I find that ftgen and ftgenonce have a problem. The orc compiles
without error, but the opcodes do not set a global variable with the
ftable number.

E.g. the following line in the orc header

gisine                  ftgen                   0, 0, 65536,    10,     1

should give a gisine of 102 (it is the 2nd ftgen), but it is 0. Then
the instr that uses gisine for a table number fails.

I suppose the problem could be with ftgen and ftgenonce, or with the
handling of global variables.

The problem occurs with both the old parser and the new parser.

This problem does not seem to occur on Linux, only on Windows 7. In
both cases I have only tested with the Csound API but I do not see why
that should make any difference.

I do not know how long this problem has been there on Windows as I
have been using Linux almost exclusively for weeks if not months.

Regards,
Mike

-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com

------------------------------------------------------------------------------
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing 
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Why would this be a problem? I do not use MSVC, only GCC.

However, what you say is very suggestive. Possibly the macro is
misbehaving on MinGW. Note that my MinGW compiler is GCC 4.6.2.
Perhaps that is the problem.

I find that USE_LRINT is not defined in the compiler command to make
ftgen.o. I will define it and see if the problem goes away.

Regards,
Mike

On Sun, Mar 4, 2012 at 5:56 PM, Tito Latini  wrote:
>> This problem does not seem to occur on Linux, only on Windows 7. In
>> both cases I have only tested with the Csound API but I do not see why
>> that should make any difference.
>
> I don't know if I can help you, perhaps there is a problem
> with `MYFLT2LRND'; in `hfgens' (Engine/fgens.c:119)
>
>  ff.fno = (int) MYFLT2LRND(ff.e.p[1]);
>
> and in H/sysdep.h
>
> #ifdef USE_LRINT
> #  ifndef USE_DOUBLE
> #    define MYFLT2LONG(x) ((int32) lrintf((float) (x)))
> #    define MYFLT2LRND(x) ((int32) lrintf((float) (x)))
> #  else
> #    define MYFLT2LONG(x) ((int32) lrint((double) (x)))
> #    define MYFLT2LRND(x) ((int32) lrint((double) (x)))
> #  endif
> #elif defined(MSVC)
> #  ifndef USE_DOUBLE
> static inline int32 MYFLT2LRND(float fval)
> {
>    int result;
>    _asm {
>      fld   fval
>      fistp result
>      mov   eax, result
>    }
>    return result;
> }
> #  else
> static inline int32 MYFLT2LRND(double fval)
> {
>    int result;
>    _asm {
>      fld   fval
>      fistp result
>      mov   eax, result
>    }
>    return result;
> }
> #  endif
> #  define MYFLT2LONG(x) MYFLT2LRND(x)
> #else
> #  ifndef USE_DOUBLE
> #    define MYFLT2LONG(x) ((int32) (x))
> #    if defined(HAVE_GCC3) && defined(__i386__) && !defined(__ICC)
> #      define MYFLT2LRND(x) ((int32) lrintf((float) (x)))
> #    else
> static inline int32 MYFLT2LRND(float fval)
> {
>    return ((int32) (fval + (fval < 0.0f ? -0.5f : 0.5f)));
> }
> #    endif
> #  else
> #    define MYFLT2LONG(x) ((int32) (x))
> #    if defined(HAVE_GCC3) && defined(__i386__) && !defined(__ICC)
> #      define MYFLT2LRND(x) ((int32) lrint((double) (x)))
> #    else
> static inline int32 MYFLT2LRND(double fval)
> {
>    return ((int32) (fval + (fval < 0.0 ? -0.5 : 0.5)));
> }
> #    endif
> #  endif
> #endif
>
> tito
>
> ------------------------------------------------------------------------------
> Virtualization & Cloud Management Using Capacity Planning
> Cloud computing makes use of virtualization - but cloud computing
> also focuses on allowing computing to be delivered as a service.
> http://www.accelacomm.com/jaw/sfnl/114/51521223/
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com

------------------------------------------------------------------------------
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing 
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

That didn't seem to help but I will try a complete update and rebuild
tomorrow. I have a feeling you are right.

Regards,
Mike

On Sun, Mar 4, 2012 at 6:37 PM, Michael Gogins  wrote:
> Why would this be a problem? I do not use MSVC, only GCC.
>
> However, what you say is very suggestive. Possibly the macro is
> misbehaving on MinGW. Note that my MinGW compiler is GCC 4.6.2.
> Perhaps that is the problem.
>
> I find that USE_LRINT is not defined in the compiler command to make
> ftgen.o. I will define it and see if the problem goes away.
>
> Regards,
> Mike
>
> On Sun, Mar 4, 2012 at 5:56 PM, Tito Latini  wrote:
>>> This problem does not seem to occur on Linux, only on Windows 7. In
>>> both cases I have only tested with the Csound API but I do not see why
>>> that should make any difference.
>>
>> I don't know if I can help you, perhaps there is a problem
>> with `MYFLT2LRND'; in `hfgens' (Engine/fgens.c:119)
>>
>>  ff.fno = (int) MYFLT2LRND(ff.e.p[1]);
>>
>> and in H/sysdep.h
>>
>> #ifdef USE_LRINT
>> #  ifndef USE_DOUBLE
>> #    define MYFLT2LONG(x) ((int32) lrintf((float) (x)))
>> #    define MYFLT2LRND(x) ((int32) lrintf((float) (x)))
>> #  else
>> #    define MYFLT2LONG(x) ((int32) lrint((double) (x)))
>> #    define MYFLT2LRND(x) ((int32) lrint((double) (x)))
>> #  endif
>> #elif defined(MSVC)
>> #  ifndef USE_DOUBLE
>> static inline int32 MYFLT2LRND(float fval)
>> {
>>    int result;
>>    _asm {
>>      fld   fval
>>      fistp result
>>      mov   eax, result
>>    }
>>    return result;
>> }
>> #  else
>> static inline int32 MYFLT2LRND(double fval)
>> {
>>    int result;
>>    _asm {
>>      fld   fval
>>      fistp result
>>      mov   eax, result
>>    }
>>    return result;
>> }
>> #  endif
>> #  define MYFLT2LONG(x) MYFLT2LRND(x)
>> #else
>> #  ifndef USE_DOUBLE
>> #    define MYFLT2LONG(x) ((int32) (x))
>> #    if defined(HAVE_GCC3) && defined(__i386__) && !defined(__ICC)
>> #      define MYFLT2LRND(x) ((int32) lrintf((float) (x)))
>> #    else
>> static inline int32 MYFLT2LRND(float fval)
>> {
>>    return ((int32) (fval + (fval < 0.0f ? -0.5f : 0.5f)));
>> }
>> #    endif
>> #  else
>> #    define MYFLT2LONG(x) ((int32) (x))
>> #    if defined(HAVE_GCC3) && defined(__i386__) && !defined(__ICC)
>> #      define MYFLT2LRND(x) ((int32) lrint((double) (x)))
>> #    else
>> static inline int32 MYFLT2LRND(double fval)
>> {
>>    return ((int32) (fval + (fval < 0.0 ? -0.5 : 0.5)));
>> }
>> #    endif
>> #  endif
>> #endif
>>
>> tito
>>
>> ------------------------------------------------------------------------------
>> Virtualization & Cloud Management Using Capacity Planning
>> Cloud computing makes use of virtualization - but cloud computing
>> also focuses on allowing computing to be delivered as a service.
>> http://www.accelacomm.com/jaw/sfnl/114/51521223/
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
>
>
> --
> Michael Gogins
> Irreducible Productions
> http://www.michael-gogins.com
> Michael dot Gogins at gmail dot com



-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com

------------------------------------------------------------------------------
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Update: I updated, cleaned, and rebuilt with useLrint-1 and verified
that USE_LRINT was defined. The problem still manifested itself.

Regards,
Mike

On Sun, Mar 4, 2012 at 5:22 PM, Michael Gogins  wrote:
> I find that ftgen and ftgenonce have a problem. The orc compiles
> without error, but the opcodes do not set a global variable with the
> ftable number.
>
> E.g. the following line in the orc header
>
> gisine                  ftgen                   0, 0, 65536,    10,     1
>
> should give a gisine of 102 (it is the 2nd ftgen), but it is 0. Then
> the instr that uses gisine for a table number fails.
>
> I suppose the problem could be with ftgen and ftgenonce, or with the
> handling of global variables.
>
> The problem occurs with both the old parser and the new parser.
>
> This problem does not seem to occur on Linux, only on Windows 7. In
> both cases I have only tested with the Csound API but I do not see why
> that should make any difference.
>
> I do not know how long this problem has been there on Windows as I
> have been using Linux almost exclusively for weeks if not months.
>
> Regards,
> Mike
>
> --
> Michael Gogins
> Irreducible Productions
> http://www.michael-gogins.com
> Michael dot Gogins at gmail dot com



-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com

------------------------------------------------------------------------------
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Yes, I always use --midi-velocity etc.

Thanks for your prompt attention and fix!

But how could --midi-velocity interact with this?

Regards,
Mike

On Mon, Mar 5, 2012 at 9:41 AM, Tito Latini  wrote:
>> Update: I updated, cleaned, and rebuilt with useLrint-1 and verified
>> that USE_LRINT was defined. The problem still manifested itself.
>
> If you use `--midi-velocity' etc, I have fixed it now.
> Sorry, it is my fault with a previous fix
>
> tito
>
> ------------------------------------------------------------------------------
> Try before you buy = See our experts in action!
> The most comprehensive online learning library for Microsoft developers
> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
> Metro Style Apps, more. Free future releases when you subscribe now!
> http://p.sf.net/sfu/learndevnow-dev2
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com

------------------------------------------------------------------------------
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

I am still experiencing problems with ftgenonce and with --midi-key
and --midi-velocity, which seem to have quite working altogether. At
the moment, I can't use the head version of Csound to make music.

I have verified that I am using your updated source for insert.c.

Regards,
Mike

On Mon, Mar 5, 2012 at 10:10 AM, Tito Latini  wrote:
>> But how could --midi-velocity interact with this?
>
> overwriting the memory; when we use --midi-*, the INSDS struct is "extended"
> (INSDS normally has p1, p2 and p3)
>
> tito
>
>
> ------------------------------------------------------------------------------
> Try before you buy = See our experts in action!
> The most comprehensive online learning library for Microsoft developers
> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
> Metro Style Apps, more. Free future releases when you subscribe now!
> http://p.sf.net/sfu/learndevnow-dev2
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com

------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Csound in the 5.16 Windows installer does not have this problem. A
diff from the time of that build (the day it was uploaded) to now
should help you locate the cause of the problems.

Regards,
Mike

On Sun, Mar 4, 2012 at 5:22 PM, Michael Gogins  wrote:
> I find that ftgen and ftgenonce have a problem. The orc compiles
> without error, but the opcodes do not set a global variable with the
> ftable number.
>
> E.g. the following line in the orc header
>
> gisine                  ftgen                   0, 0, 65536,    10,     1
>
> should give a gisine of 102 (it is the 2nd ftgen), but it is 0. Then
> the instr that uses gisine for a table number fails.
>
> I suppose the problem could be with ftgen and ftgenonce, or with the
> handling of global variables.
>
> The problem occurs with both the old parser and the new parser.
>
> This problem does not seem to occur on Linux, only on Windows 7. In
> both cases I have only tested with the Csound API but I do not see why
> that should make any difference.
>
> I do not know how long this problem has been there on Windows as I
> have been using Linux almost exclusively for weeks if not months.
>
> Regards,
> Mike
>
> --
> Michael Gogins
> Irreducible Productions
> http://www.michael-gogins.com
> Michael dot Gogins at gmail dot com



-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com

------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

I will try this patch and let you know what happens. What is the other bug?

Regards,
MIke

On Tue, Mar 6, 2012 at 4:14 AM, Tito Latini  wrote:
>> Csound in the 5.16 Windows installer does not have this problem. A
>> diff from the time of that build (the day it was uploaded) to now
>> should help you locate the cause of the problems.
>
> Try to replace (in Engine/insert.c:1925)
>
>  fltp = lcloffbas + (-indx) + pextra;
>
> with
>
>  fltp = lcloffbas + (-indx);
>
> It has resolved a bug but, if it introduces another bug,
> I revert it and look for another solution.
>
> tito
>
> ------------------------------------------------------------------------------
> Keep Your Developer Skills Current with LearnDevNow!
> The most comprehensive online learning library for Microsoft developers
> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
> Metro Style Apps, more. Free future releases when you subscribe now!
> http://p.sf.net/sfu/learndevnow-d2d
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com

------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Line insert.c:1925 is now:

          fltp = lcloffbas + (-indx); /* + pextra; */

Things are working for me now. I will do more tests later this evening.

Thanks for your continued attention to these issues, but what was the
original bug?

Regards,
Mike

On Tue, Mar 6, 2012 at 4:50 AM, Tito Latini  wrote:
> I have tried ftgenonce.csd with --midi-key-cps and --midi-velocity.
> It doesn't work (gnu/linux) also with 5.14.2, 5.15 and 5.16
> with(out) the previous fix. Today I want to definitely resolve this problem.
>
> tito
>
> ------------------------------------------------------------------------------
> Keep Your Developer Skills Current with LearnDevNow!
> The most comprehensive online learning library for Microsoft developers
> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
> Metro Style Apps, more. Free future releases when you subscribe now!
> http://p.sf.net/sfu/learndevnow-d2d
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com

------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

OK, I will try this.

Regards,
Mike

On Tue, Mar 6, 2012 at 7:05 AM, Tito Latini  wrote:
> Reverted. Try the follow instr with --midi-key=4 to see the bug
>
> instr 1
>  kres rand 1
>  ival = i(kres)
>  print ival
> endin
>
> Yes, `kres init 0' (and other tricks) works, but it is to fix.
> I'll look for another solution and I will show it to you
> before applying it to avoid further noise in git.
>
> (ftgenonce.csd works with --midi-*, my syntax error)
>
> tito
>
> ------------------------------------------------------------------------------
> Keep Your Developer Skills Current with LearnDevNow!
> The most comprehensive online learning library for Microsoft developers
> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
> Metro Style Apps, more. Free future releases when you subscribe now!
> http://p.sf.net/sfu/learndevnow-d2d
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com

------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

I have tested on WIndows and Linux with Csound, CsoundQt, CsoundVST,
and the Csound API. All works as before, thank you very much.

Regards,
Mike

On Tue, Mar 6, 2012 at 8:07 AM, Michael Gogins  wrote:
> OK, I will try this.
>
> Regards,
> Mike
>
> On Tue, Mar 6, 2012 at 7:05 AM, Tito Latini  wrote:
>> Reverted. Try the follow instr with --midi-key=4 to see the bug
>>
>> instr 1
>>  kres rand 1
>>  ival = i(kres)
>>  print ival
>> endin
>>
>> Yes, `kres init 0' (and other tricks) works, but it is to fix.
>> I'll look for another solution and I will show it to you
>> before applying it to avoid further noise in git.
>>
>> (ftgenonce.csd works with --midi-*, my syntax error)
>>
>> tito
>>
>> ------------------------------------------------------------------------------
>> Keep Your Developer Skills Current with LearnDevNow!
>> The most comprehensive online learning library for Microsoft developers
>> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
>> Metro Style Apps, more. Free future releases when you subscribe now!
>> http://p.sf.net/sfu/learndevnow-d2d
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
>
>
> --
> Michael Gogins
> Irreducible Productions
> http://www.michael-gogins.com
> Michael dot Gogins at gmail dot com



-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com

------------------------------------------------------------------------------
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing 
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Thanks for the update, and for following this through.