Re: [Cppqed-support] C++QED system size limit?

[Vukics A. <an...@vu...>]

Hi,

Thanks for your mail.

In principle, C++QED has no limit on the system arity, it’s only that
BLITZ_ARRAY_LARGEST_RANK needs to be increased for arity larger than 11 for
pure state-vector and 5 for density-operator manipulations.

So, you have hit upon a bug, for which thank you again! I have just pushed
a fix to both the Development and master branches in the git repository on
SourceForge.

We will eventually package this as a bugfix release for SourceForge and
Ubuntu, but in the meantime please try the git version, and let me know if
it works now.

In the newest git version you will find a new script 6qbits.cc in directory
CPPQEDscripts, showing what macros might be affected when increasing the
system arity, but usually it’s only BLITZ_ARRAY_LARGEST_RANK and
QUANTUMOPERATOR_TRIDIAGONAL_MAX_RANK.

Best regards,
András



On Thu, Apr 24, 2014 at 3:26 PM, Wojciech Kozlowski <
woj...@ph...> wrote:

> Dear Raimar,
>
> I changed the value by its #define everywhere where it appeared (in
> TMP_Tools.h in CPPQEDcore/utils and there are also two blitz.h headers
> with this definition)  and I recompiled everything. Didn't work.
>
> Has anybody ever used C++QED to simulate a system with more than 5
> components? I think that's my problem here as adding an additional
> interaction to wrap around and create a periodic lattice (instead of
> adding a new site) compiles and runs smoothly.
>
> Cheers,
> Wojciech
>
> On 24/04/14 13:33, Raimar Sandner wrote:
> > Dear Wojciech,
> >
> > thanks for your feedback!
> >
> > How did you set BLITZ_ARRAY_LARGEST_RANK, as a #define or as a -D
> compiler
> > flag? It has to be set before any blitz header is included.
> >
> > I think also C++QED itself has to be recompiled with this new value, not
> only
> > the script, probably András can explain the details better. He is the
> one that
> > implemented this feature in blitz and I have not used it yet :)
> >
> > Currently I am traveling and cannot check my emails regularly. Maybe
> András
> > will get in touch with you soon, otherwise I will have a closer look at
> the
> > problem next week.
> >
> > Cheers
> > Raimar
> >
> > On Thursday, April 24, 2014 12:19:41 Wojciech Kozlowski wrote:
> >> Hi,
> >>
> >> I've recently tried using C++QED for my own research which is a Bose
> >> Hubbard lattice model coupled to a cavity. I've implemented the actual
> >> BHM elements already (which are simply site elements coupled with a
> >> tunnelling interaction element between nearest neighbours), but I've
> >> encountered an obstacle when I try to make the system larger than 5
> >> sites - it fails to compile. I've attached the compiler error output to
> >> this e-mail. My own attempt to try and solve this problem got me to the
> >> BLITZ_ARRAY_LARGEST_RANK value, but changing it to a higher value did
> >> not solve the problem. Does C++QED have a very limited system size? I
> >> hope there's a workaround as at the moment it's the best tool that I
> >> found for what I need.
> >>
> >> Cheers,
> >> Wojciech
> >
> >
> ------------------------------------------------------------------------------
> > Start Your Social Network Today - Download eXo Platform
> > Build your Enterprise Intranet with eXo Platform Software
> > Java Based Open Source Intranet - Social, Extensible, Cloud Ready
> > Get Started Now And Turn Your Intranet Into A Collaboration Platform
> > http://p.sf.net/sfu/ExoPlatform
> > _______________________________________________
> > Cppqed-support mailing list
> > Cpp...@li...
> > https://lists.sourceforge.net/lists/listinfo/cppqed-support
>
>
>
> ------------------------------------------------------------------------------
> Start Your Social Network Today - Download eXo Platform
> Build your Enterprise Intranet with eXo Platform Software
> Java Based Open Source Intranet - Social, Extensible, Cloud Ready
> Get Started Now And Turn Your Intranet Into A Collaboration Platform
> http://p.sf.net/sfu/ExoPlatform
> _______________________________________________
> Cppqed-support mailing list
> Cpp...@li...
> https://lists.sourceforge.net/lists/listinfo/cppqed-support
>



-- 
*(Non nobis, Domine, non nobis, sed nomini tuo da gloriam.)*