Asynchronous loading of charts - does MetaQuotes really know how many problems this creates? - page 2
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
I am tired of polemic answers, sorry.
One should not always take ones own situation as a benchmark when someone describes a problem. From time to time, one should rather assume that a questioner knows exactly what he is doing, in other words it doesn't help if one claims that this is not a problem just because one can handle it oneself in the own tasks. Other developers, other tasks.
------------------
------------------
I can reduce all this to one single question:
1. how can I basically retrieve the required quote data at any time from any timeframe without any misbehavior?
If there is no answer, and no perfect workaround, it´s something that should be improved, because I am not the only ones who needs this. I am also in contact with other developers, not only in my team, also others who totally agree on this and who are also tired of creating workarounds, but also tired posting and describing problems like these here and who think it´s a waste of time, cause issues like these are usually ignored by MQ.
However, this is why addressed the topic mainly to MetaQuotes directly, to help them to understand that there are circumstances/tasks in which this behavior results in massive, serious problems. And ours is not the only one. If they are willing to help here, they will help many other programmers as well, also all my partners and finally all our common clients.
You already posted a lot about issues (mainly MT5/mql5) on the forum. I have personally no doubt that you really have these issues of course, but that's not the problem. As long as you don't provide ALL the relevant information for EACH specific issue to be able to reproduce this specific issue, you will never be heard by Metaquotes. And nobody on the forum will be able to help you. I have repeated you that a lot of time, but it seems you don't want to listen. It's your right, but you should listen to me if you want improvements. I will repeat it a LAST time :
The only way to get support for technical issues is through the forum, and the only way to get attention from Metaquotes is to provide them REPRODUCIBLE issue(s). They clearly stated that they will not even check or ask more details if you don't provide it from the start yourself. I will not post the reference to prove what I wrote here, it's a fact, you can find it on the Russian forum. I have helped to fix a lot of bugs, how much did you succeed to have fixed ? LISTEN !
Most of the topics you opened these last years are just wasted time. I am really interested to have all what can be fixed, fixed, and that's why I always tried to help you. (As you said correctly "they will help many other programmers as well"). But I am also tired to repeat you always the same and being never heard.
There is 0 usable information on this topic currently. You will not be heard. The bugs will not be fixed. The platform will not improve. Unless you change your mind because I am sure Metaquotes will not change its way to work for you.
People can learn from you, including myself, but you can also learn from people. I am sure in what you reported, there are real MT5 bugs, but also structural behaviour you will have to deal with (they will not change the core of their platform to facilitate your work, and when you push the platform to its limits, there are real technical difficulties) AND also errors YOU made or missed some useful techniques or whatever that you could be improve on your side. All of the 3 points, if you don't accept that, it's not even worth to say anything more.
If you want to start a constructive topic with a practical goal, I am right here to help and to invest time, as also alphatrading (though I think he missed an important point that MT5 is NOT the same as MT5) and maybe others. Up to you to decide now, I think I was very clear. All the best, whatever you will decide.
The only way to get support for technical issues is through the forum, and the only way to get attention from Metaquotes is to provide them REPRODUCIBLE issue(s). They clearly stated that they will not even check or ask more details if you don't provide it from the start yourself. I will not post the reference to prove what I wrote here, it's a fact, you can find it on the Russian forum. I have helped to fix a lot of bugs, how much did you succeed to have fixed ? LISTEN !
It seems like an easy task but in reality it is not. I experienced some issues that are not reproducible as easy as it sounds. In 95 % MQL functions work as they are supposed to be. But it's the leaving 5% which drives a programmer crazy. And as hard as I tried I could not produce a working example to illustrate those issues for 100%. Most of the time everything works fine but sometimes - and who knows why and when - it does not. I know my answer is vague and doesn't provide any additional value to the topic probably, but what's left for us programmers? The only thing in this matter is to point to a direction and ask Metaquotes take a look under the hood and think about a potential issues. At least this is what I do when a customer of mine reports a problem with my programs.
https://www.mql5.com/en/forum/313720
Samples related to this topic, much earlier (3 years old issue)
https://www.mql5.com/en/forum/332849/unread#unread
Proven sample, not fixed
https://www.mql5.com/en/forum/328410
Clear & proven, but no reaction.
https://www.mql5.com/en/forum/326220
Clear, not fixed, no reaction.
Almost all the other topics were either mistakes from my side - which I usually confirm - or could be solved by other developers or could be solved by myself - and I added the solution. By the way, thank to those who helped.
In view of reproducibility: Not all misbehavior can be reproduced by a simple sample from the scratch. This topic describes a behavior which happens in maybe 1 of 1000 cases, maybe 1 of 500, in my case almost never, but it happens and the "almost" is the problem. Now you can check out the first link again and then state that it did not happen in your first test-case and then say: "There is no bug". But actually you have to run it a thousand days to prove this. If you don't want to run it 1000 days, you could also believe what I write, because I don´t post such things just for fun. In cases like this, I have a huge amount of users in the background who deal exactly with this issue day by day. And if you don´t test it like it should be tested and if you don´t believe and just drop it, then it was a waste of time for me and you are right again - exactly.
In my opinion, the quality of some answers in this forum is the problem, not the questions. Everybody can read theirselves whats and especially how has been written here. I just can say, I appreciate any help, because I also overlook things from time to time and make mistakes, nobody is perfect. But, at least for my understanding, trying to help or support someone works different.
It seems like an easy task but in reality it is not. I experienced some issues that are not reproducible as easy as it sounds. In 95 % MQL functions work as they are supposed to be. But it's the leaving 5% which drives a programmer crazy. And as hard as I tried I could not produce a working example to illustrate those issues for 100%. Most of the time everything works fine but sometimes - and who knows why and when - it does not. I know my answer is vague and doesn't provide any additional value to the topic probably, but what's left for us programmers? The only thing in this matter is to point to a direction and ask Metaquotes take a look under the hood and think about a potential issues. At least this is what I do when a customer of mine reports a problem with my programs.
You nailed it. Thank you.
It seems like an easy task but in reality it is not. I experienced some issues that are not reproducible as easy as it sounds. In 95 % MQL functions work as they are supposed to be. But it's the leaving 5% which drives a programmer crazy. And as hard as I tried I could not produce a working example to illustrate those issues for 100%. Most of the time everything works fine but sometimes - and who knows why and when - it does not. I know my answer is vague and doesn't provide any additional value to the topic probably, but what's left for us programmers? The only thing in this matter is to point to a direction and ask Metaquotes take a look under the hood and think about a potential issues. At least this is what I do when a customer of mine reports a problem with my programs.
I fully understand you. But we have only 1 way to deal with it, it's to take the reality as it is and try our best to change what can be changed.
As I offered to Dirk, and I make the same offer to any serious programmer like you seem to be : let's stop the complaining, let's collaborate, let's identify the problems clearly. And stop thinking we know all each one on our side. If not doable publicly, let's do it privately. Let be constructive and ACT instead of talking.
Let's start with one problem at a time. Describe it as best as possible and we will see what we can do together to understand it. Identify the root cause and THEN report it to Metaquotes.
https://www.mql5.com/en/forum/332849/unread#unread
Proven sample, not fixed
Let's take this one as it's a good example. Yes it's proven, for me as for you it's clearly a bug. I took the time to discuss with Slava, which is one of the main Metaquotes developer and he clearly stated that for him it's not a bug, and nothing will be fixed.
So what to do with that now ? Nothing. Even a clearly identified, reported and reproducible issue will not be fixed, so guess what will happen to "The only thing in this matter is to point to a direction and ask Metaquotes take a look under the hood." They just don't do that.
I did all what is possible. So I will live with it and workaround it when needed. No complain, no hard feeling, let's move on.
Next problem...
https://www.mql5.com/en/forum/313720
Samples related to this topic, much earlier (3 years old issue)
I don't have any problem to believe you but I was not able to reproduce it. If you want we can try again. What would be the best way ? To run this code simultaneously on a maximum of charts ?
I don't have any problem to believe you but I was not able to reproduce it. If you want we can try again. What would be the best way ? To run this code simultaneously on a maximum of charts ?
Yes I think this would be the best method. But as mentioned, it happens rarely and it seems that it happens on specific dates, cause it´s like that, that I do not hear nothing from any client for a while, but then there are several clients within 2-3 days. It does not necessarily mean, that it does not happen every day at someones computer. This timing could be broker-server related, actually I can only guess here, but in fact its not a broker-server problem, we checked this simultaneously of course couple of times when there were more problems as usual. It seem to happen locally.
As I offered to Dirk, and I make the same offer to any serious programmer like you seem to be : let's stop the complaining, let's collaborate, let's identify the problems clearly. And stop thinking we know all each one on our side. If not doable publicly, let's do it privately. Let be constructive and ACT instead of talking.
Let's start with one problem at a time. Describe it as best as possible and we will see what we can do together to understand it. Identify the root cause and THEN report it to Metaquotes.
Thanks for the offer Alain. I will definitively come back to it :-)
https://www.mql5.com/en/forum/313720
Samples related to this topic, much earlier (3 years old issue)
https://www.mql5.com/en/forum/332849/unread#unread
Proven sample, not fixed
https://www.mql5.com/en/forum/328410
Clear & proven, but no reaction.
https://www.mql5.com/en/forum/326220
Clear, not fixed, no reaction.
Almost all the other topics were either mistakes from my side - which I usually confirm - or could be solved by other developers or could be solved by myself - and I added the solution. By the way, thank to those who helped.
In view of reproducibility: Not all misbehavior can be reproduced by a simple sample from the scratch. This topic describes a behavior which happens in maybe 1 of 1000 cases, maybe 1 of 500, in my case almost never, but it happens and the "almost" is the problem. Now you can check out the first link again and then state that it did not happen in your first test-case and then say: "There is no bug". But actually you have to run it a thousand days to prove this. If you don't want to run it 1000 days, you could also believe what I write, because I don´t post such things just for fun. In cases like this, I have a huge amount of users in the background who deal exactly with this issue day by day. And if you don´t test it like it should be tested and if you don´t believe and just drop it, then it was a waste of time for me and you are right again - exactly.
In my opinion, the quality of some answers in this forum is the problem, not the questions. Everybody can read theirselves whats and especially how has been written here. I just can say, I appreciate any help, because I also overlook things from time to time and make mistakes, nobody is perfect. But, at least for my understanding, trying to help or support someone works different.