MetaTrader 5 Client Terminal - page 26

New comment
 

When is the metra trader 5 out. Has brokers started using it yet?

 

i don't think there are any brokers using mt5 yet.

 
witsnpips:
i don't think there are any brokers using mt5 yet.

http://www.fxcentral.net/dwd/fxc5setup.exe

But i see, it is only available on demo at moment, as MT 5 is still under beta phase it is unlikely available for live traders for another few quarters

 

MetaTrader 5 Strategy Tester Released!

MetaTrader 5 Strategy Tester Released!

 
amarnath:
Hi all,

4shared.com - online file sharing and storage - download fxc5setup.exe

You can try this level 1 ECN stuff with no gimmicks and tricks like other ECN does, you can enjoy nothing but pure liquidity and fills and commission at moment reflecting is pretty not correct to international standards but soon it will be announced on their revamped site with complete details

But i see no strategy tester in this installer? doesn't it updates to strategy tester via live updates mechanism?

 

Any plans to support other asset-classes/markets (apart from Forex, options, futures, stocks)? e.g. CFDs, forwards, betting...

 
Lenar:
MetaTrader 5 Strategy Tester Released!

Hello Lenar,

In current version of MT4 build 226, strategy tester is not performing well, on 5 digit brokers. (this is not related with variable spreads of brokers) it was in a good state when brokers was 4 digit. Some real trader/users should debug the behaviour of tester, when 4->5 digit changes occur. This problem is possibly related with simulation of data with 5 points, it sounds, like some floating number steps distribution is not linear (randomly changing steps), because of crude calculation. if customer can set simulation steps ( 1/10 pip, 2/10 pip, 5/10, 10/10 pip etc.) this may produce a good behaviour, like when it was great on 4 digit tests.

Currently, I can not trust 5 digit tests. Meanwhile, I'm not a beginner..

(Good State is: if tester trades and real trades match %95+, in some months of forward test. and tester can allow to build such an expert: inconsistency of multiple TF usage, multi-pair data simulation etc. is removed/eliminated, distortions of tester removed, "tester knows future" problem removed, daily bars corrected etc. )

Thanks

Cengiz

 

Lenar,

Can you tell me:

1) are you still planning to release the new platform in Autum 2010?

2) when exactly can I start beta testing of the new client terminal and MQL5 (you wrote "in summer")?

 

if an expert is using multiple time-frames

more description

if an expert is using multiple time-frames, tester can produce different test reports for different time-frame tests. Different means, really diferent.

some of my system is using H1 trend, and some trigger from M15. and M15 tests fails, because, H1 trend indicator do not work on-time (H1 not simulated while M15 is simulating?) causing me total failure of tester.

another problem was, W1 is not simulated and tester knows the future, if W1 is used in your system (completed bars)

So, I can say that, for metaquotes, tester is not a critical tool. You need to discover it's bad points, and you need to avoid them. The most important issue is, for a pair (you can not test multiple pair experts), if an expert is using multiple time frames, (and if expert is using external indicators), tester can not (do not?) simulate all time frames perfectly. And garbage.. You are muddle. So, people do not belive back-tests, normally. Because, tester is lier, for many complex experts, if you are not aware, what is really simulated, and what is" freezed" for some periods..

Thanks to metaquotes for these tools. And hope they focus on tester, with real expert traders, to fix it's weaknesses. So, people starts to belive in back-test. The key is, [states of ALL time-frames is not correct at a given time of simulation. some TF/indicators is living in another time].

For developers: if your expert is producing different results for different TF tests (every tick mode), you are in garbage. Tester can not test your expert correctly, since, at a given time, TF's are not in the same exact time, when tester is working.. Even, I can say that, (control point mode) test can produce more correct results, for some experts, since, TF's time synchronization is better (?)

And, possibly some messaging problem exist, between tester process and indicator. If an indicator is using H1 TF, but tester is running at M15 TF, H1 indicator can not refresh itself correctly. If you run H1 tester, then indicator can refresh itself correctly. ( [Tester-> multiple time-frame indicators, messaging problem..?] )

Thanks

Cengiz

Tester in the Terminal MetaTrader 4: It Should Be Known: https://www.mql5.com/en/articles/1490

 

Here is my first template for MT5

Use 15min chart for direction of main bias and trade the 5min chart in that direction.

Just copy xard.tpl to your hard drive then from within mq5, right click on chart and select load template then browse to wherever you saved the xard.tpl file

Xard777

1...1920212223242526272829
New comment
Reason: