Execution speed on ECN accounts

 

Hello,

have an ECN MT5 (netting) PAMM account on A...ri.

Not very happy with the execution speed. Here is an example of logs

2016.10.28 20:12:25.897    Trades    '15009266': order #1000768391 sell 0.06 / 0.06 GBPUSD at 1.21730 done in 2114.218 ms
2016.10.28 20:12:25.891    Trades    '15009266': deal #1000672621 sell 0.06 GBPUSD at 1.21730 done (based on order #1000768391)
2016.10.28 20:12:23.888    Trades    '15009266': accepted market sell 0.06 GBPUSD
2016.10.28 20:12:23.783    Trades    '15009266': market sell 0.06 GBPUSD

2016.10.28 19:22:07.439 Trades '15009266': modify buy 0.11 XAUUSD -> sl: 1270.010, tp: 0.000 done in 2106.397 ms
2016.10.28 19:22:05.468 Trades '15009266': accepted modify #1000767972 buy 0.11 XAUUSD sl: 1269.380, tp: 0.000 -> sl: 1270.010, tp: 0.000
2016.10.28 19:22:05.332 Trades '15009266': modify #1000767972 buy 0.11 XAUUSD sl: 1269.380, tp: 0.000 -> sl: 1270.010, tp: 0.000

The speed of any order is about 2 seconds. During this time the price can change significantly (sometimes it is more profitable, sometimes not).

1) Ping time to server is 80 ms. Where 1.9 seconds is spent (minus the ping). Are the A...ry servers at the other end of the earth from the liquidity providers' servers? In this situation, put the terminal on a server closer to the server broker makes no sense, will not be 1.9 + 0.1 = 2 seconds as now, but 1.9 + 0.005 = 1.9005.

2) Share your data on delays on ECN, for comparison .

3) There is an assumption that trades at 0.06 or 0.11 are not sent to the liquidity provider, but a total pool is accumulated and sent to execute at least 1 lot. Is this true? Although..... probably not the reason, because the command to change the SL also took a long time to be executed (( It is not sent to the liquidity provider.

4) If trades will be in whole lots (1, 2 ... 5 ... 24 ... 51) i.e. without decimals and hundredths - will they be executed faster? Or just as slowly? Does anyone have any such observations?

5) How to make such 2 sec delay for strategy tester? So that it optimizes more correctly. Arbitrary delay is apparently much smaller (tens, hundreds of ms., but not 2 sec.).

 
elibrarius:

Hello,

have an ECN MT5 (netting) account on A...ri.

Not very happy with the execution speed. Here is an example of logs

The speed on any order is about 2 seconds. During this time the price can change significantly (sometimes it is more profitable, sometimes not).

Ping time to the server is 80 ms. Where wasted 1.9 seconds (minus the ping). Are A...ry's servers at the other end of the earth from the liquidity providers' servers? In this situation, put the terminal on a server closer to the broker's servers makes no sense, it will not be 1.9 + 0.1 = 2 seconds as now, but 1.9 + 0.005 = 1.9005.

1) Share your data on delays on ECN, for comparison .

2) There is an assumption that trades at 0.06 or 0.11 are not sent to the liquidity provider, but a total pool is accumulated and sent to execute at least 1 lot. Is this the case?

3) If trades will be in whole lots (1, 2 ... 5 ... 24 ... 51) i.e. without decimals and hundredths - will they be executed faster? Or just as slowly? Does anyone have any such observations?

4) How to make such 2 sec delay for strategy tester? So that it optimizes more correctly. Arbitrary delay is apparently much smaller (tens, hundreds of ms., but not 2 sec.).

Called du...py, they claim execution time of 30-300 ms. In practice, I did not measure it myself, but when trading, it was very fast, managed to open more than 10 deals per second.
 
At a ping of 2.50 ms or so, the minimum execution was 100 ms, the modification of the order of 20 ms or more, account type ECN-NDD real
 

What is interesting is that such a long delay appeared on 14 October. Transactions from October 13 and before were around 200ms. Here are the latest "fast" trades from October 13.

2016.10.13 23:59:07.191    Trades    '15009266': market sell 0.10 EURUSD tp: 1.10503
2016.10.13 23:59:07.289    Trades    '15009266': accepted market sell 0.10 EURUSD tp: 1.10503
2016.10.13 23:59:07.397    Trades    '15009266': deal #1000615869 sell 0.10 EURUSD at 1.10562 done (based on order #1000705910)
2016.10.13 23:59:07.398    Trades    '15009266': order #1000705910 sell 0.10 / 0.10 EURUSD at 1.10562 done in 207.859 ms
2016.10.13 23:59:15.102    Trades    '15009266': modify #1000705492 sell 0.60 EURUSD sl: 0.00000, tp: 1.10503 -> sl: 0.00000, tp: 1.10513
2016.10.13 23:59:15.200    Trades    '15009266': accepted modify #1000705492 sell 0.60 EURUSD sl: 0.00000, tp: 1.10503 -> sl: 0.00000, tp: 1.10513
2016.10.13 23:59:15.310    Trades    '15009266': modify #1000705492 sell 0.60 EURUSD -> sl: 0.00000, tp: 1.10513 done in 207.387 ms

What could be the reason? Was it because of successful trading? Have they decided to put a stick in the wheel? In those days was a huge profit of trading, scalping manually, there were about 500 trades a day and profit more than 100% per day. Well, since monday I started to lose everything - the sticks worked( Clicking - and waiting for 2 sec. reaction.... And psychology too... Fast growth makes your head spin, you start to think that erroneous trades can be outlasted. I don't rule out that the 2sec slowdown has also added its share to the psychological state.

Although DC has earned the most in this epic, more than $1000 during the week on commission and swaps. In theory, they should not have hurt such a profitable client.
 

If it was MT4, we could safely say that the anti-scalper plugin is worthwhile.

If they made such a plugin for MT5, it would not fit the title of a brokerage platform if they interfere with execution.

they are not good for the title of a brokerage platform if it is not a plugin, then the above mentioned brokerage is not good enough to be a liquidity provider, as stated recently in the news with such sluggish executions.


i have no idea what to do if i got a response from tech support about performance?

 
ivanivan_11:

If it was MT4, we could safely say that the anti-scalper plugin is worthwhile.

If they made such a plugin for MT5, it would not fit the title of a brokerage platform if they interfere with execution.

they are not good for the title of a brokerage platform if it is not a plugin, then the above mentioned brokerage is not good enough to be a liquidity provider, as stated recently in the news with such sluggish executions.


i have a good feeling i will not be able to use it in the future.

I don't know why i got so disappointed in them but they were wrong. But with the drawdown to the initial deposit it was time to analyse and analyse the situation.
I wrote to them in the forum today as well, no reply yet....
 
welcome to the ranks of the unhappy forex performance! :) that's their bread, you can't prove anything to anyone and no one will tell you anything... You have to make conclusions, it's the same everywhere. If you want something normal, you have to go straight to LP (a real one, not fake like al...). No scalping will work that way:)
 

Here's a similar example of my trading from another dtz - 3x increase in execution time and slippage. (I have been trading in the same place for a long time, but I'm not sure what to do. The quality of execution was logged all the time, so it's not a flaw of the system but banal manipulation of execution.

 
Maxim Dmitrievsky:
welcome to the ranks of the unhappy forex performance! :) that's their bread, you can't prove anything to anyone and no one will tell you anything... You have to make conclusions, it's the same everywhere. If you want something normal, you have to go straight to LP (a real one, not fake like al...). If you want a good one, go straight to LP (the real one, not fake like al...) no scalping there :).
I thought that on ECN their bread is commissions. I made $1000 in one week. And at the end I lost mine and up to 50% of my investment to the investors who appeared. What do they benefit from the delayed execution?
 
elibrarius:
I thought ECN was their bread and butter - commissions. I made $1000 for them in a week. And at the end of it, I lost mine and up to 50% of my investment to the investors who showed up. They also have a delay in execution - how do they benefit from it?
I think with such trades and 0.1 second will be a lot))).I mean that sooner or later will be lost you and 0.1 will not be satisfied))) It's like a dose of drugs))) Always want to increase))
 
Movlat Baghiyev:
I think with this kind of trade 0.1 seconds is a lot))).I mean, sooner or later you will plummet and 0.1 will not be satisfied))) It's like a dose of drugs))) Always want to increase)).
There are good ideas for the future. I was happy with 0.2 seconds - I pressed it and it worked. 0.1 sec is not really necessary for manual trading. But when it became 2 seconds it is already too much.