Phoenix - Development+Suggestions - MQ4 in Post#1 - page 10
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
To Wackena
You are also on the new Goglin thread. Good...
They use there two indicators that "show" trend : Turbo_jrsx and Turbo_jvel.
It seems that these indicators "shows the way" with Goblin.
Do you think (huuum !) that these indicators could/should help us to "show" the "turning points" (Up/down/Range) in Phoenix settings.
Discuss that with Darak : he his the "boss" in that matter and I am not so good about these technicals things... (always recognize our "limits" !)
And only helping with "ideas"...
See youbertbin,
I started testing Goblin only this week. I'm not convinced yet how well it indicates up and down trends. Seems to flip up & down signals quickly.
Wackena
All the best for you daraknor
Added the contributed PeriodGrace and PeriodForce settings from ???? (trailing stop and immediate close respectively)
This can reveal another one soul from Phoenix, so the circle keep on track.
I'll try later, so busy right now with my EA after GMT time changed in my broker.
This developments are good, no bug report since release many version.
if you still decide to use function CloseAfterHours in your setting maybe you'll add this to your code, example CloseAfterHours 36, and then your phoenix has made OP at Friday 24:00, it's dangerous because early monday the trade will close without any movement opportunity.
nb : function CloseAfterHours Disabled in 5.6.6 by Daraknor, switching to Grace/Force
//=====================SIGNAL6=======================
bool BuySignal6=false, SellSignal6=false;
if (TradeFriday ==true || DayOfWeek()!=5) {
BuySignal6=true;
SellSignal6=true;
}[/CODE]
find this code, and add in the end of condition " && (SellSignal6==true)" for sell signal, " && (BuySignal6==true)" for buy signal. See below.
[CODE]if((SellSignal1==true) && (SellSignal2==true) && (SellSignal3==true) && (SellSignal4==true) && (SellSignal5==true) && (SellSignal6==true)) return(1);
if((BuySignal1==true) && (BuySignal2==true) && (BuySignal3==true) && (BuySignal4==true) && (BuySignal5==true) && (BuySignal6==true)) return(2);All the best for you daraknor
This can reveal another one soul from Phoenix, so the circle keep on track.
This developments are good, no bug report since release many version.Thank you very much blucky! I forgot to edit the changelog to give credit. I had to look up the contributor in the history file, and it was from HerbertH again. I fell asleep instead of doing it the same night.
I'm editing the changelog to reflect another one of his contributions.
Very glad to hear no bugs so far. I reviewed the trading history from Bertbin, and since he was using 5 currencies Phoenix had been aggressively trading. Looking through his trade log, it looks good but it does confirm that a trade filter would be a good addition. Planning that for Phoenix 6.
5.6.6 Changelog
Our Second community release, another candidate for a "Stable" build of Phoenix. The 5.6 development branch will not have any changes to money management, signal trading, or major changes to modes, profit levels, etc. Those changes are waiting for version 6 of Phoenix.
I wanted to fix the syntax for the OSMA signal, but I didn't want to break compatibilty. That change is delayed until 6.x Phoenix. Please be advised the signal is reversed and may give a negative value when a positive is expected. Unless we look for magnitude based signal changes or the internal workings of the indicator change, this should not impact our performance at all.
As always, the code is availabe in Post #1. Code is posted here as well for eventual archiving of versions.
Here is the changelog for version 5.6.6:
Added comments about the changes made, who the maintainer and contributor was. Thanks Diam0nd
Added two Settings spefically for settings file management. They currently don't "do anything" but we will add sanity checking and compliance checks in later versions of Phoenix.
Documented settings more thoroughly in code for the extern area.
Disabled calls to CloseAfterHours
Added the contributed PeriodGrace and PeriodForce settings from HerbertH (trailing stop and immediate close respectively) (sorry about delay in credit Herbert!)
Heavily modified the PeriodGrace/Force code for sanity checking, correctness, syntax. I might have missed some, please review this area extensively. (I made over 50 changes)
Set signal_count default to 5 in the interest of more consistent trades between brokers, users, etc. Backtesting implies a setting of 30-50 may significantly reduce drawdown without a serious impact on trading volume.
Added the EURUSD settings contributed by EricBach. Not all settings were provided, so some settings are just copied values. My edits and the settings need confirmation. Phoenix was fighting a strong EURUSD trend, wouldn't advise using this in real money trading until trend is over.
Clarified internal documentation and what changes are necessary to turn signal count into a time based setting instead of tick based. Tick based is dynamically reactive to volatile markets. Time based drastically reduces the number of new trades.
Mode3_MoveSL_Trade_2_3 check history function is now limited to the last 20 entries in history. Older accounts would have become increasingly slow, especially with trailing stops active. The value cap of 50 for 'z' may be too low, feedback? The break and continue offered a quick escape, but I wanted a harder cap as well.
Mode3_MoveSL_Trade_2_3 now works if the first trade was a buy trade.
Added RefreshRates before any trailing stops moves, etc.
Mode3_MoveSL_Trade_3 has some history function change. Same fix for buy trades.
Mode3_MoveSL_Trade_3 will no longer generate 130 error message. Added MarketInfo StopLevel for universal broker compatibility. Needs verification.
Please forgive this blatatly rude question, but I don't have enough time to read through all of the pheonix threads and I would like to see if I can contribute to this EA, so can someone please link me to or PM me the location of the most recent version? Thanks, and sorry for the intrusion!
Please forgive this blatatly rude question, but I don't have enough time to read through all of the pheonix threads and I would like to see if I can contribute to this EA, so can someone please link me to or PM me the location of the most recent version? Thanks, and sorry for the intrusion!
Hi Nicho
Glad to have you here and hope you will contribute to the success of Phoenix
You will found all last version of Phoenix here :
https://www.mql5.com/en/forum
in post 1
Phoenix Development Thread
there :
https://www.mql5.com/en/forum
Live testing report there :
https://www.mql5.com/en/forum
See you
From the Changelog:
"Added the EURUSD settings contributed by EricBach. Not all settings were provided, so some settings are just copied values. My edits and the settings need confirmation. Phoenix was fighting a strong EURUSD trend, wouldn't advise using this in real money trading until trend is over."
We need to add anti trending code to improve Phoenix more, the fundamentals are getting in the way. As I said, the EURUSD signals aren't recommended for real money trades.
Looking at the loosing streak, do you have any suggestions? Were the signals in the wrong direction or would they have succeeded with a higher SL, etc.I tested Phoenix v5.6.6 again in mode 3 with Risk=0.5 on EURUSD and USDJPY. For last 6 months, there were two serious drawdowns on USDJPY, one caused by a trade opened in the wrong direction, and the other would succeed if third (lower) take profit was equal to the second one.
On EURUSD, there was a trade on Oct.4 that would also succeed with TP3=TP2, on Oct.10, Oct.16, Oct.20, Oct.25 trade was opened in the wrong direction (sell with up trend). Almost all of the first 50 trades in the 6 month period were excellent, though.
Goblin has an adequate response to the trends I examined, if not to notice its hesitation to make trades of noticable amount during those periods. However, I would agree that copying Goblin trend/range status code will not do.
I tested Phoenix v5.6.6 again in mode 3 with Risk=0.5 on EURUSD and USDJPY. For last 6 months, there were two serious drawdowns on USDJPY, one caused by a trade opened in the wrong direction, and the other would succeed if third (lower) take profit was equal to the second one.
On EURUSD, there was a trade on Oct.4 that would also succeed with TP3=TP2, on Oct.10, Oct.16, Oct.20, Oct.25 trade was opened in the wrong direction (sell with up trend). Almost all of the first 50 trades in the 6 month period were excellent, though.
Goblin has an adequate response to the trends I examined, if not to notice its hesitation to make trades of noticable amount during those periods. However, I would agree that copying Goblin trend/range status code will not do.I agree as well, copying other code is not ethical. Right now I need to wrap up two other EA i'm working on before I do a serious Phoenix 6 undertaking. It would be nice if Phoenix scaled up and down the lot size based on its confidence. Currently the magnitude of signals is not measured or considered. In my edit of Post#2 I listed a few possible Trend detection methods, I am always interested in more.
Hi Nicho
Glad to have you here and hope you will contribute to the success of Phoenix
You will found all last version of Phoenix here :
https://www.mql5.com/en/forum
in post 1
Phoenix Development Thread
there :
https://www.mql5.com/en/forum
Live testing report there :
https://www.mql5.com/en/forum
See youThese links aren't working. Is anyone managing a task list?