Why some great coders and trading system developers are ignoring Metatrader 5? - page 8
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
Backtesting for MT4 is related to coding? If not - no problem because I am not fan of it.
Coding for MT4 on high/low of the bar, on open bar or in MTF way and selling EAs based on backtesting results on every tick - it is realted to the coding?
Ok, no Market in mql4 - no problem.
But Job - there is some arbitration for example? Internal one or connected with some legal institutions?
Can anyone share experience with it?
I used Job but indicator was coded by my friend and I trusted him.
And what about EAs?
How simple man can know that he received normal EA according to what he described?
Does the person should describe everything on the way as "value of this indicator on bar #1 < the value of this indicator on bar #2 & .... " ....
or ... where is the section of this forum or some threads to receive the information about it?
What is general opinion of the users/clients of Job who requested EAs for example?
let's share experience how the coders are coding for the people, which rules, we like those rules or not.
=========
Hi RaptorUK,
Those my questions are not off-topic?
if it is off-topic so please - can you tell me the section of the forum inside mql4 where I can receive this information from the people?
because a lot of aggression ... you know my case ... and the other case on Job was yesterday where one client opened the job only because he wanted to punish coder publicly ...
If we can receive many opinions (any opinions, irrespective off) so we can appear to metaquotes to change something (if it is related to them for example).
And you know my opinion about profile vs anonymous.
Does the metaquotes are checking identity of the coders on Job service?
Because I do not know ... I just paid ... I am not a coder.
of course, I will check the threads here for same information.
but it may be good to collect everything in one place with discussion.
because it is the only way to change something if we want so.
If you want to talk about the Jobs section on the mql5 site you should do it in the General section of the mql5 site, this forum does not have a general section and is for mql4 coding discussion.
It seems to me that the best solution to your problems would be to learn to code mql5.
ok.
I will talk about coding for MT4.
I told about the problems with backtesting for MT4 related to the coding.
Thanks that you did not delete my post.
For now - I am going to open the thread in mql5 or to make few posts related to PR technologies.
Using this thread - the people can easy recognize about "who is who" for example, and what to do in some cases.
I hope, moderator of mql5 will not delete my thread because MT5 platform is social networks oriented.
You guys are having very narrow frame work so it is difficult to do something even if it is required for example.
:)
I didn't read this entire thread so I am not sure if these points were covered, however, I'll just provide my own personal rationale for not switching from MT4 to MT5
I have about 30 years of coding experience but I have been coding in MT4 for 2 years now. I have written 10's of 1000's of lines of MQL4 code. I make extensive
use of hedging. These are the 2 most important things to me and both were ignored in MT5. MT5 ignored 20 years of legacy code when some type of compatability
box could easily have been engineered into the product. This meant that there had to be very compelling reasons to rewrite (yes, migration doesn't quite capture the
level of effort required) existing code and the reasons just don't exist, especially when the lack of hedging essentially disallows an entire class of trading strategies.
Again, an alternative book management option should have been included (obviously, they had the code already) so that those choosing not to use an American broker
could continue to use hedging strategies. I would love to have the additional order semantics, a more fully-fleshed out coding language and, perhaps, faster execution
but the price I will have to pay for these is unacceptable to me now. Things I don't care about include better graphics, real-time news, social-media integration and
other fluff that doesn't help me make money.
Hi phredotaku,
Yes, you are right.
But it is very personal. For example - I did not use hedging.
As to the coding - I am not a coder. Of course, I understand inside the code in mql4 and I can program some simple things basicly using templates or examples of the EAs created by me some time ago. If I start to code something very new - it will take very long time for me because I will have to read ... and read ... how to code ...
In mql4 - I can take few indicators and can create just one indicator using those few ones as icustom.
As to mql5 - it may be slightly different. I understand the logic. But if I am bad coder for MT4 so I am completely nothing for mql5 :)
I remember the time when I started to learn coding for MT3 - I printed some lessons and reading ... reading ... it was not lessons from mql websites (it was the lessons when coders helped each other to learn some programming language quickly). When I almost finished - MT4 was created and I had to learn mql4 instead :) (from external websites too - as free help of coders to coders). But I do not see the lessons on forex forums uploaded on the way as 'coders are helping the coders'.
I do not want to be on the same situation once again - MT5 is already started, and if MT4 is finishing (which we all do not want) so ... I need to be dropped to MT5 anyway :)
Because in case of MT5 - I need more time. It is not as moving from MT3 to MT4. Moving from MT4 to MT5 will take more time for me.
About social networks - I am not commetrcial guy and not selling anything, but I am using social networks in extensive way. And it takes long time for me to maintain all blogs etc for good free content. If MT5 will be real generator of the content so it will be really good. Because as I see - MT5 is started it's way as social networks oriented platform. Just started. If it will be continuing so it will be good for me - I will spend may be ... 1 hour in a day for all my social accounts uploading interesting content to all my social networks friends.
Besides - backtesting.
I do not like backtesting in general.
But it may be necessary to backtest sometimes just to find some preliminary settings for the EA.
If we will compare backtesting for MT4 with MT5 - big differences.
Of course, the ticks are emulated etc ...
I was reading some post of metaquotes that they are thinking to create some kind of MT4 emulator inside MT5. That is true?
Besides - backtesting.
I do not like backtesting in general.
But it may be necessary to backtest sometimes just to find some preliminary settings for the EA.
If we will compare backtesting for MT4 with MT5 - big differences.
Of course, the ticks are emulated etc ...
Yes, I agree about tick data for MT5 ...
But I am using strategy tester for optimization only (just to find possible settings for the EA) - in case of MT4 and MT5 as well.
About backtesting ... I am having negative opinion about backtesting in general especially related to commercial EAs (irrespective off - MT4 or MT5).
For example: one person can sell EA somewhere in internet, no one knows him, and he is selling it based on backtesting results only.
He will not disclose the code unless someone will buy his EA (as ex4 file? decompillation is illegal ...).
EA can be coded on open bar, or on high/low of some bar, or indicators inside EA may have different values for different broker (iTrend indicator for example) ...
Backtesting is not valid is this case.
And backtesting is not valid in case of MTF EA (EA is attached to M1 timeframe using taking indicators's values from H1 and H4 for example).
Besides, I do not think that backtesting results can be 'evidence' that EA is profitable as commercial product.
So, I am using strategy tester just to optimize the settings for EA.
Only.
Yes, you are right about tick data for MT5 - M1 data may be real one but ticks are emulated.
But ... the brokers are having different datafeeds especially in lower timeframes so ... even if we will backtest EA for MT4 (or with MT5) - real situation with EA may be different, and it may be different from one broker to an other.
What I like in MT5 related to this subject: I can find the settings for MTF EAs (it is impossible for MT4), and I can make multi currency optimization to find the settings for all the pairs.
Besides - Cloud. I am not sure - do we have Cloud for MT4 or not ...
I tried to optimize the settings for few EA (which I created for my mql5 threads using MT5 Wizard and Milanis builder) - I spent 2 days for that.
I want to try the Cloud (as I know - it is not really expensive to use this Cloud) - just to speed up this process (to find the settings).
But ... it may be the other thinking: real robust manul trading system does not need optimization in case of EA - if we (or someone) traded this system manually with good results for the long time so ... why we need to optimize it for 'more good settings'? ...
But this case is not related to strategy tester sorry ...
And backtesting is not valid in case of MTF EA (EA is attached to M1 timeframe using taking indicators's values from H1 and H4 for example).
Besides, I do not think that backtesting results can be 'evidence' that EA is profitable as commercial product.
What I like in MT5 related to this subject: I can find the settings for MTF EAs (it is impossible for MT4),
And how MT4 strategy tester is receiving the quotes for all those timeframes in case you are backtesting it with M15 only?
You mean: are you comparing backtesting with forward testing/trading results?
I did it for Alpha and Electra EAs (those EAs are also MTF EAs) - results of backtesting and forward testing are different ...
You can backtest this EA with MT strategy tester?
And how MT4 strategy tester is receiving the quotes for all those timeframes in case you are backtesting it with M15 only?
You mean: are you comparing backtesting with forward testing/trading results?
I did it for Alpha and Electra EAs (those EAs are also MTF EAs) - results of backtesting and forward testing are different ...
Yes I tested it with the MT4 Strategy Tester. The Strategy Tester has access to the other timeframes . . . . how can you not know this and at the same time tell us how wonderful MT5 is ? it does nothing for your credibility.
From here: Testing Features and Limits in MetaTrader 4
"At every-tick modeling, tester pumps all necessary timeframes for the symbol under test independently"