Miracles with the tester. - page 4
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
All in all, things are very bad with the tester/optimiser... I'll have to exclude the first bar from work.
One more bug: It happens ANYWHERE (but quite often, though not always) on the first timer call on a weekday:
1. tester:
2. optimizer on the same parameter (time after time):
and at the same time, on other passes, both in tester and optimizer on the first trigger bid/ask = 0.
I don't know how it should be, on the tester and optimizer should work the same way.
I will attach another Expert Advisor for tests in Service Desk on request #329165.
+Sometimes the tester does not reach the end of the test - the "Results" tab does not appear, although the logs show that the test passed to the end
Upd.
I don't see any bug fixes in the new build announcement (and silence in the Service Desk), but they are essential.
Another floating glitch with the tester (occurring occasionally and inconsistently):
I can't understand why Invalid Stops, if stops are not limited, NormalizeDouble is used, numerical values of stops are correct?Another floating glitch with the tester (occurring occasionally and intermittently):
I can't understand why Invalid Stops, if stops are not limited, NormalizeDouble is used, numerical values of stops are correct?Output market Bid/Ask prices when an error is received.
Surely you will see that the stops are actually close to the market.
Output the market Bid/Ask prices when an error is received.
You'll probably see that the stops are actually close to the market.
I intentionally put the maximum number of StopLevel and FreezeLevel in the logs:
MAX(stopLevel, freezeLevel):0Which is generally correct for stockeys. Besides, I have made three more attempts to open an order increasing stops and the last one was sl=0.06933, tp=0.59208 (take was higher than the open price almost by a month range of the pair, but there is nothing to stop at)
I will finish testing and print the current bid/ask - it is unlikely anything will change.
UPD. i see on the chart a huge spike of 2011.11.16, apparently you, Renat, were right.
UPD2. Taki Renat, you are right - there was a spike and the price changed dramatically.
+ I have a strong subjective impression that optimization is slowed down in 619th build (I used to use 4 cores + 4 local "remote" agents; now there are only local agents left, maybe that's why it seems like optimization is slower)+++ optimization takes all resources of the computer and work with other applications becomes uncomfortable (e.g., watch youtube or work in Visual Studio) - in the previous build, even with 4 more local "remote" agents, everything was fine and one could do things quite comfortably
It would be much easier if the developers provided a debugger in the tester. Who can say why there is no debugger in the tester?
notused:
Optimisation and single pass results do not match (service-desk - #329165 + EA there as well)
stringo:
We will figure it out.I want to ask - have you figured it out? I do not understand how to test the Expert Advisor. And the most annoying thing is that the optimization is running in the cloud, the money is being withdrawn but one cannot use the result.
I would like to ask - have you figured it out? Because it is unclear how to test the Expert Advisor. And the most annoying thing is that the optimization is done in the cloud, the money is withdrawn, but you cannot use the result.
...
What a discrepancy. Is it only when optimising in the cloud or locally too? I've never noticed that (I've never been optimised in the cloud).