Jump to content

clovelly

Members
  • Posts

    180
  • Joined

  • Last visited

Posts posted by clovelly

  1. Re: woodie robot

     

    This version will speed up the EA. The problem was because the EA repeatedly calls the Woodie_Trend/Triple_CCI indis from the same bar whereas exactly same values will be returned, as these calls only ask for the values from past bars. Now, they are only called once at the bar open. The backtest on the GJ pair on the same period that took 50 hours now takes 5 minutes. The default settings seem to be profitable for GJ & UJ pairs.

  2. Re: woodie robot

     

    Click here to download the who EA, my settings and backtest result

     

     

    In the file Woodies_the_robot_new_5and4digits.mq4 (to cater for 4 and 5-digit brokers), there is a code segment to calculate the point value:

     

    if (Point == 0.00001) point = 0.0001;

    else {

    if (Point == 0.0001) point = 0.001;

    else {

    if (Point == 0.001) point = 0.01;

    else point = Point;

    }

    }

     

    should it be like below ?

     

    if (Point == 0.00001) point = 0.0001;

    else {

    if (Point == 0.001) point = 0.01;

    else point = Point;

    }

     

    (as Point == 0.0001 is for the old 4-digit price, and should not be changed to 0.001)

  3. Re: Megadroid Pro

     

    I am at Alpari UK and trade EURUSD GMT+1 which means trades open 22:00-00:00 Alapri time. I want the trades within the same window for EURCHF and EUGBP so I also have these set at GMT+1 but I have had trades open at 00:33 and 00:59 for these pairs. Whats going on? Are the trading windows different?

     

    Logic 2 & 3 trade between 21-24 GMT

  4. Re: Wolf(e)Waves and ZUP

     

    Merlin, your online statement is an interesting read. A long series of winning trades, very impressive. There are many small trades within a short period (eg 15 buys at 2009.07.17 14:57) to capture 1, 2 to 10 pips (most of them a few pips).

     

    How can you maintain such a focus, or do you use some tools to help you placing/managing trades, even an EA perhaps?

  5. Re: Profit Mountain has been released

     

    The EA has been released. Their website stated that the product can be purchased on a non-refund basis :-ss . If anyone manages to get a copy, please post. Thanks.

    ========

    Hello Fellow Trader:

     

    + There are a lot of traders out there looking for an Expert Advisor (EA) that makes money consistently.

     

    + We have received a lot of praise for our FX Profit Mountain trading system - and it is now available as an Expert Advisor (EA).

     

    + With one currency pair, trading one lot per trade the EA made almost 60% last year alone.

     

    + This system will work on any currency pair, it does the work for you.

     

    To find out more about this system, developed by a Millionaire Trader, go to wxw.fx?profit?mountain.com

  6. Re: MegaDroid 1.1 craaacked

     

    Did you guys have luck with the Multipair version this week so far? Couse since marked opening it didn't open one single trade for me. Last week I had many, many trades on EURJPY, USDJPY, GBPUSD and even EURUSD... Is it the marked, or is there something wrong with my MT4/MD?

    The droid does not trade the last 3 days of the month (29/30/31). It shoud be ok tomorrow...

  7. Re: Forex systems test manual by Haridas June 2009

     

    Robinhood, excellent work you did, thank you. Can you tell us how you performed the backtest? Did you go through the existing charts manually and follow the trade rules? Or did you use some Trade Simulator and check the rules real-time as the chart progressed? The reason I ask is because the SEFC08 indi (the BullBear) repaints (though lightly) and if you trade realtime, the results will be different.

  8. Re: lets make our EA!!! all programers and wise guys are welcome

     

    Yes I know, but your help and ideas would be appreciated, too. Ok finally someone wrote here, its time to reveal it. :) I´ll post my masterpiece in a few minutes. :D

    How long have you traded manually your masterpiece? If you can provide some details on its performance so far, I am certain that someone will take it on if it is worthwhile ;)

  9. Re: THV3 ea

     

    clovelly,

     

    Hi mate,

    you wrote that the "... errors are due to the indicator "THV3 Trix called" trying to draw objects in subwindow 1,..." you are right I have the same issue with another EA that use the same indicator, can you please give me an idea abt how to fix the issue?

     

    Tx in advance!

     

    You have to comment out the ObjectSet and ObjectCreate statement from the indicator. Make an extern bool flag in the indicator such as:

    extern bool DrawIndicator = true;

    then in the code:

    if (DrawIndicator)

    {

    ObjectSet...

    ObjectCreate

    }

    else

    {

    Do not call ObjectSet, ObjectCreate

    }

     

    When you display the indicator on a chart leave DrawIndicator to true. When calling it from the EA with iCustom call it as false. In this way the error will no longer occur. I wish the coders would take note of this in the future and not make these basic mistakes which make running backtests a pain in the butt for all of us on those jinxed EAs.

     

    Good luck

    gelinux,

    fxknight already replied to you, but if you want to have the code, here it is. I just commented out these Object-related lines in blocks. You may want to take up fxknight's idea as it is better, allowing you to set options.

     

    http://stashbox.org/540775/THV3%20Trix%20called.mq4

  10. Re: THV3 ea

     

    Thank you for share. ^:)^ ^:)^ ^:)^ ^:)^

    Please try this code I think it's correct of programming logic. X_X X_X X_X

     

    int buf = (Period() == PERIOD_M30);

    int buf1 = (Period() == PERIOD_W1);

     

    gi_1000 = (Period() == PERIOD_M1) + (Period() == PERIOD_M5) * 2 + 3 * (Period() == PERIOD_M15)

    + buf << 2

    + 5 * (Period() == PERIOD_H1) + 6 * (Period() == PERIOD_H4) + 7 * (Period() == PERIOD_D1)

    + buf1 << 3;

     

    I agree that this is correct. As long as the bitwise shift operation is done on an integer we are ok. Good that we can maintain the original logic of the system.

    From my high school maths many decades ago,

     

    bit shift 1 = multiply by 2

    bit shift 2 = multiply by 4

    bit shift 3 = multiply by 8

    ...

    so all is sweet [-O<

  11. Re: THV3 ea

     

    Uf, who will find good settings for this one? :-ss

     

    Hello,

    Very nice complex EA. Thank you so much for posting it and for the decompilation. It seems the developer (Damiani who wrote the code for the Damiani_volt) preferes this as a scalping tool and he shows on is website (http://www.cobraforex.com/) that he uses it on EURJPY. Too bad it cannot be backtested at the moment since it issues lots of ObjectCreate errors. These can be fixed hopefully in the Trix indicator but I really don't know how it will affect the EA. I recommend to demo test it as a scalper on EURJPY and/or GBPJPY M1 timeframe. Also check out the forum for this EA for some really good tips http://www.forexfactory.com/showthread.php?t=168347

    It seems people are trading it on EURUSD M5 to M30 timeframes as well.

     

    Good luck

    These errors are due to the indicator "THV3 Trix called" trying to draw objects in subwindow 1, useful for looking at during manual trading but irrelevant in backtesting as the EA does not need those objects, but only needs the values in the buffers. After fixing those errors, the backtesting is ok now but very very slow. It took 20 minutes just to go through 12 hours of test data. At the same rate, it will take roughly 400 hours to test 1 year of data.

     

    The development team is forward testing the EA and will release it free to the community once it is finished. Maybe you can wait till then.

  12. Re: THV3 ea

     

    Starting, I am afraid you are a bit wrong with this "<<". it's a bitwise shift, and the problem is with type conversion. As long as MQL4 does not support type cast as C++ does, should be

     

    int m30 = (Period() == PERIOD_M30);

    int w1 = (Period() == PERIOD_W1);

    gi_1000 = (Period() == PERIOD_M1) + (Period() == PERIOD_M5) * 2 + 3 * (Period() == PERIOD_M15) + m30 << 2 + 5 * (Period() == PERIOD_H1) + 6 * (Period() == PERIOD_H4) + 7 * (Period() == PERIOD_D1) + (w1) << 3;

     

     

    Not sure why the decompiler "does not know" that, but anyways.

     

    Oh, as long as it uses Ichimoku, it should work nicely on EURUSD,EURGBP and JPY pairs on H1 and M30. Will see.

    A simple code to replace it is

     

    gi_1000 = (Period() == PERIOD_M1) +

    (Period() == PERIOD_M5) * 2 +

    3 * (Period() == PERIOD_M15) +

    // (Period() == PERIOD_M30) << 2 +

    (Period() == PERIOD_M30) * 4 +

    5 * (Period() == PERIOD_H1) +

    6 * (Period() == PERIOD_H4) +

    7 * (Period() == PERIOD_D1) +

    // (Period() == PERIOD_W1) << 3;

    (Period() == PERIOD_W1) * 8;

  13. Re: Forex Derivative EA

     

    ryaz, you da man :) Now we need to see which ones need to be optimized from time to time ... I will work on this

     

    So, there are total of 42 parameters out of 101 which change annually for all 4 pairs. In general, placing them into DLL is not that bad idea for the sake of code simplicity.

     

    I would believe that they should be optimized even more often if the algorithm is so sensitive. I've tried to backtest with 2009 parameters from 2008, and drawdown in 2008 was higher than gains in 2009.

    However, without knowing what each of them means in his genius algorithm I would not give it a single try.

     

    I am also sure he will eventually change the DLL calls so that future versions of it are not compatible with older versions of EA.

    Any reasonable suggestions?

    This is the list of all 102 parameters that the EA pulls out from the DLL for each different year, to be used for making trade decisions in the past. The spreadsheet is produced by printing out these parameters returned by the DLL after each DLL call from the MQ4 codes during backtesting. The first column of the spreadsheet is the Year of test, and other columns are various parameters stored in the DLL.

     

    I do not mind if for each year/month/day (or even hourly), the EA is optimised based on past data so its performance can be improved in the FUTURE. However, there is no ethical reason why the backtest is so heavily curve-fitted (like in this case). This is very much like the repainting indicators. The FD sellers need to explain the reasons for this behaviour rather than some vague responses last week.

     

    hxxp://stashbox.org/539093/FD%20curvefitted%20data.xls

  14. Re: MegaDroid 1.1 craaacked

     

    thankyou

    i found it strange because all day and night the spread has been between 1.3 and 2 = 13 and 20 which is below the safe spread

    i have been monitoring and no further records of high activity so not sure may have been increase spread at start of london rush hour

    London Rush Hour? I thought MD only runs 21-23h GMT, when it is quiet

  15. Re: MegaDroid 1.1 craaacked

     

    Just a question can i use this cracked version1.11 on multiple pairs?

    Also i am seeing in the experts journal

    2009.06.10 20:15:18 MegaDroid USDJPY,H1: Strategy2: Safe spread limit exceeded: spread = 54

    2009.06.10 20:15:18 MegaDroid USDJPY,H1: Strategy1: Using DayDirection filter

    2009.06.10 20:15:18 MegaDroid USDJPY,H1: Strategy1: Safe spread limit exceeded: spread = 54

    2009.06.10 20:19:34 MegaDroid EURUSD,H1: Orders in history: 0 profit made: 0.00

    but my 54 should be 5.4 ?????

    so i am still not sure if this is working with ECN feed & broker Go Markets

     

    The EA is self-adaptable to brokers with 4 or 5 digit prices, so 54 here is equivalent to 5.4 (or rather between 4 and 5 :) ).

    The cracked version can be run on other pairs but you have to backtest/forwardtest and then decide whether you want to do it. The EA is designed for EURUSD, hence the safe spread is usually lower (3 or under). In this case, 54 is judged not safe (EA may require 30 to 40?), so the EA does not trade. Personally, I would not run MD with that spread.

  16. Re: MegaDroid 1.1 craaacked

     

    Hi Guys

    This may have been asked here before but i will ask again i have just set up MD to my live account i am with Go Markets

    5 digit

    it looks like its working but not sure i have the last lines reading correction detected and the 2nd last line keeps changing

    Calculating probability, Assesing Market volume,checking voltality,monitoring spread and determining direction is this normal for this?

    any feedback will be appreciated

     

    Dom

    That's fine as it shows that it is working. The first version did not show much status and traders were not happy, so they added more visibility. Good luck.

     

    Go Markets used to have fixed spread (2 for EU, 3 for EURGBP) during these quiet times, but they recently changed to ECN feed, and sometimes the spreads are widened to 2.5 /3 for EU and 5/7 for EURGBP, untradeable for scalpers. Good luck with MD!

×
×
  • Create New...