Hi Raptor. I finally got round to re-exporting the data into a csv. I did a count on the number of ticks for the entire of the 22nd Apr 2013 (7:00 to 20:00 FXCM server time), and it comes to 4690 ticks total, which is nothing like what you're seeing.
It is the FTSE that your tick count graph shows isn't it (as that is the subject of my comments)? If it is, I can only imagine they have one feed for spread bet accounts and one for CFDs and we're each looking at the other.
Edit - just looking at your chart again and you have data for the early hours of the morning. FXCM only offer FTSE from the UK open. We must be looking at different instruments, lol.
Ha, makes more sense now, no problem. Yes, it's a good visualisation.
Is that you on a track day in the pic btw? What are you driving/riding?
How long does it take to run such a backtest (tick by tick calculation)? Although it is convenient to run a full length backtest, it may be quicker to break up the files into 2 year increments. Then the last 2 years generating to the current date is much quicker, as the fxt file generation would take a while.
BTW, I use TDS and I bypass the file size limitation problem. But I found the larger files more difficult to deal with in terms of backtest speed. It was easier to run 2 year backtests than a single -8 year backtest and then splice them together. Part of that is also due to the strategy involved such that tick data is required to setup the signals and entries/exits intraday. And I get variable spreads. I still use tickstory to generate the csv files.
[rant] Most if not all of this nonsense would be much easier if mt4 had its
own tick data history importer to work with csv files directly. With all the changes they have made, they neglect the importance of accurate tick data to perform backtests. mt5 blatantly doesn't allow any data importing at all. [/rant] Birt's Tick Data Suite is the answer if you want to backtest using mt4 on a regular basis.
It's not a bug it's a limitation . . . if you aren't happy tell the service Desk.