MetaTrader 5 and MetaTrader 4 - page 4

New comment
 

A useless conversation.

 
xds:

Did you read my post?

Read. And saw that you are following the standard path of a novice tester trader.


In my TS I also checked - TS made for mt4 and mt5 go in different directions in terms of profitability without using pips

You may use stops of 100-300 pips or more... It doesn't eliminate principal differences!

Stops are not important. It's about filtering entries and exits into trades.

If you generate trade signals without protection (reconfirmation of the signal, the signal strength), and forced cooldown, then you'll have widely varying results from a single divergence of 1-2 pips.


An example of an unprotected strategy:

  • If SMA(X) crosses SMA(Y) from top to bottom, then open up

Example of a defensive strategy (roughly):

  • if SMA(X) crosses SMA(Y) from top to bottom and SMA(X)-XMS(Y) more than 1 pip and SMA(Y)-SMA(X) two bars back more than 2 pips, then open up
By simply adding a protective condition with a signal strength check you get a more robust Expert Advisor that does not generate a lot of signals on noise.
 
xds:

Based on your answer it turns out that pips ideas are only tested on real micro accounts and real money

Logically, the different ways of generating ticks in mt5 and mt4 should play both a negative role on the result, BUT an equilibrium positive. there is noise there. To accept that mt5 only generates fluffy and correct ticks and mt4 is a hacker I can't.

And still the difference in profits between mt5 (profits from my TS) and mt4 (losses) times I can't accept as an explanation for such a fundamental difference in results. The problem is in the code I think

Did you at least look at the pictures in the generation articles? It is clearly explained what happens and how it happens.

The tester generates ticks. And this must be clearly understood. Since you work as a pipsator.
Your strategy "catches" synchronization with movements of MT5 tick pattern and earns. But in MT4 the model is probably different.

Here's what stringo wrote in explanation of such grails. And here is an example of implementation of the same grail by vigor.

Here's a talk (or rather an attempt) about quality of modeling in MT4 - just look at three screenshots from the first post. And compare what awaits you in real life.

-----------

So in order to make further decisions - you need to find the exact spots (a couple of orders) where the two MTs diverge. And determine why the orders give different results. Wrong place to enter or wrong place to exit, fully derive all parameters of entry conditions, etc.

----------

I hope you are not testing WOC :)

 
Renat:
Read. And saw that you are following the standard path of the novice tester trader.


Stops are not important. It is about filtering entries and exits to the trade.

If you generate trade signals without protection (reconfirmation of the signal, signal strength) and forced cooldown, then you will have widely varying results from a single 1-2 pips difference.


Example of an unprotected strategy:

  • if SMA(X) crosses SMA(Y) from top to bottom, open up

Example of a defensive strategy (roughly):

  • if SMA(X) crosses SMA(Y) from top to bottom and SMA(X)-XMS(Y) more than 1 pip and SMA(Y)-SMA(X) two bars back more than 2x pips, then open up
By simply adding a protective condition with signal strength checks, you get a more robust Expert Advisor that does not generate a lot of signals on noise.

1. don't take the conversation to a flaw in my strategy (by the way I have two entry conditions) . I originally posed the question of mt4 and mt5 going apart with the same TS.

2. If you are so stubbornly plummeting and bombarding the pips strategy, then it becomes clear where your feet grow from.

 
sergeev:

Did you at least look at the pictures, in the articles about generation? It's clearly explained what happens and how.

The tester generates ticks. And this should be clearly understood. Since you are working as a pipsator.
Your strategy "catches" synchronization with movements of MT5 tick pattern and earns. But in MT4 the model is probably different.

This is what stringo wrote in the explanation of such grails. And here is an example of vigor's implementation of the same grail .

Here's a talk (or rather an attempt) about the quality of modeling in MT4 - just look at three screenshots from the first post. And compare what awaits you in real life.

-----------

So to make a further decision, you need to find the exact spots (a couple of orders) where the two MTs diverge. And determine why the orders give different results. Wrong place to enter or wrong place to exit, fully derive all parameters of entry conditions, etc.

----------

I hope you are not testing WOC :)


1.Thanks for the reply. I will take a closer look at your links

2. what is WOC? really don't know...

 
xds:

1. don't take the conversation to a disadvantage of my strategy (I have two login conditions, by the way) . I originally raised the issue of mt4 and mt5 going apart with the same TS.

2. If you are so stubbornly dumping and bombarding the pips strategy, then it becomes clear where your feet grow from.

After all, people prefer to learn solely from their experience. Personal rake way :)

Collective:

  1. You have asked questions, but do not provide factual material (reports) until the last moment
  2. You have not compared trades and prefer to ask questions on the final report
  3. You are not a programmer, you have not written your own EAs (this is done by someone else, who has even less experience in testing) and have not submitted any source code
  4. You have not yet seen the simple and basic mistakes in testing trading strategies
 
Renat:

After all, people prefer to learn solely from their experiences. Personal rake way :)

Collective:

  1. You asked questions, but did not provide factual material (reports) until the last moment
  2. You have not compared trades, but prefer to ask questions on the final report
  3. You are not a programmer, don't write your own EAs (this is done by someone else, who has even less experience in testing) and don't provide source code
  4. You still do not see the simple and basic mistakes in testing trading strategies

You've rolled me under the asphalt.

But the question is not closed ...

 
xds:

Based on your answer it turns out that pips ideas are only tested on real micro accounts and real money

Logically, the different ways of generating ticks in mt5 and mt4 should play both a negative role on the result, BUT an equilibrium positive. there is noise there. To accept that mt5 only generates fluffy and correct ticks and mt4 is a shoddy hack I can't.

And still the difference in profits between mt5 (profits from my TS) and mt4 (losses) times I can't accept as an explanation for such a fundamental difference in results. The problem is in the code I think

Such a difference is possible due to spread differences, for example.
 
marker:
I also struggled with one EA, MT5 and MT4, the only solution I found was to put up at Alpari, demo mt5 and mt4 (servers should be the same) and watch how and what opened with my eyes in "online" - this is the most reliable option to compare how the same system works on mt5 and mt4.
For control and objectivity I would have added a demo to MQ, I don't know how accurate Alpari's date is in MT5...
 
Interesting:
If you have a real account with Alpari, then I think you should test with them, but for experiment you can test with MQ...

If the real account is at Alpari, then I think we should test with them, but for the sake of experiment we can also test with MQ.

1234567
New comment