0
0

great, many thanks :)

  • You must to post comments
0
0

[quote="brett":25v48rut]
– Changed VB header’s boolean types to Int to be more binary compatible with the FMOD dll.[/quote:25v48rut]

This has fixed the pausing problem that Janus reported. Well it’s fixed it for me anyway! 8)

  • You must to post comments
0
0

So far, so good. I’ll let you know if I find any other stream issues.

Now, for an unrelated bug :) generatetone.exe crashes on exit if you start sinwave, or any of the other options, even if you stop the channel first. Not a big deal for me at the moment, but I thought I’d mention it.

  • You must to post comments
0
0

The problem with streams starting unpaused is indeed fixed.
The problem that I mentioned that some part at the beginning is missing seems not fixed, but I suppose it wasn’t meant to be fixed yet.
This bug also makes seeking in an mp3 not sample-accurate anymore, because sometimes you’ll get a part from about 500 milliseconds later. (So it doesn’t seem to be only at the start of the song that this bug is happening)

Also, it seems that a new bug has come up.
After re-initialize fmod (closing and reopening a new system object, and re-initializing it) it seems I can’t play any streams anymore. (Both using playSound and using readData I don’t get any data from a newly opened stream)
Opening the stream succeeded without any errors, and a sound handle is returned.
For the stream that I play with playSound, it appears that it starts playing after I seek to a new position.

  • You must to post comments
0
0

Found a serious problem (for me). FMOD_Channel_GetUserData always returns 0 in the callbacks now so I cannot change my private state information about the channel. Probably something to do with your mentioned fix to pass same channel handle to callbacks, since a call to FMOD_Channel_GetUserData right after the FMOD_Channel_SetUserData returns the right value.

  • You must to post comments
0
0

[quote="andrewboothman":1qn1dsxw][quote="brett":1qn1dsxw]
– Changed VB header’s boolean types to Int to be more binary compatible with the FMOD dll.[/quote:1qn1dsxw]

This has fixed the pausing problem that Janus reported. Well it’s fixed it for me anyway! 8)[/quote:1qn1dsxw]
IMHO, VB’s boolean type is a bit weird and should not be used. It remains a fact that this type is VB-specific and could be implemented any way the Microsoft guys fancy. Why they didn’t stick to the good ol’ (FALSE = 0 ; TRUE != 0) is beyond me.

So, I still think that, if you’re going to program in BASIC, it’s better to use -1*(Boolean condition) instead of the boolean type. But that’s coming from an old-school programmer who hasn’t touched BASIC for years (I did a year or two of BASIC after I ditched my C64 and some VB after that and that’s it).

  • You must to post comments
0
0

[quote="bloodypriest":6utj4q9a][quote="andrewboothman":6utj4q9a][quote="brett":6utj4q9a]
– Changed VB header’s boolean types to Int to be more binary compatible with the FMOD dll.[/quote:6utj4q9a]

This has fixed the pausing problem that Janus reported. Well it’s fixed it for me anyway! 8)[/quote:6utj4q9a]
IMHO, VB’s boolean type is a bit weird and should not be used. It remains a fact that this type is VB-specific and could be implemented any way the Microsoft guys fancy. Why they didn’t stick to the good ol’ (FALSE = 0 ; TRUE != 0) is beyond me.

So, I still think that, if you’re going to program in BASIC, it’s better to use -1*(Boolean condition) instead of the boolean type. But that’s coming from an old-school programmer who hasn’t touched BASIC for years (I did a year or two of BASIC after I ditched my C64 and some VB after that and that’s it).[/quote:6utj4q9a]As far as I know, VB booleans are VARIANT_BOOL, which is a signed short. CLng(True) = -1, CLng(False) = 0, Not CLng(False) = -1, Not CLng(True) = 0. If you use C++ ints and VB Longs for everything, it usually works fine. Just make sure not to go ‘== true’ in C – go ‘!= false’ or just ‘(boolean)’.

  • You must to post comments
Showing 6 results
Your Answer

Please first to submit.