Optimization cache

 

First of all, hi to everybody.


I think metatrader4 is REALLY a great software, in this moment I have 2 passions: learning forex via simulation and optimization and orchids (nothing to do)


To the point, i work in windows vista, build 220, 7 Nov 2008, and have some problems with cache. By the way I am a 57 years old, 37 years experience programmer.


As i am trying to optimize my EA, and have found that the optimizer uses the cache a lot, great idea, but I think there are some cases that it is not valid, I don´t know the logic you use, but it seems to me that there are some points that admit some review.


- After recompilation of the EA - the logic changed, so the results also changed.

- After changes in any parameters: dates, parameters of the EA, etc.


I should sugest than with an option the user can clean the cache.


I have tried in a lot of ways, erasing xx\MetaTrader 4\tester\caches\*, turning off the system, starting again, but always appears the message


"2009.03.05 18:24:24 Salidas: optimization stopped, 7419 cache records were used, 7252 cache records rejected"


I reallyze that it can be a misunderstanding of a user (me), but I have tried hard enough for 2 months before sending this message.


Please, any idea, any help,


gcarom

 

Hello gcarom,

This problem may be the result of Vista's "excessive care" about the users. I can suggest you to try to instal MT4 in another folder (not in Program Files) and start optimization again. May be this will help you.


Best regards

Rosh

 

Hello Rosh


Thanks for your answer.


Already I installed it in c:\tmp\MetaTrader4.... and the problem is persist.


I don´t know if i have to delete also experts\mqlcache.dat, when to do that, maybe if you tell me, es exactlly as possible, wich files i have to delete in order to start the optimization from scratch ?


thanks


gcaro

 

Hi


Please: what is the meaning of


"2009.03.05 18:24:24 Salidas: optimization stopped, 7419 cache records were used, 7252 cache records rejected"


There is ALWAYS records from previous optimization, or it can be records from CURRENT optimization, i.e. the genetic algoritm tries twice exactly the same point and rejects the second time ?


Thanks


gcaro

Reason: