Backtesting performances with build 600
Hi,
2 days ago I got my installation automatically upgraded to 600 (broker: Alpari) and after fixing many compile errors and warnings reported by the new compiler, I could launch back-testing again.
I see that new tester performances are now really worst than those of last version: this is a well known issue or limitation, I think...
I could test 1 year in about 1 hour before, and now in a night time the test cannot complete 1 week of history (incredible)!
Any suggestion?
Hello RaptorUK,
Thanks for your fast response. I could find some specific discussion mentioning builds 6xx tester slowness just here: https://www.mql5.com/en/forum/149452/page11, but it seems (if there are no other threads on this topic) that there is not a build solving this problem, is that correct?
It would be important to know I think:
1) if this issue of tester slowness is confirmed, and if MQ put it under analysis
2) If some workaround exists to the problem, to know which
3) If the root cause was identified (linked to 2)
Thank you again.
BR
F.
Hello RaptorUK,
Thanks for your fast response. I could find some specific discussion mentioning builds 6xx tester slowness just here: https://www.mql5.com/en/forum/149452/page11, but it seems (if there are no other threads on this topic) that there is not a build solving this problem, is that correct?
It would be important to know I think:
1) if this issue of tester slowness is confirmed, and if MQ put it under analysis
2) If some workaround exists to the problem, to know which
3) If the root cause was identified (linked to 2)
I don't think we know MQs reaction to this or if they are even aware of it . . . I have seen Stringo talking about performance on the Russian forum . . . if you want better performance tell the Service Desk and wait . . .
Good. Let's see with Service Desk, than.
Thank you very much.
LOL, is that all you want to know ? it would be progress to know 1/10th of that
I don't think we know MQs reaction to this or if they are even aware of it . . . I have seen Stringo talking about performance on the Russian forum . . . if you want better performance tell the Service Desk and wait . . .
Do you want Metaquotes to use its human resources to solve actual problems or to communicate about problems ?
Why do I have to choose ? do you think MetaQuotes should not communicate effectively what changes have been made and what has been fixed ? it would be more efficient rather than getting numerous Service Desk queries all asking the same thing . . . why do we have to trawl through posts trying to determine if something has been fixed or not ? I guess I'm expecting too much . . .
I understand perfectly your point which seems obvious. But what I mean is we are seeing things from outside, and we have a limited view on what happens. It's not so obvious to have a public view on the bugs for example. We know nothing about how Metaquotes work inside.
To summarize, my point is only to have a more equilibrate view, there are a lot of good things done by Metaquotes, and some which are not from our point of view.
I think in this particular instance (I mean the big changes from 509 to 6xx) and the serious and numerous problems caused by it, there should be an effort to better communicate bug/resolution logs to both brokers and/or end users. If that meant taking someone's time away from resolutions and onto communications, it would be worthwhile IMO. It would certainly help you mods field all the repetitious questions on here and we'd get a warmer feeling regarding the intention on MQ's part to resolve these things (or not as the case may be).
Maybe MQ don't want to divulge just how big the list is/has been.
I understand perfectly your point which seems obvious. But what I mean is we are seeing things from outside, and we have a limited view on what happens. It's not so obvious to have a public view on the bugs for example. We know nothing about how Metaquotes work inside.
To summarize, my point is only to have a more equilibrate view, there are a lot of good things done by Metaquotes, and some which are not from our point of view.
I don't want to be angry at MetaQuotes, it helps nobody . . . but I have an EA to code and I have no idea when I will have a platform that I can be confident in, and finally when I code it and come to test it I think it is going to run like a dog in Visual mode. Yipeee . . .
I could test 1 year in about 1 hour before, and now in a night time the test cannot complete 1 week of history (incredible)!
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
Hi,
2 days ago I got my installation automatically upgraded to 600 (broker: Alpari) and after fixing many compile errors and warnings reported by the new compiler, I could launch back-testing again.
I see that new tester performances are now really worst than those of last version: this is a well known issue or limitation, I think...
I could test 1 year in about 1 hour before, and now in a night time the test cannot complete 1 week of history (incredible)!
Any suggestion?
Thank you
F.