Same Account, different MT5, different amount of real tick

 

Can someone help me with that? I'm testing the same EA, on the same Account of the same Broker, on two different MT5 (same hardware), I get different amount of real ticks. 


I thought that the real tick depended on the broker and the server but it seems not to be so. 


See attached image.

Thanks

 
Luca De Andrea: Can someone help me with that? I'm testing the same EA, on the same Account of the same Broker, on two different MT5 (same hardware), I get different amount of real ticks. I thought that the real tick depended on the broker and the server but it seems not to be so. 

Yes, it should be the same, but the older setup may have tick data stored from an earlier time.

For example, say that a broker only supplies a month's worth of tick data. When you first set it up, it supplies you with 1 month of data, but as you leave that machine running over several months it starts accumulating several months worth of tick data stored on the disk drive. So it will have more available data than the a newer setup.

So, to see if this is the case, see if the setup that reports more tick data is the older one.

You can also check the main "bases" folder (not the tester one), and navigate down to the broker's tick data files and see their respective file creation dates.

 
I also see that the one reporting more ticks has a lower "quality", probably due to there being gaps in the tick data. Probably because the setup is not running 24/7 so the earlier gathered tick data is not contiguous.
 
Luca De Andrea:

Can someone help me with that? I'm testing the same EA, on the same Account of the same Broker, on two different MT5 (same hardware), I get different amount of real ticks. 


I thought that the real tick depended on the broker and the server but it seems not to be so. 


See attached image.

Thanks

Are these 2 instances using the same MT5 build ?

You should check the tester logs. You can posted them here if you like. It should give you a clue in which parts of the ticks there is a problem.

 
Fernando Carreiro #:

Yes, it should be the same, but the older setup may have tick data stored from an earlier time.

For example, say that a broker only supplies a month's worth of tick data. When you first set it up, it supplies you with 1 month of data, but as you leave that machine running over several months it starts accumulating several months worth of tick data stored on the disk drive. So it will have more available data than the a newer setup.

So, to see if this is the case, see if the setup that reports more tick data is the older one.

You can also check the main "bases" folder (not the tester one), and navigate down to the broker's tick data files and see their respective file creation dates.

Hi Fernando, thanks.

I checked that, there is just a couple weeks difference for the creation date 

 
Alain Verleyen #:

Are these 2 instances using the same MT5 build ?

You should check the tester logs. You can posted them here if you like. It should give you a clue in which parts of the ticks there is a problem.

Hi Alain, thanks,


The MT5 with 100% real tick is build 3802, the "bad one" is build 3815. See attached image

 
Luca De Andrea #: Hi Fernando, thanks. I checked that, there is just a couple weeks difference for the creation date 

Those are not the tick data folders. Those are the M1 data folders. However, there is a difference—one has data for 2012 and the other not.

Tick data folders ... \bases\[broker]\ticks\[symbol]\

By the way, what is your testing date range?

 
Luca De Andrea #: The MT5 with 100% real tick is build 3802, the "bad one" is build 3815. See attached image

Then that is another issue also. Build 3815 is a “beta” which might have bugs. The latest officially released build is 3802.

 
Fernando Carreiro #:

Those are not the tick data folders. Those are the M1 data folders. However, there is a difference—one has data for 2012 and the other not.

Tick data folders ... \bases\[broker]\ticks\[symbol]\

By the way, what is your testing date range?

Sorry Fernando you're right, 

Yes I see a big difference between those two folder. The good one has a lot more data.


Date range is the same: 01 Jan 2020 to today. 


From what can it be due?

 
Luca De Andrea #:

Hi Alain, thanks,


The MT5 with 100% real tick is build 3802, the "bad one" is build 3815. See attached image

Build 3815 is a beta, you should not use it unless you want to be a beta-tester.

What about the logs ?

 
Luca De Andrea #: Sorry Fernando you're right, Yes I see a big difference between those two folder.

You keep showing the "Date modified" of the files, but you should also be looking at "Date created" to see which is older.

Luca De Andrea #: From what can it be due?

Possible answers already given ...

  • The "bad" one is a beta build which could have bugs
  • The good one seems to have been running longer and accumulated more data (but you have not shown the creating date, so can't be certain).