Errors, bugs, questions - page 654

[pusheax]

I apologise for the insistence.

I've already written here that for EURUSD at any timeframe in the Strategy Tester

values are high, low, open, close absolutely the same, which shouldn't be.

Could you correct the history for EURUSD of Strategy Tester, because

Other brokerage companies seem to have copied your history and give the same data as you do.

Here is an example of getting EURUSD D1 timeframe data, you can see that high,low,open,close values are equal.

[Rashid Umarov]

pusheax:

Could you correct the history for EURUSD strategy tester, because

Other dealers seem to have copied the history from you and give the same data as you do.

Here is an example of getting EURUSD D1 timeframe data, you can see that high,low,open,close values are equal.

What modelling mode are you using? How this data is obtained?

Read the Basics of Testing in MetaTrader 5

[Slava]

pusheax:

I apologise for the insistence.

I've already written here that for EURUSD at any timeframe in the Strategy Tester

values are high, low, open, close absolutely the same, which shouldn't be.

Could you correct the history for EURUSD of Strategy Tester, because

Other brokerage companies seem to have copied your history and give the same data as you do.

Here is an example of getting EURUSD D1 time frame data, you can see that high, low, open and close values are equal.

Could you please describe the way of receiving this data in details?

It seems that you have started testing in "open prices" mode and analysed the current bar which has just opened.

[Mykola Demko]

tol64:

Haven't even heard of it being done yet. :)

//---

Anyway, I've done a little bit of figuring it out for myself. Came to the conclusion that the period of global code optimization began. This is probably the next step of development. Now I will optimize everything in a row.

Urain:
What would you recommend for testing a video card in that case? Is it suitable for the terminal?

Go for something like that, then.

You won't go wrong. It's never short of power.)

I have a video card, and it suits me.

The question was that man chooses the same laptop, somehow trying to understand "whether it is suitable for the terminal?

And since laptop manufacturers resort to such tricks,

it's logical to ask the developers "what can we test the suitability of the model for the terminal?" if 3D games aren't suitable for it.

[pusheax]

Rosh:

What simulation mode are you using? How is this data obtained?

Read Basic Testing in MetaTrader 5

simulation mode:

Open prices only

Data is obtained by using function: CopyRates

I get data for 22 currency pairs in loop: {"GBPUSD", "EURUSD", "USDJPY", "USDCAD", "AUDUSD", "NZDUSD", "USDCHF", "USDSEK", "CADCHF", "AUDCAD", "AUDCHF,
"AUDJPY", "AUDNZD", "CHFJPY", "EURGBP", "EURAUD", "EURCHF", "EURJPY", "EURNZD", "EURCAD", "GBPCHF", "GBPJPY"};

And only with EURUSD is there such a problem.

[Aleksey Lebedev]

pusheax:

high,low,open,close values are exactly the same, which should not be the case.

on the first tick of the current bar high,low,open,close are always exactly the same

[pusheax]

stringo:

Could you describe in more detail how you obtained this data?

It looks like you were testing in "open price" mode and analysing the current bar that has just opened.

I have been testing on GBPUSD.

Thanks for the tip, I will try to test on the previous bar.

I must have really screwed up.

[Rashid Umarov]

Urain:

And since laptop manufacturers are resorting to such tricks,

it's logical to ask the developers "what can be used to test the suitability of the model for the terminal?" if 3D games aren't suitable for it.

I won't say about the tricks, I'm not sure. Something similar has been discussed here - https://www.mql5.com/ru/forum/109284/page2

[Anatoli Kazharski]

Urain:

... then it is logical to ask the developers "what can be used to test the suitability of the model for the terminal?" if 3D games are unsuitable for this.

Well yeah, that's an interesting question. I remember I used to have a recording studio in Moscow. And I once got so tired of managers trying to push some nonsense in the "just to sell" way, that I became very distrustful of them. I urgently needed a new computer for recording, the old one was no good at all for the new project and clients were due to arrive the next day. I ended up just taking a piece of studio equipment straight to the sales office and giving it a global test run. :) It was fun. I had showed them that the "most powerful computer" that they had tried to push into me was not so powerful for my quite simple studio tasks. All the same, I had to take what I had, as I had no more time, but at least I was aware of it. The project stretched barely, but nobody noticed anything. :)

[Mykola Demko]

Rosh:
I won't say anything about tricks, I'm not sure. Something similar has been discussed here - https://www.mql5.com/ru/forum/109284/page2

Thanks for the link, then a follow-up question: can this code test graphics card performance?

#property script_show_inputs
input uint Count=1000;
void OnStart()
  {   
   uint st=GetTickCount();   
   for(uint i=0; i<Count; i++)ChartRedraw();
   Alert("Count ChartRedraw()=",Count," time=",GetTickCount()-st," mk.c.");   
  }