Bug in TA 0.984 with TimeWarp

[Ed Carstens]

TimeWarp is an excellent feature but I just encountered a bug with it today in the latest version (0.984). I noticed that one of my signals (wbuf_full) was low after the warp instead of high. The wbuf_full signal transitions from low to high before the time warp and should stay high for the remainder of time shown, but after adding the time warp, it starts low right after the "..." and remains low. Attached is the example (wbuf1.tim).

[Dan Fabrizio]

Hi Ed,

I fixed the first problem you reported in the previous message.   The last value time in the vcd file was being used for the timing diagram end time.  This caused the problem so I made the end time greater than the last value change time. 

I'm working on this reported problem now and should have it fixed soon.  These problems you are finding are a result of major changes made to the timing engine in the program in beta 0.980 which increased the time accuracy to femtoSeconds.  This was needed for true logic simulations which I"m adding now so there will be 2 modes of operation,  simulation or edit mode, in an upcoming version beta 0.990. 

I expected some problems considering the extent of the changes made but don't have regression tests in place to catch issues when changes are made so I do as much manual testing as possible and hope to get feedback from beta users like you.   I'm adding unit regression tests now as well which will help find issues before new versions are released 

Thanks for you feedback,

Dan

On Wed, Jun 3, 2015 at 4:47 PM, 'Ed Carstens' via TimingAnalyzer <timinga...@googlegroups.com> wrote:

TimeWarp is an excellent feature but I just encountered a bug with it today in the latest version (0.984). I noticed that one of my signals (wbuf_full) was low after the warp instead of high. The wbuf_full signal transitions from low to high before the time warp and should stay high for the remainder of time shown, but after adding the time warp, it starts low right after the "..." and remains low. Attached is the example (wbuf1.tim).

--
You received this message because you are subscribed to the Google Groups "TimingAnalyzer" group.
To unsubscribe from this group and stop receiving emails from it, send an email to timinganalyze...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Dan Fabrizio
The TimingAnalyzer
www.timing-diagrams.com

[dan_fabrizio]

Hi Ed,

I fixed both issues in the beta 0.985 version.  

Try it when you get a chance.

Thanks, Dan