Phoenix 2007 (new thread) - page 81

 
psnw:
I just finished reading the thread yesterday, I have nothing to do today. May I help? If so, how far back would you like to go?

If you want to? Great! Only the last 4 weeks will do. Thanks in advance!!!!

 
AZBOfin:
good info, thanks.

however, IMO it was pure coincidence that both IBFX and FXDD returned the same results. another EA i am testing returned profits for the IBFX account whereas FXDD returned a loss (same pair, same settings).

AZBOfin

Were they both "live accounts", both "demo" or mixed? It might have been an invisible hand at work

 

W*a*r*n*i*n*g*

Don't use Mode 3 until further notice. Mode 3 is not working as it should. My fellow Dutchman HerbertH (Thanks Herbert!! You're Gold man!) has spent his whole weekend to fix Mode 3. The fixed version will be posted tomorrow. Did some tests and Mode 3 is now working nice and easy (great job Herbert!).

 

options we have for improving the performance of phoenix

I think we have two options:

1- Considerable amount of work and test has been done, on Phoenix with Hendrick and other ppl here, but something left unchanged from the first version without any improvement is the engine of phoenix, i mean the function that generate buy and sell signal, but first we need to understand it by bone.

all right, we all can guess about the reasons of why they have been choosed by Hendrick, but it's worth to hear from Hendrick why they are choosed, and discussed about them, may be we need another signal (useSignal6) for some currency pairs, or may be ...

then alot of good ideas will be generated by smart ppl in this forum, so we need your manual about phoenix engine Hendrick!

2- forward test show us, that manual exit is better than mechanical exit, I've heard it somewhere that EA's are good for entering the market but not good for exit, so if professional manual traders can tell us how they exit from a position, we can make something good from it, or may be we need another extra engine for phoenix for generating exit signals.

thanks all of you and sorry for mistakes that i probably made bcz of my bad English.

 

daraknor:

great analysis above! i think it also has to do with as when someone starts an EA. if there are already open trades and a new signal is present, the EA won't open another trade. this is probably the main reason for different results.

daraknor:
Were they both "live accounts", both "demo" or mixed? It might have been an invisible hand at work

they are both demo accounts.

my live account (which is a mini account) did "only" -19 pips compared to the -68 of the demo (which are standard accounts).

"invisible hand"? definately yes! it's big brother's hand.

AZBOfin

 
 
 
fikko:
Hi Hendrick,

Yes, I need to learn how to do excellent setting for phoenix. I am interested to know where the figures are coming from. the "The Ultimate Guide To Backward Test Phoenix" will do. After reading it, I believe we can do our own setting for many pairs or even modify the setting when the market trend changed. Is that Right?

if you do not support Phoenix again, we are just like sheep without shepherds. It is very blue out there. However, I'll try to keep giving my contribution, though my contribution gives a lot of troubles at your side.

Phoenix is a good tool for doing trading. Just like a weapon, it must be used in a right way. Running 24 hours is the standard procedure. To increase level of success, I think, it needs human assistance. I am trying to find out when Phoenix needs assistance so Phoenix makes no more mistakes

Hi Fikko!

No, I'm still here and will support Phoenix (forget my previous mail). Phoenix is a good EA and we will work all together to make it even better in the future!

 
daraknor:
I have created virtual environments for my MetaQuotes installation. I will use a professional 24/7 connected data center with redundant power companies, onsite generators, redundant fiber connections. I also have tools to detect if MetaQuotes crashes and restart it. I plan to offer this as an EA hosting service. Probably nowhere near as good as Metaquotes offers for the contest you are winning, but very reliable and still something that can be controlled. I'm also going to look into the multiple terminals concept more, so I could manually close trades at home while the datacenter runs the EA. The datacenter will likely have half as many machines to cross, which would lower the requote rate I expect. That said, I would probably have SL and TP levels far in excess of what I actually think is reasonable in case of war, flood, huge news events, etc. Basically these would be "no poverty" style settings, and not my regular trading signals.

Hi there,

Are you creating one by yourself or there is a company to offer that kind of EA hosting service?

Sounds interesting but it may not be affordable by everyone.

Regards

Wilson