Hi,
A while ago (like half a year), we post some bug/problem that was "added to your defect list".
I will like to know:
- If I can consult this bug list (i didn't found it)
- When the bug will be fixed. (more or less)
My problems aren't "vital", but I will like to know if the release v8 will fix them.
Thanks for your time,
Link and defect ID:
http://www.teechart.net/support/viewtop ... polydegree
TF02011495
http://www.teechart.net/support/viewtop ... a+overdraw
TV52011730
Pending Bug (Where is the defect list?)
-
- Site Admin
- Posts: 14730
- Joined: Mon Jun 09, 2003 4:00 am
- Location: Banyoles, Catalonia
- Contact:
Hi Mariano,
We don't have an automatic bug tracking system for the clients. However providing our internal tracker ID makes easier for everyone to track issues.
TF02011495: We reviewed this issue and found that this is an algorithm limitation and can only be solved by using another algorithm for polynomial fitting. The problem is Gaussian algorithm is not all powerful and cannot deal with any number combination. For some combinations (when matrix elements are too similar) the algorithm will fail. For this cases additional check is done before the new values are calculated. If matrix elements are too similar, an exception is raised:
exception (cause « Abs(M[i,i]) < Error » in GaussianFitting). The problem is systematic with this chart.
We have added a request to our wish-list to add other algorithm implementations for polynomial fitting.
TV52011730: This is not a bug, you should add a Tools->ClipSeries tool associated to the area series. This tool does what you want.
We don't have an automatic bug tracking system for the clients. However providing our internal tracker ID makes easier for everyone to track issues.
TF02011495: We reviewed this issue and found that this is an algorithm limitation and can only be solved by using another algorithm for polynomial fitting. The problem is Gaussian algorithm is not all powerful and cannot deal with any number combination. For some combinations (when matrix elements are too similar) the algorithm will fail. For this cases additional check is done before the new values are calculated. If matrix elements are too similar, an exception is raised:
exception (cause « Abs(M[i,i]) < Error » in GaussianFitting). The problem is systematic with this chart.
We have added a request to our wish-list to add other algorithm implementations for polynomial fitting.
TV52011730: This is not a bug, you should add a Tools->ClipSeries tool associated to the area series. This tool does what you want.
Best Regards,
Narcís Calvet / Development & Support Steema Software Avinguda Montilivi 33, 17003 Girona, Catalonia Tel: 34 972 218 797 http://www.steema.com |
Instructions - How to post in this forum |
Hi,
Poly Fitt:
ok. juste so you, know. Meanwhile I did this
For the Area, I will look at it later.
thanks for the answer
Poly Fitt:
ok. juste so you, know. Meanwhile I did this
Code: Select all
procedure TCustomFittingFunction.AddPoints(Source:TChartSeries);
[...]
try PolyFitting(tmpCount,FPolyDegree,IXVector,IYVector,IAnswerVector);
except
// added cause in some case (PolyDegree > 7) this raise.
// (Exception.Create(TeeMsg_FittingError))
on e:exception do begin
if e.Message = TeeMsg_FittingError then begin
if FPolyDegree < 4 then raise; // in case of the problem don't come from this.
dec(FPolyDegree);
AddPoints(Source);//retry
end else
raise; //other problem
end;
end;//try
thanks for the answer