Phoenix 2007 (new thread) - page 117
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
Just so everyone knows, default is now mode 1, we're calling it stable. Mode 3 should still be tested, it has much higher profit potential long term. Testing is a big help! I test on my machine too, but since my trades differ sometimes I don't have a chance to see bugs before others do. I have been including little messages and tags in error messages lately, so posting error messages helps along with your great explanation of what happened. As soon as I had a decent explanation of the problem, we had a fix. Fixes are posted in the Dev thread often before they are released as a new version. (on upper left click Forex Trading > Trading systems > Phoenix for a list of all related threads)
That was a bug
and it is fixed in 5.6.8 which has just been released. https://www.mql5.com/en/forum/175560
I'm setting up the site with downloadable history files this week, on the PhoenixFund website. PhoenixFund is a nonprofit that will donate profits to charities, and is approved to be an IB. We can standardize our backtesting on the same broker that gives us trades.
Thank you for your excellent writeup, enjoy the new version!Awesome! Glad to help in some way! Dude.....You are FAST!
And THANKS!!
Will be looking for the Phoenix Fund Site, will Egold donations be accepted?
Hi to all!
i do not know if somebody else has already written about this error (I try to read everything but time is always so little...).
Today, 12/12/06 at 19:45 GMT P5607a mode 3 opened 3 trades on eurjpy both on my live and demo account at IBFX. All the trades were opened at 155.16 and SL of 155.96 but immediately after the open my live account changed trade 3 moving SL to 155.64 and after 36 minutes trade 2 was changed moving SL to BE. After that also trade 3 was moved to SL at BE (155.16).
First trade remained at 155.96 as opened. I checked demo account but there SL of trade 2 and trade 3 were left at original 155.96
After seeing it I moved manually SL of my real trade 2 and 3 to SL 155.96 but I really did not understand what's happened.
RafThat was a bug
please update to 5.6.8 - you will find it on the first page of this thread. Mode 3 is considered "in testing" and mode 1 is considered stable.
I think we need new settings, Phoenix is having problems in this market - especially on IBFX. FXDD it makes money for several people rather well still.
Awesome! Glad to help in some way! Dude.....You are FAST!
Welcome and thank you
Generally as soon as I have enough information I can fix the issues. This is much easier than debugging 7 layer inheritance with template C++ objects. That... just... horrible. C# is ok to troubleshoot with objects, the real tests are when you can't look inside or perform the tests yourself. Backtesting gives me some means of testing code but not enough 
I will remember to include EGold as a means of donating. The IB method will take a few more days to set up, but I will try to get the basic ways to donate set up soon.
That was a bug
Well, I did it and SL for that trade has not changed but.... another trade USDCHF opened on 12/12 at 19:40 GMT with 5.6.7a (but managed by 5.6.8 overnight) closed trade 2 changing SL to BE this morning at 7:15 GMT. Trade 1 and 3 are correctly managed.
???
Raf
Best regards from Russia, Hendrick!
I tested your phoenix on demo in mode2 and risk 0.3 on GBP/USD. I have a 212% clean profit on period 13.11.2006 - 12.12.2006.
very good result!
I want to start the real money test on the next week. When did you plan post a new version adjustment for phoenix?
PS: Sorry, my english is very bad =))
Input Settings in Phoenix
Hi testers,
Where can I find a description of all the input settings in Phoenix? I have been backtesting with by-the-tick data and getting good results. But are they reliable? It might help if I knew the meaning of the knobs I am twisting.
Boola
Hi testers,
Where can I find a description of all the input settings in Phoenix? I have been backtesting with by-the-tick data and getting good results. But are they reliable? It might help if I knew the meaning of the knobs I am twisting.
BoolaYou can get a list of all threads at www.forex-tsd.com/phoenix/ and I would recommend the first 3-5 posts in FAQ and almost all of the optimization thread (which is short and on topic). A few indicator settings were noted as being more important/reactive than others.
Not included in the older documentation is "signal_count" which looks for consecutive signals before making a trade. If signal_count =1 (default version < 5.6.6) then it will trade on choppy, quickly reversing signals. If signal_count is 5+ (default in version 5.6.6+) then the signals are smoothed out and only consistent signals are traded.
When the number is higher, we will enter trades *later* than normal. Since Phoenix is a trend reversal system, this time delay will reduce drawdown and can turn loosing trades on good signals into winning trades on good signals. I would recommend experimenting with "very high" numbers of 50-200 but I haven't done a thorough analysis.
I hope that gets you started!
Best regards from Russia, Hendrick!
I tested your phoenix on demo in mode2 and risk 0.3 on GBP/USD. I have a 212% clean profit on period 13.11.2006 - 12.12.2006.
very good result!
I want to start the real money test on the next week. When did you plan post a new version adjustment for phoenix?
PS: Sorry, my english is very bad =))Dmitry,
The community is creating and maintaining the settings as well as the code. Version 5.6.08 has been released as a bug fix. Mode 1 is stable, Mode 3 is testing. New settings will come very soon, but it doesn't make sense to optimize bugs so we have been a little slow about settings.
If you wait a week or two it would be good, since new settings need to be made and we should test the new version a little more.
We all wish Hendrick well! He drops by every once in a while but now he has time with his family again.
Mode 3 troubleshooting
If you are using Mode3, you should definitely be using 5.6.08. 5.6.03 will trade without errors, but Mode3 is not working the way it should.
I wrote post on the dev thread about what Mode 3 is supposed to do.
"Mode 3 on the other hand is designed to work with one trade at 0.5 TP, one trade at full TP, and another trade at 1.5 the TP. If the 0.5TP succeeds, then the other trades are protected. Their SL is set equal to the entry point, and they won't loose us any money. That we can take 24 pip * 1/3 lot size as profit. If we take the full 48 pip TP, then the last trade has its SL moved up to a 24 pip profit. Essentially if the first trade wins, we get something. If we were trading 3 lots overall, and we had a 48 pip TP, we would make
24 pips if trade 1 TP, 2 & 3 SL
96 pips if trade 1 & 2 TP, 3 SL
144 pips if trade 1 2 & 3 TP
The pip winnings are calculated based on a single lot size, not 3 lots. "
Explained another way, Trade1 hits the TP, Trade2 and Trade3 should have breakeven stoploss.
If the SL for Trade2 or Trade3 move before Trade1 closes, this is a bug and please post it here with the trades before this time and the current price if possible.
If Trade1 closes at a profit and the SL for Trade2 and Trade3 don't move, this is a bug and please post it here with the trades before this time and the current price if possible.
If Trade 2 closes aat a profit, then Trade3 is supposed to move the SL to a profitable level. If this does not happen, please post it here because it is a bug.
If you are using Mode3, you should definitely be using 5.6.08. 5.6.03 will trade without errors, but Mode3 is not working the way it should.
Darak,
I have done quite a bit of optimizing with Mode3 on the GainCapital tick data. Switched up to .8 as soon as it came out, but it reacts differently from .7a, so I'm not sure if the .7a results stand. Would it be of any help for you to see the journals from the tests? Perhaps you could point out settings that I am not using optimally, and you might find cases where the SL are not behaving as they should.
Should I post the reports to this thread or another one?
We all owe you and Hendrick a tremendous vote of thanks. After only a couple of weeks at this I am beginning to appreciate the magnitude of what you have accomplished.
Boola