I can only agree with that. That's why I bought the Tickstory subscription on the weekend. So I am at least sure about the quality of the data now. It was scary how my strategies produced on the bad data, with the good data now look ...

Yes you have to export to mt4 with Tickstory. Than you open mt4 with Tickstory. Than open the chart and you will see the chart of exported Data. Now you put the eastudio Script into chart..this files you find in mql folder and you upload to eastudio.

I do it as follow.

1.First you need a clean MT4 Backtest Installation. Make it as portable

So i add in my shortcut this.

M:\MT4Backtest\terminal.exe /portable /skipupdate

Than you are save that the MT4 Folder will always be same.

2. You need also from your Broker an MT4 Installation.
For example call it MT4Live and use same extensions in the terminal shortcut /portable

3. In Tickstory you set in settings the file where tickstory will put his downloaded files.

4. The Next Settings like in pic for the launcher. Here the Folder of your Backtest MT4 and add here also /portable in command as in pic. Now you have done all settings to get exported data to your MT4 Backtest Terminal.

5. Now you need to get your Broker Forex data like digits etc. which you get a script here. Put it into the MT4LIive experts Folder.
And attach this to a chart. Now you will get a file called xxx.mt4config.

6. In Tickstory choose your symbol right click export to mt4.

7. On Metatrader Info you now load the mt4config file to get the symbol specifications.
8. Choose your Start End date and check your timezone and now its downloading. Before close your MT4Vacktest Terminal
9. Now when is ready exported. Click on the symbol

Now your MT4Backtest Terminal is opening with only the Dukascopy charts in it.

10. To get this data for EA Studio download this script from eastudio
Once compiled you put it on your e.g exported EURUSD Chart and the script download the files, which you find in M:\MT4Backtest\MQL4\Files Folder.

Now you can upload it.

I hope it is clear.

The Problem is when you use tickstory with a connected demo account than you broker will overwrite the dukascopy backtest data everytime you run the terminal. So thats why you have to use a seperate mt4 install only for backtesting without beeing connected to a broker.

You can use Tickstory to export the data to your MT4. And with Popov's tool then download the data for EA Studio down there.

When you a registered User, then you can the Limit set to 500k

We are working on a "Premium data feed" service for EA Studio and later for FSB pro. It will provide quolity historical data for more than 20 pairs.

We hope the new data feed will be ready for testing in July.

It will work properly but it will behave differently, performance will be altered and that will be defined by the extent of difference between your data sets. I think basic quantification of data differences is doable, whether it will help to go forward and find a solution is another story.

Basically if you plan to employ (regular) optimization and you have successfully accumulated the amount of data you plan to run optimization on, then it is just a matter of running your strats on the broker's data in the optimizer and then go live with them.

Let me explain further if I don't make sense: for example, you develop the strats on 50k duka data and you plan to re-optimize your strats every weekend on the latest 20k bars. So, assuming you have got the latest 20k bars from your broker, you put your newly developed strats on that 20k newest bars from your broker and optimize them. Then your strats should be adapted in the best possible way to your broker's data.

If you don't use this kind of re-optimization workflow and/or you don't have enough new data from your broker, then there's not much else you can do unfortunately. If backtest data characteristics are similar to your broker's data, then it should be ok. If your strats are really good, then slight offset in the nature of data should not have a pronounced effect.

Come to think of it, maybe you can acquire like 3 sets of data from different sources and pick only those strats, which perform well over all 3 data sets. That will theoretically give you strats, which should be more immune to slight data discrepancies.

I also tried a lot of variations.
The most effective for me after a long time now is the following:
I have acquired tickstory data to develop the strategies. Then I take the time and test, as Footon says, the last xx data bars with my broker data. The strategies left over then, using me, with relatively good success. Of course, this does not protect you from market changes. I ALSO USE THE WEEKEND IF I ALWAYS OPTIMIZE OR HAVE SLEYTUS TO CALIBRATE AND TO EXCHANGE.
I wish you success

P.S .: My special project, which I show here in the forum again and again, has nothing to do with it.