Interesting topic for many: what's new in MetaTrader 4 and MQL4 - big changes on the way - page 36

 
sanyooooook:

That's the way it works )

ZS: lied a bit, some strategies run in the tester for a general idea of the system, but only if the system is interesting

I don't look at the status quoted by the tester (I'm also lying a bit), I look at the chart from afar, if the blue icons are below the red ones the system will work, if not - it won't.

When I look at the system, I get a feeling that the MQ system is not that good, but it has a real effect.

SZS and if the tester is not used, then Cloud is not involved, twice ;(

So what would a developer vs a trader need? Maybe you want R or something else the lady wants :)

ZZZY the question is not rhetorical.

 
JJerboa:

is it a bug or is robofx messing with you?

No, it's just bad luck. You got one bar with tick volume == 1 :) When this bar moves out of the left edge, everything will be fine.

It's because of a small inaccuracy in my tool. Change one line here:

   if(NewBar())
     {
      DeltaLoAskBuffer[1] = ((low[1]+spread[1]*_point) - LoAsk) / _point;
      DeltaLoAskColors[1] = (int)(DeltaLoAskBuffer[1]<0);
      DeltaLoAskBuffer[0] = 0;
      DeltaLoAskColors[0] = 0;
   // LoAsk=DBL_MAX;     // это убрать
      LoAsk=tick.ask;  // это вставить
      OldTime=CurTime();
     }
   else
     {

Or download the new (corrected) one right here:

Files:
 
MetaDriver:

It only seems provocative to you, in fact there is no professional alternative. Bars are a surrogate.

Open/Close prices have nothing to do with the logic of the quote, they are randomly pulled out of the row. // And one of them (any of them) is unnecessary.

As for intra-bar extremums - I have already stated it.

Only ticks remain as a reliable source for tests. They are not perfect (there are an awful lot of them, for example), but there is nothing to replace them without falling into tester idiocy.

--

There are many ways of thinning out tics that can be suggested. Ivan once suggested his way, I smoked it, and it was a reasonable one at first glance. There's more to come. If anybody likes it, let them test/optimize with reneko/cagi; if I want equitemporal bars, flags to hand; if I want equi volumetric ones, let them generate and test them; if I want, let the optimizer use raw ticks.

HOW to realize it is a second quite difficult question (taking into account the necessity of multicurrency synchronization!), but it can be solved if desired.

You're right on the money.

I've been watching the ticks yesterday and I want to say I strongly dislike the way the bar slice breaks the micro-trends into different spots (neighboring bars), which makes it completely impossible to understand if there was a micro-trend or not.

That said, a bunch of small shifts are being pushed into one bar.

ZZY again in MT5 we have a link to quotes of a particular broker, by the same observations yesterday for Alpari Robo and MQ, the ticks are different for all.

 
Urain:

On the whole the same thing, that's a shame for developers MQ, all the same tried, but it's like that :(

SZS and since the tester is not used, Cloud is not involved, twice ;(

So what would a simple developer vs a trader need for it to work? Maybe you want R or something else the lady wants :)

ZZZI The question is by no means rhetorical.

I dare say a thought. )

We need to think of an alternative to real ticks for the tester which we already have. How would it be possible to implement such a thing, to bring the tester ticks and all other data in general, as close as possible to the real one? I think it could even be done better than just loading and using the custom history, which is always what it is. We need to solve the problem of lack of data in general, which in the absence or insufficient amount of data you have to keep supplementing/accumulating or looking for from where to download.

To do this, you should not use the algorithm of ticks generation (fixed), which is given by developers, but build a configurable generator of data. In which you can configure the size and time ranges of the duration of trends, flags, spreads, volatility and generally any data, where you can simulate any possible reality. The history problem would then be solved definitively.

 

They are handsome people:

  • One declares that he has never written a line on mql5, that the tester is useless and he does not use MT5 at all. He has his own tester!
  • A couple of others are telling tales about the tester that it is none at all, too. Like we do not use it.
Who do you want to fool?

But it is clear with hrenfx, in fact, he acts as the face of FXOpen by mutual consent. And he doesn't need and never will need any MT5. Tiki is his favourite song, which he tries to present as "I root for all traders and want it to be staff". He doesn't want staffed and won't use it, as the subject of performance will disappear. Instead there will be a lot of complaints that the story is not enough, that the glass is not given and he loses something.

Not to mention the fact that hrenfx knows very well that he tells tales of free retail arbitrage and scalping. And there's no need to wave a storyteller with the history turned off. After all, he hasn't even put a single fairy tale account in signals here just to show off, but only hides it in hidden pams.

 
Renat:

They are handsome people:

  • One declares that he has never written a line on mql5, that the tester is useless and he does not use MT5 at all. He has his own tester!
  • A couple of others are telling tales about the tester that it is none at all, too. Like we do not use it.
Who do you want to fool?

But it is clear with hrenfx, in fact, he acts as the face of FXOpen by mutual consent. And he doesn't need and never will need any MT5. Tiki is his favourite song, which he tries to present as "I root for all traders and want it to be staff". He doesn't want staffed and won't use it, as the subject of performance will disappear. Instead there will be a lot of complaints that the story is not enough, that the glass is not given and he loses something.

Not to mention the fact that hrenfx knows very well that he tells tales of free retail arbitrage and scalping. And there's no need to wave a storyteller with the history turned off. After all, he hasn't even put a single fairy tale account in the signals here just to show off, but only hides it in hidden pams.

If this paragraph is about me, including - I declare responsibly, I haven't run a tester on MT4 or MT5 for exactly half a year. Although I do active research.

And about retail arbitrage it's really not that simple. But this subject should not be approached head-on. Information on how to bypass the restrictions is available in the public domain.

 
Renat:

They are handsome people:

  • One declares that he has never written a line on mql5, that the tester is useless and he does not use MT5 at all. He has his own tester!
  • A couple of others are telling tales about the tester that it is none at all, too. Like we do not use it.
Who do you want to fool?

But it is clear with hrenfx, in fact, he acts as the face of FXOpen by mutual consent. And he doesn't need and never will need any MT5. Tiki is his favourite song, which he tries to present as "I root for all traders and want it to be staff". He doesn't want staffed and won't use it, as the subject of performance will disappear. Instead there will be a lot of complaints that the story is not enough, that the glass is not given and he loses something.

Not to mention the fact that hrenfx knows very well that he tells tales of free retail arbitrage and scalping. And there's no need to wave a storyteller with the history turned off. After all, he hasn't even put a single fairy tale account in signals here just to show off, but only hides it in hidden pams.

I'll add to the beauties.

Although I've written "a couple of lines" on mql5, but due to limitations of quotes history and lack of Ask history MT5 tester is not used and I don't plan to use it yet.

I am completely agree with hrenfx, that for proper testing (especially for scalping systems) we need HighBid and LowAsk. If we have LowAsk then the majority of algotraders will not needtick history. And I do not understand how after such a long explanation you and other algotraders do not understand.

I cannot say anything about free arbitrage and free scalping. But scalping is probably 90% of all strategies that work in Forex.

 
MetaDriver:

Only ticks remain as a reliable source for tests. They're not perfect (there are an awful lot of them, for example), but there's nothing to replace them without falling into tester idiocy.

Well if that's the case, you need a ticking asc-bid story. And if you're going to get really bogged down, you also need liquidity at the edges.
 

What are the history limitations when there is a detailed minute history of spreads over a dozen years in MT5?

What kind of fairy tales are you telling? Are you deceiving yourself as well or are you just playing for the public in support of the general trend?

 
TheXpert:
If that's the case, you need a ticking asc-bid story. And if you're going to get really bogged down, you need liquidity on the edges as well.
What do you want liquidity on the edges of? The edges of the tick?