bug found. attempting to make xfade on ping-pong loop leads to crash

It doesn't crash here during a quick test. I loaded an xm, turned the loop of the first sample into pingpong, marked some data around the first loop point and used xfade, works fine. Also works on 16-bit samples. How did you use it?

started ft, loaded 16bit sample (~300kb), set loop points, selected some area around the first marker, pressed xfade and ft crashed, several times all the same

seems that it crashes if the selected area is too big and it happens only with pingpong loop, checked it on two dos computers

I'll look into this tomorrow or something.

I still can't trigger any crash, not even on a Pentium 233 laptop using the latest version of FT2.11. I try to load a rather big 16-bit sample, I set a pingpong loop, I mark some data on the left loop pin and press xfade. Works every time. Do you use a GUS sound card, by any chance? Are you also using the latest version? When you quit the program, it should say "released 3rd of April 2020".

i'm using 2.11 with no midi prg change, you built it for me some weeks ago. sound blaster awe 64 inside p233 mmx with 128mb ram. another pc is ibm thinkpad 600e with on-board sound. i can film short video of the issue if that helps. not sure how to upload it though...

8bitbubsy wrote on 2020-05-17, 15:17:

I still can't trigger any crash, not even on a Pentium 233 laptop using the latest version of FT2.11. I try to load a rather big 16-bit sample, I set a pingpong loop, I mark some data on the left loop pin and press xfade. Works every time. Do you use a GUS sound card, by any chance? Are you also using the latest version? When you quit the program, it should say "released 3rd of April 2020".

and yes, it says "released 3rd of april 2020"

i've just noticed this crash report!
https://drive.google.com/open?id=1DowOhq8slg1 … QnD0iibqFPYlW9R

i did almost the same but instead of loading sample i recorded it

I can't fix this issue for now, since I can't trigger it at all!

the same crash happens in the last "official" version too... i'll put the video here in a while, perhaps i'm doing something i can't comprehend myself...

Ok. Are you running FT2 from a true MS-DOS environment? (not while Windows is running).

sure! i don't even have windows on those machines, just pure dos 6.22

8bitbubsy wrote on 2020-05-18, 10:04:

Ok. Are you running FT2 from a true MS-DOS environment? (not while Windows is running).

here's the video of what happens
https://drive.google.com/open?id=1rNkMJTZU8iN … wziVMJdYmGbrYmn
hope this will help somehow

traffkin wrote on 2020-05-21, 05:06:

here's the video of what happens
https://drive.google.com/open?id=1rNkMJTZU8iN … wziVMJdYmGbrYmn
hope this will help somehow

I tried loading a 16-bit stereo sample like you did, selecting the left part, making a similar pingpong loop, select about as much as you did and press xfade. No crash...

EDIT: Oh, the GUI actually locks up, so I guess I'm hitting the same bug as you, just without a crash to DOS. Will try to look into the pingpong 16-bit part of the xfade code today, but I can't promise that I figure it out.

that's odd but it never happens on 8bit samples. just converted 16bit sample to 8bit and it crossfaded flawlessly

Yes, it's because of a division by zero. I'm trying to fix it. I also found out that the xfade algo is wrong on 16-bit poingpong, not just buggy.

Bug fixed.

Changes in v2.11:

• ...
• XFade didn't work correctly on 16-bit pingpong samples and could crash the program
• ...

https://16-bits.org/etc/FT211.ZIP
"No MIDI-out program change" version for traffkin: https://www.dropbox.com/s/b624qekij49get9/FT2 … DI_PRG.ZIP?dl=1

IT WORKS!!! thank you so much!!! you're genius 😀

Thank you so much for updating FastTracker II! This was a very unexpected project. I recall learning how to use trackers in the late 90s with version 2.04. It was my introduction to music altogether. To see any updates since 2.09 is like a dream come true. I can't wait to return home and try this out.