A minute and a half difference between local time and fresh tick time. What to do. - page 9

 
pivomoe:

No. Better then to open a second broker's terminal and extract ticks from two servers at once. I'm thinking of digging in the following directions:

1. Try to pause if SymbolInfoTick takes too long to execute.

2. Reduce number of calls on illiquid characters.

3. change windows settings.

I think we should stop at this point. The main thing for me was to get rid of pauses in minutes. I already have a 5 second delay on a hundred characters. (really it's an evening ).

Maybe you have any ideas about identifying SymbolInfoTick's rebellion ( refusing to give ticks )

I have a suspicion of a lack of resources (network or otherwise).

That's why I'd start with a split EA, each one working in its own thread.

MaybeSymbolInfoTick won't be so slow.

 
Andrey Khatimlianskii:

I suspect a lack of resources (network or otherwise).

That's why I would start with EA separation, each EA would work in its own thread.

Maybe SymbolInfoTick won't be so slow.

Came up with a question, but I think I found the answer in the thread, thanks!
 
Andrey Khatimlianskii:

That is why I would start exactly by splitting up the EAs, each one working in its own stream.

I have tried it. One terminal with three identical EAs exchanging data about the last tick. The result is zero. The ticks are completely identical.

Two terminals from the same broker configured on the same server. One computer. The result is null. Ticks are absolutely identical.

But when the brokers are different, there is an effect. One terminal often has fresher data than the other especially in terms of lagging ticks.

 
pivomoe:

But when the brokers are different there is an effect. One terminal often has fresher data than another especially in terms of lagging ticks.

Interesting, show the lag statistics for different brokers, please.

 
Aleksey Vyazmikin:

Interesting, show the lag statistics for different brokers, please.

What different brokers? There are only two of them.

Very briefly, the difference between BCS and Otrytie is 400 milliseconds at this time on 110 futures in the market review.

Calculated approximately this way: I ran one EA simultaneously on two brokers. Took TimeLocal( in milliseconds) and subtracted from it .time_msc of new tick by symbol. Everything was put in an array. Then I'd arrange them in ascending order. And compared the middle of the array for each broker. The difference is just over 400 milliseconds in favour of BKS.

 
pivomoe:

How is it different? There are only two of them.

Very briefly, the difference between BCS and Opening is 400 milliseconds at this time on 110 futures in the market review.

Calculated approximately this way: I ran one EA simultaneously on two brokers. Took TimeLocal( in milliseconds) and subtracted from it .time_msc of new tick by symbol. Everything was put in an array. Then I'd arrange them in ascending order. And compared the middle of the array for each broker. The difference is a bit more than 400 milliseconds in favour of BKS.

Assumed that MQ have already expanded the number of clients. Well gradually there is progress, I see on smartlab charts from MT5 started to appear.

And the ping is the same - isn't that the point?

Actually I would be primarily interested to know their slippage - BCS is a market maker on Si, and I wonder if this gives an advantage on stops...