New MetaTrader 4 Client Terminal 387 and MetaTrader 4 Data Center build 387 - page 11
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
So, what Slava is trying to say is that the 225 build shows exactly the same effect with butterfly removal.
Please check your code carefully. Apparently, there is no effect from re-initialization of buffers.
Thank you. I'll check it out.
The recalculation with the wrong number of bars pointed out by AlexSTAL has been corrected. But this is a very old bug.
Thanks for checking, it's now clear which way to dig.
We'll try to find it, fix it and issue an update as soon as possible.
What about off-line charts? When updating off-line graphics using PostMessageA(hwnd,WM_COMMAND,33324,0); re-initialisation occurs on every artificial tick!!!
There's also a complete refresh taking place. Refresh command
There's a complete refresh taking place. The Refresh command
A full refresh takes place there. Refresh command
This has never happened before. My indicator has stopped working with optimisation. Now I have to fill the buffers for every such tick. Wrote about it above.
Now window refresh clears indicator buffers.
Strange.
Started checking on 229 build yesterday after Slava posted the logs. There was the same error as in 388. I got the same crashing pictures in 229 build as well.
The main thing there is not the butterflies. The zigzag disappeared, i.e. buffers were zeroed. That is why graphic constructions were disappearing.
I decided to think it over. Today I tried again. I also put indicator with default settings. It works fine in both build 229 and 388. The only difference from yesterday's test is this. I have decided to test it since old versions to analyze which version the error has appeared. All versions of the indicator work properly. No errors. Buffers are not reset. I have not encountered this yet.
The indicator does not work with global variables. It does not save information for the next session in global variables. Therefore, old versions, launched first for testing, could not leave any information in the terminal that can affect the operation of later launched versions of the indicator.
May be there is something wrong with your computer? But how can the computer affect the operation of the terminal? Affect it in such a way that during testing the buffers are reset at random points in time. It's a mystery.
Now testing in the 388 build is running smoothly with absolutely the same initial data as yesterday when there were failures.
Strange.
Started checking on 229 build yesterday after Slava posted the logs. There was the same error as in 388. The same crashing pictures came out in the 229 build as well.
When PPC and I were testing ZigZag, we stumbled upon a huge number of glitches related specifically to the starting point of building ZigZag.
Probably just a combination of factors - starting point, number of bars, etc.
This has never happened before. My indicator has stopped working with optimisation. Now I have to fill the buffers for every such tick. Wrote about it above.
Now window update clears indicator buffers.
This didn't happen because there was a bug with the change counter count. A full override means that some data within the buffer may have changed. We cannot guarantee the integrity of the data in the offline graph.