Strategy tester. - page 3

New comment
 
Dear Renat.
I am trying to use a very handy thing - a regression channel.
From EA it is created and modified, you can get the values of its middle,
But when I run it through the tester =
exp2 EURUSD,M1: kanal_regres1=0
exp2 EURUSD,M1: object name passed to ObjectGetValueByShift function is missing in the objects list
exp2 EURUSD,M1: object name passed to ObjectCreate function is missing in the objects list

Can this channel be implemented as an indicator and if so, is there a code?
 
Then why are you writing?

Fool, that's why I'm writing.
For the third time I am writing about errors in the tester, but you probably do not care ...

Run your expert from your distribution kit and get complete nonsense in the output of your tester.
Precise and detailed questions and descriptions of the situation I have already provided you (2 times).
Or do you need as in that anecdote about cops (N times)?

Frankly speaking, I don't care about your experts and your tester ...
The tester is IMHO useless, and I don't need it ...
I just wanted to help you in your search for glitches.

You don't want to?
Me neither ...
I have already described it all to you (after spending my time) ...
I don't have any selfishness ...
 
Then why are you writing? You know very well what the answer to "crooked and so on" claims is. <br / translate="no"> Provide exact and detailed questions, please.

Quite an interesting situation turns out =)))


I'm not saying and will not say that MT (or tester) is glitchy. But. )))
I'll give you an example from life (mine):
- I download a new build, install it.
- I read everything on the forum and in the guide about the tester
- I start testing the Expert Advisor that has been tested on a demo account for half a year.
- I leave to have a rest.
- After about 36 hours I look at the result - and there's just no result, the testing is not completed (it's interesting, the load of the CPU is constant - terminal.exe is 50% - but that's not what I mean)


I understand that:
- the history file has been generated (it's not increasing in size, the traffic is not increasing, the tester doesn't say anything about it anymore)
- the tester works (with other people, with other experts)

And then finally this "but":
nowhere does it say that the expert to be tested should differ from the one really trading
(with the exception of the mentioning of the impossibility of portfolio trading)


I understand that this is about my Expert Advisor, "but".....
 
Sorry, what does he do with the memory? Sticking it somewhere, maybe? :)
I got 645M of virtual memory and didn't want to free it all when I closed. It stayed a process in memory.
 
Sorry, what does he do with the memory? Maybe shoving it somewhere? :)<br / translate="no"> I had it take 645M of virtual memory and didn't want to free it all when I closed. Stayed as a process in memory.

same story =(
Linux is really bumpy =)))) I feel dreadful toriosity. Can be fixed by rebooting...
 
Gentlemen, so that you do not have doubts, do a few things. <br / translate="no"> 4. when the EA is deinitialized, print the entire history and look at commissions, swaps and profits. calculate the results. compare it to what the tester said. think about it.
And then make a claim.


It's a good tester, no doubt about it. It's just glitchy as hell.
699	2005.02.04 14:55	sell	72	1.00	1.1969	1.2044	0.0000		
700	2005.02.04 14:55	s/l	72	1.00	1.2044	1.2044	0.0000	-621.89	20367.86	
701	2005.02.04 14:55	buy	73	1.00	1.2060	1.1985	0.0000		
702	2005.02.04 14:55	modify	73	1.00	1.2060	1.2030	0.0000		
703	2005.02.04 14:55	s/l	73	1.00	1.2030	1.2030	0.0000	-250.65	20117.21	
704	2005.02.04 14:55	sell	74	1.00	1.1969	1.2044	0.0000		
705	2005.02.04 14:55	s/l	74	1.00	1.2044	1.2044	0.0000	-621.89	19495.32	
706	2005.02.04 14:55	buy	75	1.00	1.2060	1.1985	0.0000		
707	2005.02.04 14:55	modify	75	1.00	1.2060	1.2030	0.0000		
708	2005.02.04 14:55	s/l	75	1.00	1.2030	1.2030	0.0000	-250.65	19244.67	
709	2005.02.04 14:55	sell	76	1.00	1.1969	1.2044	0.0000		
710	2005.02.04 14:55	s/l	76	1.00	1.2044	1.2044	0.0000	-621.89	18622.78	
711	2005.02.04 14:55	buy	77	1.00	1.2060	1.1985	0.0000		
712	2005.02.04 14:55	modify	77	1.00	1.2060	1.2030	0.0000		
713	2005.02.04 14:55	s/l	77	1.00	1.2030	1.2030	0.0000	-250.65	18372.13	
714	2005.02.04 14:55	sell	78	1.00	1.1969	1.2044	0.0000		
715	2005.02.04 14:55	s/l	78	1.00	1.2044	1.2044	0.0000	-621.89	17750.24	
716	2005.02.04 14:55	buy	79	1.00	1.2060	1.1985	0.0000		
717	2005.02.04 14:55	modify	79	1.00	1.2060	1.2030	0.0000		
718	2005.02.04 14:55	s/l	79	1.00	1.2030	1.2030	0.0000	-250.65	17499.59	
719	2005.02.04 14:55	sell	80	1.00	1.1969	1.2044	0.0000		
720	2005.02.04 14:55	s/l	80	1.00	1.2044	1.2044	0.0000	-621.89	16877.70	
721	2005.02.04 14:55	buy	81	1.00	1.2060	1.1985	0.0000		
722	2005.02.04 14:55	modify	81	1.00	1.2060	1.2030	0.0000		
723	2005.02.04 14:55	s/l	81	1.00	1.2030	1.2030	0.0000	-250.65	16627.05	
724	2005.02.04 14:55	sell	82	1.00	1.1969	1.2044	0.0000		
725	2005.02.04 14:56	s/l	82	1.00	1.2044	1.2044	0.0000	-621.89	16005.16	
726	2005.02.04 14:56	buy	83	1.00	1.2060	1.1985	0.0000		
727	2005.02.04 14:56	modify	83	1.00	1.2060	1.2030	0.0000		
728	2005.02.04 14:56	s/l	83	1.00	1.2030	1.2030	0.0000	-250.65	15754.51	
729	2005.02.04 14:56	sell	84	1.00	1.1969	1.2044	0.0000		
730	2005.02.04 14:56	s/l	84	1.00	1.2044	1.2044	0.0000	-621.89	15132.62	
731	2005.02.04 14:56	buy	85	1.00	1.2060	1.1985	0.0000		
732	2005.02.04 14:56	modify	85	1.00	1.2060	1.2030	0.0000		
733	2005.02.04 14:56	s/l	85	1.00	1.2030	1.2030	0.0000	-250.65	14881.97	
734	2005.02.04 14:56	sell	86	1.00	1.1969	1.2044	0.0000		
735	2005.02.04 14:56	s/l	86	1.00	1.2044	1.2044	0.0000	-621.89	14260.08	
736	2005.02.04 14:56	buy	87	1.00	1.2060	1.1985	0.0000		
737	2005.02.04 14:56	modify	87	1.00	1.2060	1.2030	0.0000		
738	2005.02.04 14:56	s/l	87	1.00	1.2030	1.2030	0.0000	-250.65	14009.43	
739	2005.02.04 14:56	sell	88	1.00	1.1969	1.2044	0.0000		
740	2005.02.04 14:56	s/l	88	1.00	1.2044	1.2044	0.0000	-621.89	13387.54	
741	2005.02.04 14:56	buy	89	1.00	1.2060	1.1985	0.0000		
742	2005.02.04 14:56	modify	89	1.00	1.2060	1.2030	0.0000		
743	2005.02.04 14:56	s/l	89	1.00	1.2030	1.2030	0.0000	-250.65	13136.89	
744	2005.02.04 14:56	sell	90	1.00	1.1969	1.2044	0.0000		
745	2005.02.04 14:56	s/l	90	1.00	1.2044	1.2044	0.0000	-621.89	12515.00	
746	2005.02.04 14:56	buy	91	1.00	1.2060	1.1985	0.0000		
747	2005.02.04 14:56	modify	91	1.00	1.2060	1.2030	0.0000		
748	2005.02.04 14:56	s/l	91	1.00	1.2030	1.2030	0.0000	-250.65	12264.35	
749	2005.02.04 14:56	sell	92	1.00	1.1969	1.2044	0.0000		
750	2005.02.04 14:56	s/l	92	1.00	1.2044	1.2044	0.0000	-621.89	11642.46	
751	2005.02.04 14:56	buy	93	1.00	1.2060	1.1985	0.0000		
752	2005.02.04 14:56	modify	93	1.00	1.2060	1.2030	0.0000		
753	2005.02.04 14:56	s/l	93	1.00	1.2030	1.2030	0.0000	-250.65	11391.81	
754	2005.02.04 14:56	sell	94	1.00	1.1969	1.2044	0.0000		
755	2005.02.04 14:57	s/l	94	1.00	1.2044	1.2044	0.0000	-621.89	10769.92	
756	2005.02.04 14:57	buy	95	1.00	1.2060	1.1985	0.0000		
757	2005.02.04 14:57	modify	95	1.00	1.2060	1.2030	0.0000		
758	2005.02.04 14:57	s/l	95	1.00	1.2030	1.2030	0.0000	-250.65	10519.27	
759	2005.02.04 14:57	sell	96	1.00	1.1969	1.2044	0.0000		
760	2005.02.04 14:57	s/l	96	1.00	1.2044	1.2044	0.0000	-621.89	9897.38	
761	2005.02.04 14:57	buy	97	1.00	1.2060	1.1985	0.0000		
762	2005.02.04 14:57	modify	97	1.00	1.2060	1.2030	0.0000		
763	2005.02.04 14:57	s/l	97	1.00	1.2030	1.2030	0.0000	-250.65	9646.73	
764	2005.02.04 14:57	sell	98	1.00	1.1969	1.2044	0.0000		
765	2005.02.04 14:57	s/l	98	1.00	1.2044	1.2044	0.0000	-622.56	9024.17

.

 
Would you be so kind as to point out the error, please?
Just don't forget that postic testing mode without appropriate timeframes is unreliable. Or you do not want to think about it? It is easier to enable quality "to the max", not to fill our head with thoughts of how everything works and where it all comes from, and just go out on the forum and shout. It's much easier, isn't it?
 
Second post in this thread - Mak 02.07.05 08:50
 
The tester took a look. Acceptable.

I would like to suggest a few additions to the interface. I may have looked at it badly and missed something.

1. After testing, I would like to either automatically open a window with a chart, or work in the current window.
2. I would like to be able to assign default chart templates, because when you assign a template, all the marks on the chart disappear. And green on black is not familiar to me personally.
3. in my opinion it is more convenient to read the chart on time axis, rather than the number of deals, or enter in the tooltip the time of a deal.
4. There is no possibility to clear the log.
5. The report does not contain the time frame, which is not correct in my opinion.
6. It is not clear how to use different time periods ("Use date" does not work correctly)
 
Would you be so kind as to point out the error, please? <br / translate="no"> Just don't forget that postic testing mode without appropriate timeframes is unreliable. Or don't you want to think about it? It is easier to enable the quality "to the max", not to fill our head with thoughts of how everything works and where it all comes from, and immediately get on the forum and shout. It's much easier, isn't it?

Yes please.
I will repeat it again. And I'll even explain, if it's not obvious.
All trades have tp=5 and profits from -51.25 to 50.00

He's got a funny way of counting. It cannot be so. EURUSD currency.
.................................... 225 2005.04.15 16:29 buy 113 1.00 1.2920 1.2847 1.2925 226 2005.04.15 16:30 t/p 113 1.00 1.2925 1.2847 1.2925 31.25 12360.00 .................................... 259 2005.05.06 14:45 sell 130 1.00 1.2862 1.2935 1.2857 260 2005.05.06 15:01 t/p 130 1.00 1.2857 1.2935 1.2857 50.00 12406.25 261 2005.06.02 08:57 buy 131 1.00 1.2248 1.2175 1.2253 262 2005.06.02 09:14 t/p 131 1.00 1.2253 1.2175 1.2253 -51.25 12355.00 263 2005.06.02 09:27 buy 132 1.00 1.2263 1.2190 1.2268 264 2005.06.02 09:52 t/p 132 1.00 1.2268 1.2190 1.2268 50.00 12405.00 .................................... 291 2005.06.29 17:08 buy 146 1.00 1.2087 1.2014 1.2092 292 2005.06.29 17:43 t/p 146 1.00 1.2092 1.2014 1.2092 8.75 12273.25



And as for -

Just don't forget that postic test mode without appropriate timeframes is unreliable. Or don't you want to think about it?


This is exactly what I'm talking about - unreliable modelling. And it's clear that it will be unreliable in any case, but what he reproduces is out of the question.
He doesn't do that on all simulated minutes. But only on selected ones, and you say it's OK, it should be like that.
1234567891011
New comment