, non-automotive) members of ECP5 family is gone on Lattice's pages.

Questions:

1. Is there process advancement comming ( 40nm ->28 nm or similar)
2. Are we to see iCE50, MachXO4, ECP6 shortly ?
3. How much of this is caused by process advancement (like 28 nm becoming more cost-effective than 40nm for the purpose etc) ?
4. How much of this is caused by IoT and AI expansion ?
5. HOw much of this is caued by new names and offers ( Effinix, new Chinese names etc) ?

Article: 161548
Subject: Re: Efinix and their Trion FPGAs
From: Brane2 <brankob@avtomatika.com>
Date: Fri, 29 Nov 2019 00:00:16 -0800 (PST)
Links: << >>  << T >>  << A >>
One more question:

At least on the first glance, I've got the impression that Trion is not as fast as ECP5. Is this about right ?
If so, why do you think this is - power optimisation or perhaps consequence of unified routing/LUT/MEM approach ?

Article: 161549
Subject: Re: Lattice's ECP5 - half of the program went MIA - WTF ?
From: Michael Kellett <mk@mkesc.co.uk>
Date: Fri, 29 Nov 2019 09:06:52 +0000
Links: << >>  << T >>  << A >>
On 29/11/2019 07:56, Brane2 wrote:
> I've noticed that literally over night ordinary ( non-SERDES, non-automotive) members of ECP5 family is gone on Lattice's pages.
> 
> Questions:
> 
> 1. Is there process advancement comming ( 40nm ->28 nm or similar)
> 2. Are we to see iCE50, MachXO4, ECP6 shortly ?
> 3. How much of this is caused by process advancement (like 28 nm becoming more cost-effective than 40nm for the purpose etc) ?
> 4. How much of this is caused by IoT and AI expansion ?
> 5. HOw much of this is caued by new names and offers ( Effinix, new Chinese names etc) ?
> 
Near panic - just about to design in LFE5U-45 in 256pin 0.8mm pitch BGA,

Still on the website today:
under the bold heading "0.8mm Spacing I/O Count/SERDES", 4th - 7th 
columns are parts with no serdes and not automotive.

Mouser had stock yesterday.

MK



Site Home   Archive Home   FAQ Home   How to search the Archive   How to Navigate the Archive   
Compare FPGA features and resources   

Threads starting:
1994JulAugSepOctNovDec1994
1995JanFebMarAprMayJunJulAugSepOctNovDec1995
1996JanFebMarAprMayJunJulAugSepOctNovDec1996
1997JanFebMarAprMayJunJulAugSepOctNovDec1997
1998JanFebMarAprMayJunJulAugSepOctNovDec1998
1999JanFebMarAprMayJunJulAugSepOctNovDec1999
2000JanFebMarAprMayJunJulAugSepOctNovDec2000
2001JanFebMarAprMayJunJulAugSepOctNovDec2001
2002JanFebMarAprMayJunJulAugSepOctNovDec2002
2003JanFebMarAprMayJunJulAugSepOctNovDec2003
2004JanFebMarAprMayJunJulAugSepOctNovDec2004
2005JanFebMarAprMayJunJulAugSepOctNovDec2005
2006JanFebMarAprMayJunJulAugSepOctNovDec2006
2007JanFebMarAprMayJunJulAugSepOctNovDec2007
2008JanFebMarAprMayJunJulAugSepOctNovDec2008
2009JanFebMarAprMayJunJulAugSepOctNovDec2009
2010JanFebMarAprMayJunJulAugSepOctNovDec2010
2011JanFebMarAprMayJunJulAugSepOctNovDec2011
2012JanFebMarAprMayJunJulAugSepOctNovDec2012
2013JanFebMarAprMayJunJulAugSepOctNovDec2013
2014JanFebMarAprMayJunJulAugSepOctNovDec2014
2015JanFebMarAprMayJunJulAugSepOctNovDec2015
2016JanFebMarAprMayJunJulAugSepOctNovDec2016
2017JanFebMarAprMayJunJulAugSepOctNovDec2017
2018JanFebMarAprMayJunJulAugSepOctNovDec2018
2019JanFebMarAprMayJunJulAugSepOctNovDec2019
2020JanFebMarAprMay2020

Authors:A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Custom Search