When is the new version of MT5 and where do you find out what's expected in it? - page 19

 
Karputov Vladimir:

Could it just be that the load percentage is not displayed correctly? I.e. is it really 10% and 20%?

Did the tick download continue after this message ((200%) download)?

This is what happened next:

2016.04.03 19:37:11     Core 1  agent process started
2016.04.03 19:37:11     Tester  EURUSD: ticks data begins from 2016.03.01 00:00
2016.04.03 19:37:11     Tester  EURUSD: preliminary downloading of history ticks completed
2016.04.03 19:37:10     Tester  EURUSD: 94% ticks downloaded
2016.04.03 19:37:07     Tester  EURUSD: 70% ticks downloaded
2016.04.03 19:37:06     Tester  EURUSD: 55% ticks downloaded
2016.04.03 19:37:05     Tester  EURUSD: 40% ticks downloaded
2016.04.03 19:37:03     Tester  EURUSD: "bases\MetaQuotes-Demo\ticks\EURUSD\201603.tkc" (200%) download
2016.04.03 19:37:01     Tester  EURUSD: "bases\MetaQuotes-Demo\ticks\EURUSD\201604.tkc" (100%) download

If this is a continuation, they could have designed it in the same style.

Unfortunately, no increase in test speed was noticed in 1295. As it was 10 minutes to run the test and remains so. To be honest, I do not understand why we have to download the tick history and spend a lot of time on its synchronization. Neither symbol, nor timeframe, nor period, nor testing type is changed!

Yes, another phrase seems to be a gaffe:

2016.04.03 19:37:01     Tester  EURUSD: preliminary downloading of history ticks started, it may take severe minutes

Probably meant "several".

 
Yes, plus, in the tester, in the data window there is no line with real volumes. But this was also the case in previous builds.
 
Oleg Tsarkov:
So mt4 won't be needed anymore if brokers scratch their heads and start serving mt5 hedge accounts.

Many people thought the same back in 2009 09 09 - that's the release date of mt5! and it's been 7 years already

brokers don't give a shit ... they don't care ... they're fine as they are ... why change something?

They don't have any plans to change MT5.

And if a business feels that they will lose a bunch of customers from a hard forced migration from mt4 to mt5 - business won't kill mt4!

Instead of killing MT4 + MQL4 MetaQuotes, they just kept on developing the MT4 and MQL4 capabilities - not stupid people are sitting there, you see how they crossed your mind .


You see, they crossed out your idea. And no one rushed to rewrite a lot of mathematics from MT4 to MT5 - but it's a well-known fact that MT5 unfortunately does not have the whole spectrum of what has been developed for MT4

--

I am glad that MT5 has appeared the hedging element, but this solution is 7 years late.

no broker still supports hedging - until they update the maths - if it's not a hard requirement - believe me no one is in a hurry - and even if they do ,

no one will give up mt4 until the client leaves mt4 and this is a long process.

And even if there is no real ticks in MT4 - well, no problem - they can always be added to the tester's FXT files using third-party tools.

 
Alexey Kozitsyn:

Discovered an interesting inscription while testing on real ticks:

Can anyone explain the 200% load? What does it mean? Build 1295 x64.

No one can explain it. Remains from the very first debug builds when we started testing tick load.

Thanks for the comment. We'll fix it

 
Alexey Kozitsyn:
Yes, plus on top of that, in the tester, there's no row with real volumes in the data window. But this was also the case in previous builds.

In previous builds there was also a line with the actual volume in the data window. You just didn't configure it to show it. It's exactly the same as in the client terminal, using the context menu


 
Slawa:

In previous builds there was also a line with the actual volume in the data window. You just didn't configure it to show it. It's exactly the same as in the client terminal, using the context menu

Yes, sorry, didn't take that into account.

Slawa, are you planning on speeding up the tests by real ticks? 10 minutes now from start to start testing, with the ticks being reloaded/synchronised every time.

And also, if after switching to another testing method, the same procedure of loading/synchronizing all ticks. Just as long.

 
Alexey Kozitsyn:

Slawa, are you planning to speed up the tests by real ticks? 10 minutes now from start to start of testing, with ticks being reloaded/synchronised every time.

And also, if you switch to another testing method afterwards, the same procedure of loading/synchronising all ticks. Just as long.

How is it?

There was a "real ticks" method. Then you replaced it with a method, say "m1 ohlc", but the real ticks are still synchronised?

 
Slawa:

How's that?

There was a "real ticks" method. Then you replaced it with a method, say "m1 ohlc", but the real ticks are still synchronised?

Yes, it's still as long!

I restarted the terminal to check, here's the log:

2016.04.04 14:51:33     Tester  quality of analyzed history is 99%
2016.04.04 14:51:32     Core 1  EURUSD: history downloading canceled
2016.04.04 14:46:32     Core 1  EURUSD: history synchronization started
2016.04.04 14:46:32     Core 1  common synchronization completed
2016.04.04 14:46:32     Tester  EURUSD,M5 (MetaQuotes-Demo): visual testing of Experts\Examples\Moving Average\Moving Average.ex5 from 2016.03.01 00:00 to 2016.04.03 00:00
2016.04.04 14:46:32     Core 1  authorized (agent build 1295)
2016.04.04 14:46:32     Core 1  connected
2016.04.04 14:46:28     Core 1  connecting to 127.0.0.1:3000
2016.04.04 14:46:28     Core 1  agent process started
2016.04.04 14:46:28     Tester  EURUSD: preliminary downloading of history canceled
2016.04.04 14:41:27     Tester  EURUSD: preliminary downloading of history started

As I said, same 10 minutes, only on OHLC on M1.

 

The question is - where will mt5 get these real ticks from or will it have to feed them to the mt5 itself and in what format? With mt4 it is clear how to feed it real ticks, but not with mt5. It would be good to write a separate article on this subject.

 
Real ticks are on the MT5 server
Reason: