EA's - 3 accounts, identical setups, all different results . . .
Mar 08, 2010 at 02:54
1,253 Views
4 Replies
forex_trader_7697
Member Since Feb 24, 2010
4 posts
Mar 08, 2010 at 02:54
I've been testing some EA's, and have had 3 of the same EA with identical setups running on 3 accounts: 2 demo, and 1 live. The only difference has been that the live account is running on a VPS.
Results have been puzzling. On one account, there have been lots of trades the past week, on the other two, varying amounts of trades. And, the accounts are not all trading the same pair (choices of EUR/USD, GBP/USD, EUR/GBP, EUR/CHF).
I contacted the developer of the EA's, and the response was that the scalping EA in question is very sensitive to timing; that is, ticks may not arrive at the same times on different MT4 accounts -- it depends on the speed of the broker's server.
For this week, I've changed things around. I put the live account on my office system to see if the VPS might be making a difference, and swapped the demos around as well. We'll see.
Does anyone have any experience with such an issue? Is running on a VPS slower than through my computer at the office?
Thanks.
Doc
Results have been puzzling. On one account, there have been lots of trades the past week, on the other two, varying amounts of trades. And, the accounts are not all trading the same pair (choices of EUR/USD, GBP/USD, EUR/GBP, EUR/CHF).
I contacted the developer of the EA's, and the response was that the scalping EA in question is very sensitive to timing; that is, ticks may not arrive at the same times on different MT4 accounts -- it depends on the speed of the broker's server.
For this week, I've changed things around. I put the live account on my office system to see if the VPS might be making a difference, and swapped the demos around as well. We'll see.
Does anyone have any experience with such an issue? Is running on a VPS slower than through my computer at the office?
Thanks.
Doc
forex_trader_8139
Member Since Mar 04, 2010
423 posts
Mar 11, 2010 at 14:24
hi,
I have few account(demo, live) running on same EA and all setting are same. However the result will be very much different some time. But most of the time be small different. Never be same.
They have been running for about 1year.
SIM
I have few account(demo, live) running on same EA and all setting are same. However the result will be very much different some time. But most of the time be small different. Never be same.
They have been running for about 1year.
SIM
Mar 29, 2010 at 18:13
(edited Mar 29, 2010 at 18:14)
We have run a number of accounts on different VPS' with same settings on the same number of charts, just a few minutes of starting of each other and we found that just a few minutes of entering the market can make a huge difference over time in the return achieved.
One example of the high returns we are achieving with MIDAS1, a return we have not been able to duplicate exactly as yet, even starting a back test to the same time we went live with this one!
One example of the high returns we are achieving with MIDAS1, a return we have not been able to duplicate exactly as yet, even starting a back test to the same time we went live with this one!
When everything else out there is just not good enough, design your own! MIDAS
Member Since Oct 28, 2009
1424 posts
Mar 29, 2010 at 18:59
It depends to some extent on the EA.
How it enters trades, how many trades it enters and what time period all contribute. In general though I find that the more complex you make your EA the more inconsistent the results will be across different installations.
Of course this is for my own EA's for an EA developed by someone else there could be all sorts of issues in the code, like use of repainting indicators, which are very prone to latency differences.
How it enters trades, how many trades it enters and what time period all contribute. In general though I find that the more complex you make your EA the more inconsistent the results will be across different installations.
Of course this is for my own EA's for an EA developed by someone else there could be all sorts of issues in the code, like use of repainting indicators, which are very prone to latency differences.
11:15, restate my assumptions: 1. Mathematics is the language of nature. 2. Everything around us can be represented and understood through numbers. 3. If you graph these numbers, patterns emerge. Therefore: There are patterns everywhere in nature.
*Commercial use and spam will not be tolerated, and may result in account termination.
Tip: Posting an image/youtube url will automatically embed it in your post!
Tip: Type the @ sign to auto complete a username participating in this discussion.
