This is not exactly true. One well written expert advisor on MT will show correct result. The real problem is that there are special cases that have to be found out and calculated correctly (which is not the case of 90% of publicly available experts).

One example is the following logic:
Buy at Open(t) when Close(t) > Open(t)
Obviously, we don't know Close price at bar opening and we cannot live trade this logic. But we can perfectly fine write such logic in an expert and MT will not complain. On the other hand, FSB will not allow you to make such strategy.

I can say that MT4 provides almost unlimited possibilities to a skilfully and experienced trader/programmer. But , as we know, great power comes with great responsibility. It's not difficult to make a logical error in the coding.

FSB allows limited possibilities (due to the limited slots and logical conditions), but up to the limit of my knowledge and understanding, it's not possible to make and test a logically incorrect strategy which produces a false backtest.

any way to make generator resume the last process when we close the software and reopen it

The available feature now is generate a new strategy at every start if checked so it will start a new generation , if not checked so it will resume the last generation but with one condition'' the software didnot close''

lets say when we re open FSB and press generate so we receive a msg before generation a new strategy { do you wish complete last generation?}

another option needed its not a portfolio testing but at least it will find strategies that give positive returns for the majority of instruments easily

generator use one currency per strategy and if we need to test the same strategy on other currencies we go to currency drop list and manually go one by one

so if we have an additional step to automate that process instead-of manual selection of currency by currency it will be better 

this will test the winner strategy against the whole currencies we have  and show it as a column graph 1st column is the total returns followed by currencies  one by one  and from that step we can build a new top ten list for strategies as a total returns for the selected portfolio

is it applicable?

could we have a predefined groups of selected indicators in the generator window , it will make the fsb handy and more easier to use

fantastic 
also another idea regarding to '' strategy properties window''  which if we think to rename it , it will be Money management window  

the idea is making different predefined profiles from that specific window to test same strategy with different appetites of risk taking '' changing maximum number of open slots'' and/or '' high frequency trading appetites by altering ''handling of additional entry signal'' 

thanks in advance

Adam (ab) is developed custom sorting algorithm with different criteria for Generator Top10. I'll include it in the FSB release this week.

The feature that Popov have mentioned has some initial screenshots and ideas at http://forexsb.com/forum/topic/4033/cus … generator/ for anyone who is interested.  Once the code has been incorporated into the main release I will get some documentation completed.

The custom generator sorting mentioned above uses an external DLL file for the calculations, allowing for the code in the DLL file to be modified externally to the main FSB release to do exactly what Blaiserboy is suggesting.  While this is not really a 'Plugin' or 'Addin' that will be available in FSB Pro, the concept is similar.

You can make this. Save your desired strategy configuration as New.xml.

So i save some new empty strategies with different strategy properties and when starting new simulation i recall the needed one and start to add other slots !!!!

The first that come in my mind is to simply measure % difference between both balance lines

delta = 100 * Abs(Green-Red) / Balance;

[x ]Max red / green balance deviation % [40 ]     ---> delta < 40

i think i got the point now  , that concept will work with the end point of each curve of three as a total  but will not measure deeply the controlling factors  '' limitations'' so it will a part of a bigger solution