Download MetaTrader 5
To add comments, please log in or register
Write your articles using our editor. It is convenient!
nanobot
46
nanobot 2010.09.27 06:13 

Please add your vote or comment to this question...

I would say YES. After all the championship starts in about 5 days. I got my robot to compile 1 hr after registration. I know we had plenty of time, 4 months, but some of us decided late and had to get acquainted with the new language an completely new API... I for instance have an MT4 robot ready for long time but unfortunately it was too complex for translation so i decided to create a completely new robot to my demise.


Please comment on this... let's hope Metaquotes gives this a serious thought.

Pix
45
Pix 2010.09.27 06:33  

My Vote: NO

I still have ideas on how to optimize my robot and I wish I could put those in, however, part of real world projects and competitions are deadlines. If this contest is the Forex trading counterpart of the Olympics, then it should be treated in the same manner. If you do not make it past the first round of Olympic pre-qualifications which is on a set date and time, then you do not get to compete in the Olympics. Nanobot said it, we had 4 months! Nonetheless, I bet you will have your bot ready for the next championship and I will be eager to watch it compete!

- Patrick 

rforex
15
rforex 2010.09.27 08:36  

If the testing was done correctly without errors before close of registration, then it is fair. But when the programmers are given clean pass while the submission period is open and then reject them when it is closed, it is grossly unfair. This is called TRANSPARENCY.

LEE Chun Wing
747
LEE Chun Wing 2010.09.27 09:19  
rforex:

If the testing was done correctly without errors before close of registration, then it is fair. But when the programmers are given clean pass while the submission period is open and then reject them when it is closed, it is grossly unfair. This is called TRANSPARENCY.

absolutey, rforex is right

ATC.tester should responsible for their statement before

Dumbledore
114
Dumbledore 2010.09.27 15:51  
It is called a scandal if there are mistakes in Olympic pre-qualifications and no one feels responsible.

Slawa
Moderator
6798
Slawa 2010.09.27 17:51  
Dumbledore:
It is called a scandal if there are mistakes in Olympic pre-qualifications and no one feels responsible.

You have programming error in your EA. You don't check lots amount before send order. This error became obvious after history data adjustment. But this is your error not our.

We can check EAs on any history data to detect programming (not algoritmic or trading) errors.

You had about 4 months to write your EA correctly.

rforex
15
rforex 2010.09.27 17:59  
stringo:

You have programming error in your EA. You don't check lots amount before send order. This error became obvious after history data adjustment. But this is your error not our.

We can check EAs on any history data to detect programming (not algoritmic or trading) errors.

You had about 4 months to write your EA correctly.

How about mine? I strongly believe there is no programming error, only some time taken to download past data that will not affect realtime trading...
Slawa
Moderator
6798
Slawa 2010.09.27 18:14  
rforex:
How about mine? I strongly believe there is no programming error, only some time taken to download past data that will not affect realtime trading...
Please wait. There are too many EAs to manual recheck. We check they one after one
Dumbledore
114
Dumbledore 2010.09.27 20:50  

Sorry I know you are very busy, but can you look at my issue a little more carefully and try to understand it? THE AUTO TESTER TESTED THE WRONG EA AND MARKED THE PROBLEMATIC EA "APPROVED". (sorry for the cap).

I had plenty of time to test all the versions, I just wanted to submit to you the one and best and CORRECTLY TESTED EA. If I knew that EA didn't pass the test I would have submit the one that passed and even don't bother to worry about whether it will pass or not.

Thanks god I still had proof for that. Otherwise who would believe this??

My EAs relies on long term MA, and 1H volatility, you think that will be influenced by adjustment of historical data?? Can you go back to get the version that the tester really examined and see how "bad" it is under historical data adjustment?? THANKS!

Yes I had about 4month, and had a good use of it. I developed 5 different EAs, passed 11 tests, one of the tests a mistake, not bother to catch up the deadline, and refused admission. Thanks for the mentioning.

stringo:

You have programming error in your EA. You don't check lots amount before send order. This error became obvious after history data adjustment. But this is your error not our.

We can check EAs on any history data to detect programming (not algoritmic or trading) errors.

You had about 4 months to write your EA correctly.


stringo:
Please wait. There are too many EAs to manual recheck. We check they one after one

rforex
15
rforex 2010.09.28 04:29  
stringo:

I'm sorry. Your EA is extremely slow - 60% of testing take about 30 minutes. We cannot approve your EA

1) Why testing speed is a criteria for qualification? It is not trading speed but backtesting speed which is not going to affect real time trading.

2) Why is this not mentioned in the rules?

3) If testing speed is a criteria for qualification, MUST have mentioned this in the rules and I would have PLENTY of time to correct it! Why only informing this criteria AFTER the registration close and when the initial EA has been approved? The CPU & memory use on real trading is minimal.

 I disagree with this reason given as it is like pulling the carpet from down under, most importantly, the backtesting speed is IRRELEVANT to the real trading! This is UNFAIR

As the EA contained no errors, it should be allowed to participate in the ATC. My EA is built for trading, not testing and that's why I did not bother about the slow testing process. Only if you guys have mentioned it, I'd have immediately corrected that.

/
To add comments, please log in or register