Questions from a "dummy" - page 80

[Andrey Dik]

x100intraday:
So if you can't be bothered to read all the history back in time, then I don't see the problem. Find out the opening and closing times of each bar and look at the number of seconds in those intra-bar ranges. If less than expected, write a "fake" bar. This will be the tipping point, after which all other bars will be incomplete. There is no point in searching further.

You can meet such "fake" bars in the middle of "normal" history (in which case most of the history will be rejected for practical purposes) or even entire series of such bars. I say there is no reliable way.

[Mykola Demko]

joo:
Such "fake" bars can also be found in the middle of a "normal" story (in which case most of the practical story will be discarded), and even whole series of such bars. I am telling you that there is no reliable method.

But several consecutive bars only at the end of the story, so there is a reliable way.

[x572intraday]

joo:
Such "fake" bars can also occur in the middle of "normal" history (in which case most of the useful history will be rejected), and even entire series of such bars. I'm telling you, there's no reliable way.

Then the statistics should be done by looking out where the amount of "fake" bars is not occasional (read: accidental), but regular (constant), and most likely the pattern will not be simple: when several consecutive bars consistently have more than the Open - Close range, we can assume that this is not an accident and feel free to stick a ski stick in it.

There is no reliable - perfectly working - method, Forex is not reliable. There is only a common sense criterion to proceed: how many consecutive bars in a series must not correspond to the condition of the number of seconds to be considered not as an accidental "failures" in the history, but as the beginning of the preceding era - the era of less detailed history making.

[Andrey Dik]

x100intraday:

There is no reliable - ideally working - way, Forex is not reliable at all. We can only use a common-sense criterion: how many consecutive bars in a series must fail to meet the condition of the number of seconds to be considered not as a random "failures" in the history, but as the beginning of the preceding era - the era of less detailed history keeping.

Have you worked on MT4? Have you encountered a similar problem? - There is no such a problem there, which means there is a reliable way, but it is in the developers' hands. I have already suggested one solution.

Another solution, expressed by tol64, is not to display incorrect data at all.

All other methods, but on the part of the user, will be unreliable.

[[Deleted]]

joo:

Have you worked with MT4? Have you encountered such a problem? - There is no such a problem there, so there is a reliable way, but it is in the developers' hands. I have already suggested one solution.

Another solution, expressed by tol64, is not to display incorrect data at all.

All other tricks on the part of the user will be unreliable.

You can make an additional functionality at the language level that will return the quality of history in percentage for a certain pair and TF, for a certain period.

For example, analyze the minute history for the year/month, if it is not 100% then there will be problems.

[x572intraday]

joo:

Have you been working on MT4? Have you encountered a similar problem? - There is no such a problem there, which means there is a reliable way, but it is in the developers' hands. I have already suggested one solution.

The other solution, expressed by tol64, is not to display incorrect data at all.

All other methods, but on the part of the user, will be unreliable.

What do you mean, unreliable? Of what? Honest and blatant falsification of minute bars in the distant past with the substitution of the bars of older TFs? Or no lack of minute bars in the distant historical past? If we are speaking about the first one, then yes, in MT4 the developers have not implemented such a compromise substitution and they act directly: if there are earlier minute bars, they are downloaded from the server, otherwise finita la comedy. That is, the minute history stopped being loaded from the more distant past at a later date than the history of older TFs, but I certainly have never seen any substitution. Maybe, MT has a built-in check of free disk space, I don't know exactly yet... but I have never had as deep history of minutes in MT4 as I have in older TFs. Minutes take the most disk space, and I have just enough space... And the terminal may have decided not to download them from the server, although they may have been waiting for me there. But to make it clear, I want to tell you the following: while people are talking about how MT5 1-minute history starts in 1999, it incomprehensibly starts in 2009, and it hasn't been loaded for some reason. That's why I thought the terminal was checking for free space. I have no idea whether they are not really available on MetaQuotes server or I simply cannot download them (I use MT5 mode but as MQL5 programmer I do not know, I have not tried so far). In settings: "Display number of bars on chart: Unlimited". It still does not want to load more bars.

[[Deleted]]

Gentlemen, can you give me a hint?

- There is a dll,

- I know it has the function fn(...),

How to know what type of parameters should be passed to it?

[---]

220Volt:

Gentlemen, can you give me a hint?

- There is a dll,

- I know it has a function fn(...),

How to know what type of parameters should be passed to it?

Open documentation or source code and read.

[x572intraday]

I'm still embarrassed to ask, although the question has been bothering me for a long time...

TheCopyTime copies directly into the buffer what is on the specified timeframe of the called indicator handle. I can be mistaken, but ideologically there is no programmatic algorithmic "slots", that could squeeze in the preliminary calculations of the primary refinement of date-time (we know that non-M1 timeframes have inaccurate time values). I would very much like to know how to defeat such a straightforward copying just through the above function.

[Anatoli Kazharski]

Can you tell me in which cases the value of a tick may differ depending on whether the position is currently in profit or loss?

SYMBOL_TRADE_TICK_VALUE

SYMBOL_TRADE_TICK_VALUE_PROFIT

SYMBOL_TRADE_TICK_VALUE_LOSS