MetaTrader 5 and MetaTrader 4 - 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.e. the conclusion is still:
1. maybe the code is not reliable
2. the strategy is not reliable and cannot be coded to get the results i need on mt4 and mt5 at the same time
3. 1 и 2
1. If we assume that the strategy was maximally fitted to the trading conditions of both terminals (in the logic of experts, there is no great difference) when performing a large number of transactions, the results on both terminals will be "identical" (with a certain margin of error).
2. Almost all MT4 strategies are transferable to MT5, of course there are exceptions.
1. Assuming that the strategy has been adjusted as much as possible to the trading conditions of both terminals (there is no great discrepancy in the experts' logic), the results on both terminals will be "the same" (subject to a certain margin of error) when a large number of trades are made.
2. Almost all strategies that trade on MT4 can be transferred to MT5, of course there are exceptions.
Ahh...
My code is the simplest one, without indicators.
I tried to use it with some simple indicators and at first I tried to use it for mt5.
I think the programmer is messing around and cannot find his error. I have already paid the money.
I understand that the results may differ by 10-30%, but mt5 makes a profit, and mt4 just blows away... The difference is several times bigger...
For example, after three years of testing mt5 gives profit +X, and mt4 gives loss -2X
A test on real quotes is real trading with real money. To get reliable data on the profitability of the selected TS you will need to trade for a year or two
Or I do not understand something?
1. Why do we need testers at all?
2. For example: "History is stored in bars". Then for minutes that would mean "All ticks" . There are no lower TFs.
1. For clarification what I mean by real quotes - This is the current quotes provided by the broker/dealer on a real or demo account (taking into account that there may be differences between the real and demo).
2. the tester is needed to test your ideas and for a relatively fast run on the history (which is important) of the Expert Advisor, implementing a particular TS.
At that MT4 tester is inferior in some sense to MT5 tester in the access to certain data (that is reflected in testing of multicurrency EAs and EAs using several TFs).
Some of the limitations of the MT4 tester become less noticeable on the demo or real (as the EA can then address other TFs and symbols).
Eh...
My code is the simplest one, without indicators.
I have a small profit, but I have it on mt5, and I am losing money on mt4.
I think the programmer is messing around and cannot find his error. I have already paid the money.
I understand that they may have 10-30% different results, but mt5 makes a profit, and mt4 just pisses you off... The difference is different...
for example during three years of testing mt5 shows profit +X, and mt4 shows loss -X
The programmers on this forum are mostly quite competent and qualified.
The reason of the loss in MT4 will be very difficult to understand without demo tests and detailed analysis of all trades (preferably on the terminals of one brokerage company so that the differences in trading conditions were minimal).
It is quite possible that TS is competently implemented in both terminals, but for some reason the difference in the results will be quite large (there are many factors and it is hard to tell at once what and how).
For example, while working on the copyer of deals from MT5 to MT4 I faced a limit in the number of orders (or rather in the number of current deals and orders), I think there are a lot of things that may affect the final result.
1. What's the point of having testers in the first place?
For self-reassurance :)
In general, your case is not suitable for a tester. Pipsetters adapted to the tester will lose in real time.
If you take profit / stop loss on one minute bar, then you can easily discard the results of your tester. They have nothing to do with real trading.
2. "History is stored in bars". Then for minutes, that would mean "All ticks" . There are no lower TFs.
Exactly! The ticks are GENERATED. That is, they are created on the basis of data about previous bars. Open an article here on the forum on modelling ticks. It's all there in details.
But the generation method is different in MT4. That's why you see the difference in the results.
---------------
The only trick that is available in MT4 - you can download a tick history of some broker (like Dukascopy etc.) and instead of the one generated by MT itself - give your downloaded kind of real history. This is the ticks you can talk about a slight approximation to the real one.
But even in this case not completely, since there is also a variable SPRED, and RECVOT. They take away much profit from scalpers.
For reassurance :)
In general, your case is not suitable for a tester. Pipsetters adapted to the tester will lose in real time.
If you take a profit / stop loss of one minute bar, then you can safely discard your tester results. They will have nothing in common with the real trade.
Exactly! Ticks are GENERATED. That is, they are created based on data about previous bars. Open an article here on the forum on modelling ticks. It's all there in details.
But the generation method is different in MT4. That's why you see the difference in the results.
---------------
The only thing that is available in MT4 - you can download a tick history of some broker (like Dukascopy etc.) and instead of the one generated by MT itself - give your downloaded kind of real history. This is the ticks you can talk about a slight approximation to the real one.
But even in this case not completely, since there is also a variable SPRED, and RECVOT. They take away much profit from scalpers.
It follows from your reply that pips ideas are tested only on real microaccounts and real money
Logically, the different ways of generating ticks in mt5 and mt4 should play both a negative role on results, BUT an equilibrium positive one. 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.
sergeev,
"Tick generation has nothing to do with reality"...
Go read the article Tick Generation Algorithm in MetaTrader 5 Strategy Tester. If you have a complaint about the tick generator, provide a clear proof.
xds,
Your question about the difference in testing results should be immediately followed by full transaction reports, not an empty flame with no factual material. Also, you still haven't made any post-test comparisons.
sergeev,
"The generation of tics he has in reality has nothing to do with"...
No need to be so belligerent in showing your ignorance. Go read the article Tick Generation Algorithm in MetaTrader 5 Strategy Tester. If you have a complaint about the tick generator, provide a clear proof.
xds,
Your question about the difference in test results should be immediately followed by full transaction reports, not an empty flame with no factual material. Also, you still haven't made any post-test comparisons.
sergeev,
"The generation of tics he has in reality has nothing to do with"...
No need to be so belligerent in showing your ignorance. Go read the article Tick Generation Algorithm in MetaTrader 5 Strategy Tester. If you have a complaint about the tick generator, provide a clear proof.
xds,
Your question about the difference in test results should be immediately followed by full transaction reports, not an empty flame with no factual material. Also, you still haven't done a post-deal comparison.
Here are the files I sent to my programmer
The settings of the EA in MT4 and its analogue in MT5 are identical.
The reason of the discrepancy in the results was not found
I don't understand the "worries" about ticks in the tester either.
A lie repeated many times becomes a reality.
That's why I have to continually reprimand those who make incorrect claims about the tick generator.