How was your first EA developed? - page 3
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
In about 2010 when MQL5 was initially released, MQL4 was like MQL5─for the most part.
At that time, MQL4 remained rather skeletal and didn't support OOP. If I remember correctly, the first deviations from MQL4 in MQL5 were the advents of indicator handles followed by more nuanced order sending code─and MQL5 supported OOP before MQL4 did so. I never used OOP in MQL4 because I switched over to MQL5 to use OOP. As the MQL5 language expands at breakneck speed, MQL4 still remains rather skeletal in comparison to MQL5.
Even though I have a bit of nostalgia for the MQL4 language as an easy start in algo coding, moving forward meant leaving it behind. The faster execution and greater control of MQl5 certainly influenced my thinking as well. IMHO, there is no reason to use antiquated MQL4/MT4 unless of course, you only have access to MT4 in your given region of the world.
I see. When I first started trading, many people didn’t like MT5 very much—they were quite attached to MT4. In my case, I also preferred MT4 at the beginning, but once I started coding in MQL5, I realized it’s actually better.
What is OOP? Object Orientd Programming? (Here ir México is POO, Progrmación Orientada a Objetos)
Can't Imagine MQL5 without it.
What is OOP? Object Orientd Programming? (Here ir México is POO, Progrmación Orientada a Objetos)
Yes─unless a coder is bad at object-oriented programming. For a bad coder, its OOPsie Daisy in the U.S... or literally POO in Mexico.
It took me a lot of time to really translate what I usually do when manual trading into an algo.
Manual trading includes a lot of quick mental math and judgements.
However, when translating all that thoughts and strategies into EA, it does not actually produce a result that I expected.
Therefore, the hardest parts are:
1) having a fundamental logic solid enough when building your EA
2) embedding strict risk management
3) translating manual trading strategies into your EA while recognizing the emotionless execution of an algo which is really different from manual trading
Manual trading includes a lot of quick mental math,,,
Unless you're doing math in superhuman milliseconds, an algorithm is faster.
However, when translating all that thoughts and strategies into EA, it does not actually produce a result that I expected.
You could always try strapping on a bunch of AI driven body monitoring devices, and piping that data into an EA. Eventually, AI might deduce your current mood.
It took me a lot of time to really translate what I usually do when manual trading into an algo.
Manual trading includes a lot of quick mental math and judgements.
However, when translating all that thoughts and strategies into EA, it does not actually produce a result that I expected.
Therefore, the hardest parts are:
1) having a fundamental logic solid enough when building your EA
2) embedding strict risk management
3) translating manual trading strategies into your EA while recognizing the emotionless execution of an algo which is really different from manual trading
I also found very hard the first encounter with translating human logic to mql5, firstly backtesting a pattern in TV then trying to replicate in mql5.
Let me guess... Because TV doesn't support automated trading?
Yeap!
Also the strategy tester is key to optimizing.
Somedays I still check TV, but nothing like mt5.
Somedays I still check TV, but nothing like mt5.
I stopped checking in on TV years ago. Once I saw that coding a "hook" to a paid 3rd party automation website was the only way to automate trading, I declared it a kid's toy.
I even gave TS a go at one point. Nope─it printed duplicate Renko bars. So much for integrated features. It's a sad day for TS when free custom utilities here blow TS's integrated features out of the water.
I started my journey with MT4 back in 2005. At that time, I knew a little programming, but there was still so much I didn’t understand.
I don’t quite remember what my very first EA was, but one of the earliest ones I do recall came together pretty quickly. It was a simple test of an idea — trading based on indicator signals with a very tight stop loss and high risk. Classic "search for the Grail" stuff.
I spent countless hours in the strategy tester back then. Days and nights just tweaking parameters, watching how things behaved, chasing that perfect curve.
Years passed. My interests shifted, and now I work on completely different things — professional trade copying solutions, modern GUI, architecture that actually makes sense for real trading.
But looking back, those early days were genuinely exciting. A time of experiments, mistakes, and endless curiosity.