Why the developers failed to make a proper tester - page 9
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
Nindza is much more reliable ... and in G forex, if you follow the rules for its work (following the algorithm of the tester)
Most of the commercial testers (the ones you can buy) are also collecting history ...
( decomposing ticks in matcad is the perfect tester... except it's a hassle)
I am amazed at your ability to work on so many platforms!
This is how many years of programming one needs to do, I've only just mastered MT - look, it's already more than half a year has passed, maybe next year I'll start working on Ninoza - another six months at least
would believe...
if your experts were working adequately in these systems.
But you yourself said about incomplete conversion.
What are we discussing then?
That your new pips algorithm in the Ninza demo is better than the MT4 demo in Al Spore?
And all testor and all on demo.
Sad Vyacheslav...
This isn't research anymore.
This is masochism.
Andrew, do you seriously think that tick history will save you? Then your system is probably based on the analysis of price movement within minute bars - and probably built on trades within a minute bar.
This is not an area of activity that DCs approve of. Because it's a naked pipsing, with all that it implies.
But all fans of tick history, understand the most important thing: Metaquotes, very probably, has made its product according to the strict requirements of its clients, i.e. brokerage companies. This ToR obviously did not include providing the client with the exact tick history of this broker - simply because this broker is not profitable due to the specifics of the business. This is my imho.
One more thing: before you say "How can I rely on my robot", you should understand the extent to which you should trust a tester on even "the most accurate tick history" (that term makes me smile, frankly). Do you realise that the history is virtually variable (simply because tick behaviour on small TFs is a random process, and your so-called accurate tick history is only one implementation of it) - and hence the test results should be just as variable too?
The DTs have already learned how to effectively fight the pips players by making the spread and stopgap floating. In fact this is the actual killing of the piper. But the stubborn DC piper still thinks that this is not the point, but that, you see, he is not given an accurate tick history. It is more pleasant for him to fool himself. Well, good luck with the self-deception!
Why would I need MTS on H1-D1??? I can spare 10 minutes a week to trade myself! Robots have all the power with speed! M1-M15 is the maximum for robots, and by depriving them of ticks you just neuter the very idea!
PS
And I want to know how the fucking spread will affect my system!
The editor crashes the first time I try to copy something to buffer by pressing Ctrl-Ins or Ctrl-C, it happens randomly.
If I agree to just-in-time debugging in VS2010, I get a message from the studio
"Av unhandled exception occurred in MetaEditor.exe [5976]".
Then the debugger gives "Unhandled exception 0x7734dc9b in MetaEditor.exe: 0xC0000374: Heap has been damaged".
The funny thing is that if you debug further, you fall back into the editor and it often starts working without errors! That is, I always start the studio first and then the meta-editor :)
Basically, I need to take a .map file and see what's at the excerpt address.
Please advise how best to make a bug report.
I'll tell you. It will kill it - along with the requotes and the busy trade flow that will suddenly appear on the real. If the system is such that its result is highly dependent on the spread, the tester results are better not to be trusted - even if you were testing it on a "perfect tick history". This is the main thing that for some reason pip testers don't want to understand.
Any testing must take into account the conditions of play allowed in the DC. Pipsing does not meet those conditions. Full stop.
I'll tell you. It will kill it - along with the requotes and the busy trade flow that will suddenly appear on the real. If the system is such that its result is highly dependent on the spread, the tester results are better not to be trusted - even if you were testing it on a "perfect tick history". This is the main thing that for some reason pip testers don't want to understand.
Any testing must take into account the conditions of play allowed in the DC. Pipsing does not meet those conditions. Full stop.
Well, if you take various cheap kitchens - I agree. I'm afraid it's almost 2011. If you want to be sure of the correctness of the market you've got to be careful when choosing a broker who may allow you to trade with more or less points. The execution is on the market - i.e. requotes are basically excluded, but slippage is possible and on major events they are usually minimal - 0,-0,2 pips.
That's the thing, I don't deal with crazy kitchens, pipsing is no problem, the problem is in the implementation of the idea
confirm the problem, server 2008 x64, it's true I crash after some time of editing, the interesting thing is if I save after every body move, the problem does not appear...
Hmm.
Win7 x64 - no problems so far. However, one "but" - I launch MT4 in compatibility mode with WinXP (not to climb in a fucking mirror %APP_DATA%\Virtual Store\Program Files\...\experts, but to work in MT4 normal native directory from C:\Program Files\...) and I use F4 from there to open the editor. And I'm in the habit of pressing CTRL+S after every sneeze...