AmiBroker 6.02.0 BETA released

Sirius

October 30, 2015
AmiBroker 6.02.0 BETA released
Stability: Rank 3 – Regular BETA, pretty stable, should work fine in most environments

This is a BETA version. Make a backup first

A new beta version (6.02.0) of AmiBroker has just been released.

32-bit version:
http://www.amibroker.com/members/bin/ab6020beta.exe
(2 190 928 bytes)

64-bit version:
http://www.amibroker.com/members/bin/AmiBroker6020x64.exe
(10 095 488 bytes)

If you can not log in into members’ area please get your new password by filling the form at: http://www.amibroker.com/login.html

IMPORTANT: Unauthorized copying and/or distribution of materials found on members’ only page is STRICTLY PROHIBITED and will result in IMMEDIATE termination of license.

Remember to BACKUP YOUR FILES FIRST !
Note that this version can only be installed onto previous full installation of version 6.00 or higher from http://www.amibroker.com/download.html

CHANGES FOR VERSION 6.02.0 (as compared to 6.01.0)

•     AFL Editor: Implemented Line comment/uncomment feature (Edit->Line Comment, Ctrl+Q) to automatically add/remove // line comments from single or multiple lines
•     AFL: If any matrix cell has Null value then matrix product operator @ produces Null respective row/column of the result matrix.
•     AFL: if user called (a) GetPerfomanceCounter( 1 ) then (b) GetPerformanceCounter( 0 ) then subsequent call to GetPerformanceCounter(0) returned cumulated time not from call (a) but from system boot. Fixed.
•     AFL: matrix identifier can now be used in if-else statement. Such condition checks whenever very first element of matrix [0][0] is NOT NULL. This is useful for checking output of functions like MxInverse/MxSolve that would return NULL in all cells if matrix is singular
•     AFL: MxSolve/MxInverse now return a matrix filled with Nulls as a result if source matrix can not be inverted and produce warning level 2 instead of an error.
•     AFL: On Windows Vista and higher static variables use slim read-write (SRW) lock instead of critical section. This gives 5% performance increase in multithreading scenarios.
•     AFL: StaticVarAdd( “name”, value, keepAll = True, persistent = False ) – an atomic addition (interlocked read-add-write) operation for static variables
•     AFL: Study() returned NULL array when line’s start date was greater than end date. Fixed (now it works for lines drawn from right to left too).
•     Docs: Example polynomial fit formula shows how to gracefully handle singular matrix and overflow/not-a-numbers in polynomial calcs
•     Persistent static variables are now saved to PersistVars.temp and once write is successful the file is renamed to PersistVars.bin. This is to prevent data loss when writing very large sets of persistent variables.
•     When more than one error is detected in single line of the formula then the first error message is displayed instead of last one in chart/commentary/analysis.