Wishes for MT5 - page 24
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
Here you can see that minutes from 1993 (actually, until 1999 instead of minutes are days) take 231 megabytes, compressed and ready to work. The adjacent clocks, however, take up only 4 mb.
If we multiply 52 bytes by 4 400 000 bars, the resulting value is about 230 MB. A simple Moving Average indicator with one working buffer (double) will require 4 400 000 * 8 = 35 megabytes of memory.
When the trader sets "Maximum bars in the window" in the terminal settings, he explicitly specifies "not more than the specified number of bars to be loaded into memory for working". If the entire minute history is loaded (I'm working on M1 chart), some heavy multi-buffer indicators are applied and Expert Advisors are used, the memory cost can be significant. And we cannot think "I work with one window and see 500 bars, so I can unload the rest". You cannot unload - indicators, Expert Advisors, etc. live and work on this data.
So, is that why I recently failed to get up to 1996 on M1 and see detailed minute bars, which were substituted with bars of higher TFs? If so, then the reason of "wrong history extraction" disappears, especially because we are not talking about awry attempts of MQL-script extraction, but the terminal one, and here I am not involved as a programmer, being a simple user (unless I should have set a larger value of max number of bars as a user?)
Regarding the live recalculation of the indicators... It was predictable, just didn't pour water on it, but here's an extra confirmation from the developers themselves. It's clear that to calculate the indicator you have to take into account the bars that are now "out of the picture" from the user. But, if they mean it and not only the visual aspect, then, in my opinion, there's nothing that contradicts the analytical recalculation of indicators at any change of the situation, and not the whole stack, but by using a small memory buffer of historical sector by sector, and then the subresults are glued in one result and the final formation of indicator readings. Isn't it? And if it were multi-threaded (one thread for each subsector), it would be a beauty, so that you would not have time to blink. Well, LIFO would at least relate to displaying a visible part of the graph to the user, although it could also be useful for subcomputing chunks of data, depending on what math approach to implement.
Anyway, all this, I think, should not interfere in any way with a wonderful concept of "single source data - several charts".
And it's not kosher to put all history needed for research into memory and have it all computed at every tick - and in one thread, of course!
...Hmm! But multithreading immediately kills the idea of narrowing the memory segment for calculations. That means as many threads as there are sub-sectors to compute, and the memory occupied would be multiplied in such a way (unless all this cunningly and automatically optimizes during linking build and compilation... However, I'm not quite sure whether we should speak about MQL calculation of bars as it is done in scripts and indicators or such calculations [should] be implemented at terminal code level). As a result, we will probably have to use the same 2 GB. It turns out that multi-threading is good for speed, while sequencing is good for memory consumption reduction.
If I've got it wrong, you can correct me.
It would be good if the built-in indicators had a "Select indicator timeframe" option. The default is the current timeframe. It is clear that we can create any indicator, but there are those who have just started using TM5. And in general, such an option would be useful.
how do you imagine a 30 minute 5 periodic moving average displayed on a daily chart?
is it like a bar graph inside a daily candlestick?
how do you imagine a 5 periodic 30 minute moving average on a daily chart?
it's like a little notch in a daily candlestick?
I think it is very useful to provide each graphical object with date and time of its creation (+ can be the last modification) in the properties. Sometimes it's very important to know, because sometimes you draw something, and then there is something similar next to it, you hesitate... - and you can't remember how long ago you drew that, how long ago you drew that... and which of them stinks of dubious relevance at the moment and needs to be removed to unload the schedule. After all, not everything is always obvious with regard to the market situation itself. So there you go.
A trifle, but pleasant.
О?.. In the explanatory dictionary of the Russian language by S. I. Ozhegov these words on the letter o... I'd like to make a detective investigation of what the developers are hiding in the 4th column of the "List of Objects"... What if it's something of national importance? How do you sleep in the dark? I can't get it out. I've done it this way and that way, with a chisel and a chisel.
This is the number of the window in which the object is located. From the terminal's help (read the help, it's all there):
В данном окне отображается список всех объектов, наложенных в данный момент на график. Наложенные объекты представлены в виде таблицы со следующими полями:
If you tick the "Object" field, the object will be highlighted in the chart. If you press the "Show" button, the chart will be moved to the selected object. If you press the "Properties" button, you will go to editing the properties of the selected object. If you press the "Delete" button, the selected object will be deleted. Use Ctrl+A hotkey to select all objects.
This is the number of the window in which the object is located.
Oh, I see. Thank you. It would even be more accurate to say sub-box, that is, for example, an indicator window (which is not directly superimposed on the chart) within the general window of a single chart.
About the window "List of objects"... Well, maybe it would be logical to make it traditional, with "form memory", the separator to be fixed at the position when you release the mouse button?
Please comment on the appearance of a vertical frontal line in the gap area on some TFs and, if you do not consider it necessary to eliminate it, justify your decision. The same artifact appears with wider candlesticks. I may be wrong, but something like this occurs in MT4, but very rarely. Until now we often see another strange effect in MT4 - a vertical chart gap on small TFs, but not at the front itself, but in the depths of history. That is, there was no gap, but the gap is observed. Maybe some small part of history is "swallowed" by incorrect downloading, or maybe there is another reason.
Thank you.
Wishes!!!! Note!!!!!
Do you have a Documentation, please make the end of the description of the function, etc. by request to open the forum strictly according to this function, very helpful, sometimes you read the description but do not quite understand the effect or influence on the desired construction of the EA and starts losing time, searching the forum maybe someone has already experienced this function, but if you do not understand something here below as Code Base you have done read!!!! What do you say!!!! I think many need it!!!