MetaTrader 5 Strategy Tester: bugs, bugs, suggestions for improvement - page 12
![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
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
Error in the Tester when selecting "All history". The last day of history (< the current date) of the custom symbol does not fall into the test interval.
It turns out that if you set the whole history manually, the single run will have more days than the "Whole history" mode.
The last 24 hours were never tested.
For market products demo - last 7 days
The last 24 hours have never been tested.
For market products, the demo is the last 7 days.
You misunderstood. The custom symbol contains history up to and including the 2nd of October. I do "All history" and the interval is up to and including the 2nd of October. And today is the 10th.
Please fix the Tester's work with custom symbols. They are not selectable in the Tester. I have to reload the Terminal for that. And so every time I need to change a symbol.
It just won't reproduce like that.
Describe the exact steps to reproduce the problem
Isthe custom symbol accurately selected in the market overview?
It just won't reproduce like that.
Describe the exact steps to reproduce the problem
Is the custom symbol accurately selected in Market Watch?
Running single runs alternately on each custom symbol. From a certain point it is impossible to see the path to them in the Tester, but they are still selectable via drag and drop from the Market Watch. Then after a while this doesn't work either: you can select either a normal symbol, or only one custom symbol. The others are not possible. The animation below shows this situation.
Started 3 testers in run mode, scattered them on different CPUs, ended up withPAGE_FAULT_IN_NONPAGED_AREA screen of death
I've added more memory (32GB) and each tester is running on its own SSD.
I'll try rolling back to a previous version, maybe it won't crash. I'll try to collect logs and other stuff after the runs, because I need the results now and the runs last 12-20 hours each :( My multicurrency run goes straight to 27 characters on 4 years... During the week 4-5 times the computer already went down, it's a shame...
On my old tester I didn't notice such things. Maybe it is the new hardware (memory and ssd) - I will look at it again.
On HDD when searching genetic algorithm in 12 threads in 15 years the computer frightened, moreover, old tester have coped with it, new one just rebooted the computer after few messages about insufficient swap file (up to 48 GB) and was eating memory. As soon as I'm done with the runs, I'll move back to 15 again, but on SSD (need about 1TB on the drive for that).
If Metakvotes has a supercomputer with 16+ cores and 64+GB RAM + 1+TB SSD, can you let me have it for tests? I'll run my bot there in a genetic algorithm for 15-20 years (1 character per search unit). And then there will be a run of 27 pairs)))) Here will be the logs for you)
PS. On the settings - every tick on the real ones) Without any acceleration. Only genetics, only hardcore)
PPS. Preliminarily, 2 new memory cards didn't start (or one of them). I'm remote yet and can't exactly figure it out. Will try running the runs again)
Ran 2 testers - memory already at 100%... Pumped up 2x48GB selected. Won't run the third one yet and take any chances.
Started 3 testers in run mode, scattered them on different CPUs, ended up withPAGE_FAULT_IN_NONPAGED_AREA death screen
I've added more memory (32GB) and each tester is running on its own SSD.
I'll try rolling back to a previous version, maybe it won't crash. I'll try to collect logs and other stuff after the runs, because I need the results now and the runs last 12-20 hours each :( My multicurrency run goes straight to 27 characters on 4 years... During the week 4-5 times the computer already went down, it's a shame...
On my old tester I didn't notice such things. Maybe it is the new hardware (memory and ssd) - I will look at it again.
On HDD when searching genetic algorithm in 12 threads in 15 years the computer frightened, moreover, old tester have coped with it, new one just rebooted the computer after few messages about insufficient swap file (up to 48 GB) and was eating memory. As soon as I'm done with the runs I'll move back to 15, but on SSD (need about 1TB on the drive for that).
If Metakvotes has a supercomputer with 16+ cores and 64+GB RAM + 1+TB SSD, can you let me have it for tests? I'll run my bot there in a genetic algorithm for 15-20 years (1 character per search unit). And then there will be a run of 27 pairs)))) Here will be the logs for you)
PS. On the settings - every tick on the real ones) Without any acceleration. Only genetics, only hardcore)
PPS. Preliminarily, 2 new memory cards didn't start (or one of them). I'm remote yet and can't exactly figure it out. Will try running the runs again)
Ran 2 testers - memory already at 100%... Pumped up 2x48GB has been selected. Won't run the third one yet and take any chances.
Andrew, BSODs occur when there is a failure in kernel mode.
Cause driver or HW
Check for overclocking, especially if it is NoName memory.
Bad SSD could also be the cause, e.g. if there are problems with swap file (but this case is kinda not your case - NONPAGED_AREA).
And in the Tester it appears as follows
Made
And in the Tester, it appears as follows
I have bool UseOption = false;
Sometimes it can be 0 in the tester, and sometimes it can be false as it should be
Andrew, BSOD happens when kernel mode crashes.
The cause is driver or HW.
Check for overclocking, especially if it's a NoName memory.
Bad SSD could also be the cause, e.g. if there is a problem with swap file (but this case is kinda not your case - NONPAGED_AREA).
I have corsairs, but the ryzen and their mothers have problems with the slats. I got the exact same ones as the other two though. I'll have to see what's wrong.