i edited my initial post to exclude EAs mentioned.
ok you say it is inpossible to run at the same time. I trust you, but tell me I would like to fully understand why is that. Is it because of the margin? Do you think there will be more then 20-40 open orders at the same time and used margin will get too high? Or there is some technical charateristic of meta trader that won't allow that many EA'a running?
Off the bat, two problems I see with having tons of EAs running are:
1) Trade server is busy. Not sure if you've ever run into this error, but this happens when you're trying to open/close/modify too many orders at one time. Not a huge deal when you only have one EA, but when I was forward testing several EAs on my demo account, I ran into this problem often. Also, if you're dealing with grid/scalping EAs you're going to be sending a lot of orders down the pipe. Now, the end result of this error message may be small, say, you missing a few pips by entering in slightly after you signal told you to enter. But even that can wreak havoc on your EA if it goes after small quick pips. A bigger problem is when the EA doesn't handle this error right, and thinks it did put its order through, and updates internal values, etc... This will probably only affect more complex EA's that keep a detailed state of everything.
2. Magic Numbers misuse Anytime I decide to forward test a new EA, I almost always have to re-write portions of it to make sure it plays nice with my other EAs. If you don't handle Magic Numbers right then you'll have EA's stomping all over each other, closing/modifying orders that don't actually belong to them. The more refined an EA is, the less you'll run into this, but I still see a lot of active EAs that don't handle this properly.
My 2 cents.
good information here. So lets talk about possible soutions. Running multiple terminals? Having account funds spread out to a few brokers or having multiple accounts with the same broker? (not sure if that is technically possible).
I beliave newdigital when he says that it is not possible to have this much EAs on 1 account however I would like to know what is a concrete problem that can not be solved. Magic number I am sure can be solved. "Server is bussy" is beyond my knowladge (like everything else isn't). Margin call and overal float are things to consider and are potential account killers.
I am not coder as well.
So you don't need to believe in me.
Just open many orders (perding orders normal orders) in any demo account and then try to delete all the orders using some script or EA. Probable you have such scripts or EAs to delete all open orders?
You will see that the orders will be deleted one by one:
- one order will be deleted;
- then second order will be deleted;
and so on.
The orders will not be deleted simultaniously. MetaTrader can not do it in the same time. Because of Metatrader by itself.
MetaTrader can not modify two orders in the same time.
EAs is the robots semding some commands to Metatrader/ Which commands? Commands to open the order or close the order, or modify and so on. If two EAs sent commends in the same time so uit may be a problem: Metatrader will not know what to do first.
Try to open many orders on demo account the then to delete them using script and you will see. But real account is not demo account. It may be more slowly on real account.
Many people are trying to use many EAs in one Metatrader. And they had to create special library files to do that.
- "trade context is busy" error fixing file: library file
- Pause between trading operation in Metatrader: this file and this file.
- all the events monitoring in Metarader: this file and this one.
- "all orders controll" library file: here.
- "3 variation of money management for every EAs in account" file: here.
- 29 variations of money management for all EAs in account using one file: here.
And so on.
Article from Metaquotes.
Error 146 ("Trade context busy") and How to Deal with It:
Just go to MetaQuotes website to get all informatuion about it.
So, to create Portfolio manager is very very difficult task. What is Portfolio manager?
It is some tool to manage all EAs inside one account:
1. to automatically detect the condition of the market sending/saving all the information in txt file;
2. "to predict/forecast" of the price movement with support/resistance and so on;
3. to switch EAs on and of depends on the market conditions;
4. to controll and monitor all events in Metatrader including margins;
5. to controll all EA in one Metatrader;
6. use one Money Management for all EAs working in Metatrader.
7. re-act on the news;
8. to generate the signals (email for example).
9. "trade with me" function: "copy" trades on the other Metatrader.
One year ago one person sent PM to me on this forum. He said that he is part of some commercial project to create Portfolio manager. This project is fully financed and he is having very good money as a salary. And he said that this tool will never be sold because company wants to use it.
I don't think that somebody will create Portfolio Manager here in this public forum for free.
Some parts of Portfolio manager tools were created by Russians and this tool is using in some russian forums for their elite sections to test portfolio EAs. Almost all big forex forums are having elite sections to have some freedom from sponsors (brokers, banks and commercial people). And they are testing EAs using some kind of Portfolio manager tool. They will never sell it. If they sell their tools they will sell their freedom.
And because of that this big project will be started in elite section. Many coders wanted to participate (Igorad, Codersguru, Kalenzo and so on). It is already started but not coded yet. BTW we have 9 pages in portfolio manager thread in elite section already.
It's all that I know about this subject.
Thanks for so many valuable informations ND!
Ok, I posted something earlier from my Palm Treo, but I thought I did a double posting, hence I erased my first message.
No offense vpopovic, but it sounds like you are not ready to go live yet, based on the questions you initially asked. My advice to you is to wait, seriously. Get a second computer, demo trade every good EA on this forum for 2 to 3 months, and then go live (use real money.) DON'T DO IT RIGHT NOW!
I'm sorry , but if you are scared of tinkering with magic numbers, then you are not ready to go live with EAs. That's like wanting to be in an auto race, but not knowing EXACTLY what the wind shield wipers do.
great info newdigital!!! I also hate repeating myself but again it is imppossible to read every single post in this forum,.
i could agree more with you Yoda_Glenn, and that is exaclly why I posted looking for somebody to customize these things for me in exchange for a small fee and a percentage from profits. Beeing not particularly inteligent and ignorant costs money and I do now that. But I need to disagree on your professional driver theory. I am sure that if the shield wiper question was on the test Ricky Boby would't give a correct answer.
I HAVE A AN IDEA! Why not set up a fund, hire a master C, mql whatever coder. This EA could be available only to contributers, or later FOREX-TSD might want to charge a fee to individual investors that want to use it. Honestly I think this is the greatest idea ever!
It will be some difficulties:
1. When somebody is lossing his own money so it is bad but understandable. But if this somebody is lossing the people's money so it is very different story.
2. Past EAs' performance is not guarantee of future performance.
3. Some EAs were created especially to be used with particular broker. Generally to say, almost all EAs were created for particular (but different) brokers and every broker is having different data.
I think we will create Market Condition indicator first. If I posted some market analysis during the many months on our forum using Ichimoku chart for many timeframes with some prediction (if price ... buy ... but if ... sell, breakout levels ...) so EA/indicator can do it as well. And there are some other systems with this possibility (6 ma system - forgot the thread sorry; Nina system and some others).
It is not very difficult by programming.
Once we create Market Condition EA/indicator so next step: we may identify EAs which work good in particular market condition. We have great statistical testing data in elite section because many EAs were tested live during the more then 1 year. So we may identify them as trending, counter trend and so on.
And just create some EA Switcher which will switch EAs between each other depends on market condition. With some probability of course. And we may reduce the risk in this case.
If we add some possibility to have "trade with me" function so it will be exactly what you want:
you will need just attach some tool to the chart, select the options "trade as Newdigital's StepMAExpert_v145" for example and your metatrader will open and close the orders exactly in the same time as StepMAExpert145 metatrader from elite section. If I am testing some EAs and those EAs are profitable so all that you will need to do is to open/close the order exactly together with those EAs. If we will have "market condtion" function (as true and false in the settings) so it will be exactly what everybody needs: StepMAExpert opened buy at 12:30 for example in elite section Metatrader and in many other Metatraders for many different people in the same time and the order will be closed for all different Metatraders simultaniously.
But it is not difficult to do. Difficult to start this development.
And it will be some problems with market condition and with "trade with me function" because my EA may open buy at 1.9650 for example at 12:30 GMT and this tool will copy this order at 12:30 GMT to your Metatrader but your price to open the order may be 1.9655 because of different broker's data. So it may be a problem: my Alpari rus demo Metatrader will have +10 pips in profit but your real FXDD will have -10 loss. May be, not every EAs should be included in this 'trade with me" possibility ...
Many people are waiting for this development I know that.
i am seting up on VSP (darknor is hosting?) FXDD demo, GridMACD on EURUSD and EUDCHF plus StepMAExpert 1.5 on GBPUSD... Test about 2 weeks then compare results with newdigital alpari results. If they are not drastically different I am going live baby!
As far "ea trander" goes I wish best of luck. Again, I think there are professional coders out there that can transform any idea into reality and I would be more then willing to pay for something that I am going to make money with!
VSP is different. There some VSP that your Metatraders are working only if you login and after logout - stopped. We have two good thread about VS in this forum where people discussed this and other subjects concerning VPS. Or check with darknor.
You should have big account to use GridMACD. Because this EA is openning many orders and not trading often. Probable you will get the same results with GridMACD with any broker. Two weeks is nothing for this kind of EA.
As to Step EAs so it is different situation. I am testing this EA with Alpari rus and with Fibo Group having different results. If you are planing to test any Step EA with different broker (FXDD or IBFX) so you probable should find your own settings. Besides two weeks and two months are nothing for this EA as well: those kind of EAs are not trading often.
Just remember what I said few posts about: "start with broker". Not a county "to put money". With broker's data comparing and so on. Because most significant in this case (in case of EA's trading) is the data. Usually people are selecting the broker according to the country and so on. But in case of automated trading you should select the broker according to the data first. Because most of EAs were coded to use with particular broker and it is necessary to find the other settings sometimes to use them with other broker.
I have one example with one EA (i will not name it sorry because it is public well-known EA created long time ago outside of our forum). This EA were coded using Alpari data and profitable for some pair during the testing live. But not profitable with IBFX with any settings. And I know exact opposite situation: one well-known EA were coded using IBFX data (and seems profitable with IBFX) and not profitable if testing with Alpari rus broker the same or similar settings.
People discovered this subject long time ago before this forum was created. But you may check and find some discussion in our forum concerning DayTrading3, Envelope 2.11, Brainwashing EAs and Goldwarrior EA.
Better to find all the posts the user Sadaloma with his testing results of Golrwarrior EA and Brainwashing EAs with different brokers because he discussed with me about this subject one year ago (i was not agree with him in that time but later on i recognized that he was right).
Just make sure that the person, you are willing to pay to, knows all those subject.
When somebody is trying to sell EA to me (nobody really, but just for example) so first question which I will ask when I see the performance: "which broker did you use?"