From: Fredy konig <fredy.konig@vo...>  20071130 16:12:49

Hi Nec (in fortran) or nec2c++ (translated in c++) is an open source EM using Method of moment, a sleeping project nowdays but is the based of many commercial applications. they is also Meep from MIT. I am using Qucs and Kicad and I have to say that they complement each other I always wanted to contribute towards EM under QUcs, but never have the time until now, And I m not used to program under make , in fact a tutorial/guide on this (c++ programming QUCS simple example) would help me to start, I am more used of easy tools of borland or M$ to program C or VB. I'm more RF/mw designer, but like to progrm as well, so If you decide to start something I ll contribute too Fred 
From: Fredy konig <fredy.konig@vo...>  20071210 09:47:37

Hi all, My experience=20 I work for some time designing power amp and mw circuit, using ADS, Mwoffic= e, cadence. I am now back to phd study (parttime) for EM stuff BTW 2.5D is more like a 2D simulator with some assumption to make it almost= a 3D...but not quite a 3D, therefore a 2.5D .Calculation only around the 2= D plane. It may be messy around all the contributors so we have to organize ourself. Cheer Fred > Message du 08/12/07 =C3=A0 12h34 > De : "wernher korff" <0775118@...> > A : "Bastien ROUCARIES" <roucaries.bastien@...> > Copie =C3=A0 : qucsdevel@... > Objet : Re: [Qucsdevel] want to contribute >=20 > Hi Bastien >=20 > Thanks for your reply. >=20 > > > Hi there. > > > > > > You guys are going to have to forgive me. I am trying to figure out w= here=20 > a > > > good starting point would be to beable to figure out how qucs is writ= ten, > > > that is, the layout of functions and header linkages. can somebody pe= rhaps > > > give me a kick start on this? > >=20 > > I am quite sorry but I think we are misundestanding. I suppose you want= to=20 > > create a EM simulator isn't ? >=20 > Yes i would like to contribute to the qucs project by helping create a EM= =20 > simulator. >=20 > >=20 > > If yes that I want to say is in a first step you do not need to know ho= w=20 > qucs=20 > > work :) > > Because You will need to write from zero the EM simulator. The EM simul= ator=20 > > could be considered in first approximation as independant and as a seco= nd=20 > > order approxiamation as black box for qucs :) Therefore you are not bo= unded=20 > > to qucs library header etc. >=20 > thanks, the black box analogy got the picture across to me :) >=20 > >=20 > > Could I ask what is your technical background and education? Personnall= y I=20 > > came from microwave engineering field. >=20 > I am a second year electrical engineering student. We have covered such t= hings=20 > as C programming and electromagnetic fields, my C++ experience is on a lo= w,=20 > but i have to learn it anyway. At school we have not covered such things = as=20 > microwave engineering. Thus, my experience is way lower than that of your= s,=20 > and probably that of anyone else that is working on qucs. >=20 > >=20 > > I am volunter to talk with you about tehnical aspect and how to begin a= EM=20 > > field simulator. I think for the beginning you need to write 2.5D EM fi= eld=20 > > simulator. >=20 > Do you mean 2.5D, as in it looks 3D but cannot be rotated? >=20 > >=20 > > For languanguage and header you could use http://lapackpp.sourceforge.n= et/ > > for matrix computation.=20 > >=20 > > I recommend you to read something like > > @Book{ fieldharrington:1993, > > =09title =3D "Field Computation by moment method", > > =09series =3D "IEEE press series on electromagnetic waves", > > =09author =3D "Roger F. Harrington", > > =09publisher =3D IEEEpress, > > =09edition =3D "2", > > =09year =3D "1993", > > =09isbn =3D "0780310144", > > =09language =3D "langenglish", > > } > > In order to understand method involved >=20 > thanks, i have found a copy and am diving into it, its a little above my= =20 > level, but i'll read it again after i have read it once. >=20 > Another question, unrelated to the technical side. > How exactly do you guys all work together on an open source project where= so=20 > many people are contributing at one time? How do you know that 2 or more= =20 > people are not doing the exact same thing and are therefore doing double= =20 > work? >=20 > To Fred. > What is your experience? How can we work together? >=20 > Kind regards, > Wernher >=20 >  > SF.Net email is sponsored by:=20 > Check out the new SourceForge.net Marketplace. > It's the best place to buy or sell services for > just about anything Open Source. > http://sourceforge.net/services/buy/index.php > _______________________________________________ > Qucsdevel mailing list > Qucsdevel@... > https://lists.sourceforge.net/lists/listinfo/qucsdevel >=20 >=20 Fredy Konig 
From: Bastien ROUCARIES <roucaries.bastien+<qucs@gm...>  20071130 17:57:18

Le vendredi 30 novembre 2007, Fredy konig a =C3=A9crit=C2=A0: > Hi > > Nec (in fortran) or nec2c++ (translated in c++) is an open source EM using > Method of moment, a sleeping project nowdays but is the based of many > commercial applications.=20 Nice nec2c++ is under GPL really nice/ > they is also Meep from MIT.=20 I know it is really nice > I am using Qucs and Kicad and I have to say that they complement each oth= er Yes we know. And it will be better integer if we could convert qucs netlist= to=20 kicad netlist > I always wanted to contribute towards EM under QUcs, but never have the > time until now, And I m not used to program under make , in fact a > tutorial/guide on this (c++ programming QUCS simple example) would help me > to start, I am more used of easy tools of borland or M$ to program C or V= B. You do not need to to knwo qucs internal for doing this. You could create a= n=20 external program that take a layout and transform this layout to S param This is the hard part,=20 Using this S param and backannotate origin schematic is really easy :) =46or now consider that you could use C++ as in standard with standard c++= =20 library (complex type etc). If you need matrix or something else knot us we= =20 will give some good pointer. You could use I think LAPACK (atlas) if you ne= ed=20 this kind of stuff. If you need bibliography pointer I can you some good hint. And we could eve= n=20 compare our work to comercial software > I'm more RF/mw designer, but like to progrm as well, so If you decide to > start something I ll contribute too > > Fred =2D=20 "ROUCARIES Bastien" roucaries.bastien+qucs@...= com =2D= =2D DO NOT WRITE TO roucaries.bastien+blackhole@... OR BE BLACKLISTED 
From: Lisandro
<perezmeyer@gm...>  20071130 18:25:06
Attachments:
Message as HTML

El Viernes 30 Noviembre 2007, Bastien ROUCARIES escribi=C3=B3: > Le vendredi 30 novembre 2007, Fredy konig a =C3=A9crit=C2=A0: > > Hi > > > > Nec (in fortran) or nec2c++ (translated in c++) is an open source EM > > using Method of moment, a sleeping project nowdays but is the based of > > many commercial applications. > > Nice nec2c++ is under GPL really nice/ That's actually the backend program running in QAntenna, the program I am=20 involved in. Seems to work pretty fine. > > they is also Meep from MIT. > > I know it is really nice > > > I am using Qucs and Kicad and I have to say that they complement each > > other > > Yes we know. And it will be better integer if we could convert qucs netli= st > to kicad netlist I hope to be in that in January. But if anyone wants to start first, I will= =20 not have any problem :D Reagards, Lisandro. =2D=20 Los errores ortogr=C3=A1ficos y de redacci=C3=B3n fueron insertados con la = =C3=BAnica intenci=C3=B3n de testear sus conocimientos de la lengua castellana. Lisandro Dami=C3=A1n Nicanor P=C3=A9rez Meyer http://perezmeyer.com.ar/ #bblug irc.freenode.net 
From: wernher korff <0775118@st...>  20071206 21:02:35

Hi there. You guys are going to have to forgive me. I am trying to figure out where a= =20 good starting point would be to beable to figure out how qucs is written,=20 that is, the layout of functions and header linkages. can somebody perhaps= =20 give me a kick start on this? What i exactly want to know is where do i start looking at source code so t= hat=20 i can make a logical walk through the setup of the qucs source code and=20 understand how or where i can logically implement code. kind regards, Wernher On Friday 30 November 2007 06:57:59 pm you wrote: > Le vendredi 30 novembre 2007, Fredy konig a =C3=A9crit=C2=A0: > > Hi > > > > Nec (in fortran) or nec2c++ (translated in c++) is an open source EM us= ing > > Method of moment, a sleeping project nowdays but is the based of many > > commercial applications.=20 > Nice nec2c++ is under GPL really nice/ >=20 > > they is also Meep from MIT.=20 >=20 > I know it is really nice >=20 > > I am using Qucs and Kicad and I have to say that they complement each=20 other >=20 > Yes we know. And it will be better integer if we could convert qucs netli= st=20 to=20 > kicad netlist >=20 > > I always wanted to contribute towards EM under QUcs, but never have the > > time until now, And I m not used to program under make , in fact a > > tutorial/guide on this (c++ programming QUCS simple example) would help= me > > to start, I am more used of easy tools of borland or M$ to program C or= =20 VB. >=20 > You do not need to to knwo qucs internal for doing this. You could create= an=20 > external program that take a layout and transform this layout to S param > This is the hard part,=20 >=20 > Using this S param and backannotate origin schematic is really easy :) >=20 > For now consider that you could use C++ as in standard with standard c++= =20 > library (complex type etc). If you need matrix or something else knot us = we=20 > will give some good pointer. You could use I think LAPACK (atlas) if you= =20 need=20 > this kind of stuff. >=20 > If you need bibliography pointer I can you some good hint. And we could e= ven=20 > compare our work to comercial software >=20 > > I'm more RF/mw designer, but like to progrm as well, so If you decide to > > start something I ll contribute too > > > > Fred >=20 >=20 >=20 > =20 >=20 > "ROUCARIES Bastien" > =20 roucaries.bastien+qucs@... > = =2D > DO NOT WRITE TO roucaries.bastien+blackhole@... OR BE BLACKLISTED >=20 =2D=20 I no longer accept word attachments as i do not support the tyranical and=20 monopolising policies of microsoft regarding the patenting of word formats. We Can Put an End to Word Attachments : http://www.gnu.org/philosophy/nowordattachments.html Download OpenOffice : http://download.openoffice.org/2.2.1/contribute.html?product=3DOpenOffice.o= rg&os=3Dwinwjre&lang=3DenUS&version=3D2.2.1 
From: Bastien ROUCARIES <roucaries.bastien@gm...>  20071207 10:57:22

Le jeudi 6 d=C3=A9cembre 2007, wernher korff a =C3=A9crit=C2=A0: > Hi there. > > You guys are going to have to forgive me. I am trying to figure out where= a > good starting point would be to beable to figure out how qucs is written, > that is, the layout of functions and header linkages. can somebody perhaps > give me a kick start on this? I am quite sorry but I think we are misundestanding. I suppose you want to= =20 create a EM simulator isn't ? If yes that I want to say is in a first step you do not need to know how qu= cs=20 work :) Because You will need to write from zero the EM simulator. The EM simulator= =20 could be considered in first approximation as independant and as a second=20 order approxiamation as black box for qucs :) Therefore you are not bounde= d=20 to qucs library header etc. Could I ask what is your technical background and education? Personnally I= =20 came from microwave engineering field. I am volunter to talk with you about tehnical aspect and how to begin a EM= =20 field simulator. I think for the beginning you need to write 2.5D EM field= =20 simulator. =46or languanguage and header you could use http://lapackpp.sourceforge.net/ for matrix computation.=20 I recommend you to read something like @Book{ fieldharrington:1993, title =3D "Field Computation by moment method", series =3D "IEEE press series on electromagnetic waves", author =3D "Roger F. Harrington", publisher =3D IEEEpress, edition =3D "2", year =3D "1993", isbn =3D "0780310144", language =3D "langenglish", } In order to understand method involved > What i exactly want to know is where do i start looking at source code so > that i can make a logical walk through the setup of the qucs source code > and understand how or where i can logically implement code. > kind regards, > Wernher > > On Friday 30 November 2007 06:57:59 pm you wrote: > > Le vendredi 30 novembre 2007, Fredy konig a =C3=A9crit=C2=A0: > > > Hi > > > > > > Nec (in fortran) or nec2c++ (translated in c++) is an open source EM > > > using Method of moment, a sleeping project nowdays but is the based = of > > > many commercial applications. > > > > Nice nec2c++ is under GPL really nice/ > > > > > they is also Meep from MIT. > > > > I know it is really nice > > > > > I am using Qucs and Kicad and I have to say that they complement each > > other > > > Yes we know. And it will be better integer if we could convert qucs > > netlist > > to > > > kicad netlist > > > > > I always wanted to contribute towards EM under QUcs, but never have t= he > > > time until now, And I m not used to program under make , in fact a > > > tutorial/guide on this (c++ programming QUCS simple example) would he= lp > > > me to start, I am more used of easy tools of borland or M$ to program= C > > > or > > VB. > > > You do not need to to knwo qucs internal for doing this. You could crea= te > > an external program that take a layout and transform this layout to S > > param This is the hard part, > > > > Using this S param and backannotate origin schematic is really easy :) > > > > For now consider that you could use C++ as in standard with standard c++ > > library (complex type etc). If you need matrix or something else knot us > > we will give some good pointer. You could use I think LAPACK (atlas) if > > you > > need > > > this kind of stuff. > > > > If you need bibliography pointer I can you some good hint. And we could > > even compare our work to comercial software > > > > > I'm more RF/mw designer, but like to progrm as well, so If you decide > > > to start something I ll contribute too > > > > > > Fred > > > >  > > > > "ROUCARIES Bastien" > > roucaries.bastien+qucs@... > > > = =2D > > DO NOT WRITE TO roucaries.bastien+blackhole@... OR BE > > BLACKLISTED =2D=20 "ROUCARIES Bastien" roucaries.bastien@...= com =2D= =2D DO NOT WRITE TO roucaries.bastien+blackhole@... OR BE BLACKLISTED 
From: wernher korff <0775118@st...>  20071208 11:34:53

Hi Bastien Thanks for your reply. > > Hi there. > > > > You guys are going to have to forgive me. I am trying to figure out where a > > good starting point would be to beable to figure out how qucs is written, > > that is, the layout of functions and header linkages. can somebody perhaps > > give me a kick start on this? > > I am quite sorry but I think we are misundestanding. I suppose you want to > create a EM simulator isn't ? Yes i would like to contribute to the qucs project by helping create a EM simulator. > > If yes that I want to say is in a first step you do not need to know how qucs > work :) > Because You will need to write from zero the EM simulator. The EM simulator > could be considered in first approximation as independant and as a second > order approxiamation as black box for qucs :) Therefore you are not bounded > to qucs library header etc. thanks, the black box analogy got the picture across to me :) > > Could I ask what is your technical background and education? Personnally I > came from microwave engineering field. I am a second year electrical engineering student. We have covered such things as C programming and electromagnetic fields, my C++ experience is on a low, but i have to learn it anyway. At school we have not covered such things as microwave engineering. Thus, my experience is way lower than that of yours, and probably that of anyone else that is working on qucs. > > I am volunter to talk with you about tehnical aspect and how to begin a EM > field simulator. I think for the beginning you need to write 2.5D EM field > simulator. Do you mean 2.5D, as in it looks 3D but cannot be rotated? > > For languanguage and header you could use http://lapackpp.sourceforge.net/ > for matrix computation. > > I recommend you to read something like > @Book{ fieldharrington:1993, > title = "Field Computation by moment method", > series = "IEEE press series on electromagnetic waves", > author = "Roger F. Harrington", > publisher = IEEEpress, > edition = "2", > year = "1993", > isbn = "0780310144", > language = "langenglish", > } > In order to understand method involved thanks, i have found a copy and am diving into it, its a little above my level, but i'll read it again after i have read it once. Another question, unrelated to the technical side. How exactly do you guys all work together on an open source project where so many people are contributing at one time? How do you know that 2 or more people are not doing the exact same thing and are therefore doing double work? To Fred. What is your experience? How can we work together? Kind regards, Wernher 
From: Bastien ROUCARIES <roucaries.bastien+<qucs@gm...>  20071208 17:22:19
Attachments:
Message as HTML

Le samedi 8 d=C3=A9cembre 2007, wernher korff a =C3=A9crit=C2=A0: > Hi Bastien > > Thanks for your reply. > > > > Hi there. > > > > > > You guys are going to have to forgive me. I am trying to figure out > > > where > > a > > > > good starting point would be to beable to figure out how qucs is > > > written, that is, the layout of functions and header linkages. can > > > somebody perhaps give me a kick start on this? > > > > I am quite sorry but I think we are misundestanding. I suppose you want > > to create a EM simulator isn't ? > > Yes i would like to contribute to the qucs project by helping create a EM > simulator. > > > If yes that I want to say is in a first step you do not need to know how > > qucs > > > work :) > > Because You will need to write from zero the EM simulator. The EM > > simulator could be considered in first approximation as independant and > > as a second order approxiamation as black box for qucs :) Therefore you > > are not bounded to qucs library header etc. > > thanks, the black box analogy got the picture across to me :) Moreover ever it is the UNIX philosophy. One tool one use :) > > Could I ask what is your technical background and education? Personnally > > I came from microwave engineering field. > > I am a second year electrical engineering student. We have covered such > things as C programming and electromagnetic fields, my C++ experience is = on > a low, but i have to learn it anyway. At school we have not covered such > things as microwave engineering. Thus, my experience is way lower than th= at > of yours, and probably that of anyone else that is working on qucs. Do not decrease your knowledge. Every man (or woman) here was once student = :)=20 > > I am volunter to talk with you about tehnical aspect and how to begin a > > EM field simulator. I think for the beginning you need to write 2.5D EM > > field simulator. > > Do you mean 2.5D, as in it looks 3D but cannot be rotated? No 2.5D is 3D but with one axix (say z) ivariant by part. IE multi layer=20 cicruit pcb. It is a really useful approxiamtion. > > For languanguage and header you could use > > http://lapackpp.sourceforge.net/ for matrix computation. > > > > I recommend you to read something like > > @Book{ fieldharrington:1993, > > title =3D "Field Computation by moment method", > > series =3D "IEEE press series on electromagnetic waves", > > author =3D "Roger F. Harrington", > > publisher =3D IEEEpress, > > edition =3D "2", > > year =3D "1993", > > isbn =3D "0780310144", > > language =3D "langenglish", > > } > > In order to understand method involved > > thanks, i have found a copy and am diving into it, its a little above my > level, but i'll read it again after i have read it once. Ok but do not try to understand all the whole thing. This book is the absol= ute=20 reference but really outdated. You should read also: http://www.ntu.edu.sg/home/mwtang/bem.htm And (in high frequency electronics we solve helmotz equation) http://www.boundaryelementmethod.com/helmholtz/manual/chap1/index.htm > Another question, unrelated to the technical side. > How exactly do you guys all work together on an open source project where > so many people are contributing at one time? How do you know that 2 or mo= re > people are not doing the exact same thing and are therefore doing double > work? We do not know, but darwin will help. If you do a good job you will survive= =2E=20 Nothing more. I you want to done true colaborative workI recommend you to use: =2D git as source control tool (you could make a project on tux familly for= =20 this). Git is very nice because you could work in parallel in contrary of=20 cvs. =2D use doxygen with LaTeX comment for documenting your work =2D Make the manual and code in parallel. Do not say I will document later =2D Use well isolation between core (ie the method of moment) and mathemati= cal=20 tools used. =2D isolate graphical interface from computationnal part. The best think to= do=20 is to use two different program. If you need help I can help you in order to learn ho to use this tools. > To Fred. > What is your experience? How can we work together? > > Kind regards, > Wernher =2D=20 "ROUCARIES Bastien" roucaries.bastien+qucs@...= com =2D= =2D DO NOT WRITE TO roucaries.bastien+blackhole@... OR BE BLACKLISTED 