Errors, bugs, questions - page 903
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Is it realistic to get previous x32 builds of MT5? Really need to get back into the summer. One of my recent posts in this thread complaining about "MemoryException 180772428 bytes not available" in Log tab and "out of memory" in Experts tabhaven't got rid of them after installing clean OS and new MT5 (which I managed to do only recently), so I conclude the problem was not with littered and damaged old operating system and not with my code (that was last modified in April), but with next build of MT5 somewhere in second half of summer. And then it went on and on... No errors had occurred before that build.
I did not remember it at once, because, firstly, I was afraid of my own local problems, and secondly, I had very little time to do it in time. Shifting my code to developers for analysis - I do not believe that they want to change it from bad to worse (besides, developers have their own things to worry about), but even if they agree, I will have to take the trouble to simplify the code - and that for me now is almost harder than to develop it.
I also want to mention, that DDR was run with test programs and no errors were found; moreover, one 1Gb module was replaced by 4 new 1Gb cards of another famous vendor, they are all right too, but error keeps creeping up. The system is not overclocked. Even if my code is not written optimally, that's not the main thing now, as it worked without errors before anyway.
One of my recent posts in this thread complaining about "MemoryException 180772428 bytes not available"....
On x64 (I don't use x32 anymore) this suggests that the memory is not enough and the swap file size needs to be increased (with all that implies - slowdown during optimization for long periods, etc.).
I will definitely give it a try, but I'm still fundamentally curious as to what kind of screws the developers have put in place that are now different. Before, with the same paging file size and other things being equal, everything ran smoothly.
By the way, one of these days I'll have to test the code on an x64 8-pack too, but I suspect my 4Gb will turn out to be something like 2Gb for it, which again is not enough. And physically all the DIMM slots are full, and also, according to the User Guide on the motherboard, a technical total maximum of 4GB has been reached (although that may just be a limitation of that era, not an absolute one).
build 740, x86
I optimize my Expert Advisor in "Balance+min Drawdown" mode and it feels like it only tries to optimize the drawdown. As a result of the optimization I get a drawdown of 1% and profit of 8% of the initial deposit for the year. A week earlier the results were quite different.
build 740, x86
I optimize my Expert Advisor in "Balance+min Drawdown" mode and it seems that it tries to optimize only the drawdown. As a result of the optimization I get a drawdown of 1% and profit of 8% of the initial deposit for the year. A week earlier quite different results were obtained.
We have excluded results with a negative balance from the calculations.
Let's talk to Service Desk, maybe we have corrected something wrong. But please give us the most detailed calculations possible (so that we can get to the bottom of it quickly)
The results with a negative balance were cut off from the calculations.
Let's talk to the Service Desk - maybe we've got something wrong. Please give us as much detail as possible (so that we can sort it out quickly)
Opened application #617331
You didn't provide any data there.
I'll repeat my question once again - it seems that Custom Max result is counted incorrectly in the optimizer.
The Expert Advisor in OnTester() returns Profit * Trades * (1/(MaxEquityDDPercent+1)) * RecoveryFactor.
Here are the optimization results:
It was expected that with more profits, number of trades and Recovery Factor, and with less drawdown, the result would be larger. Which is not the case.