Posts made by fxDreema

Sure, let's see it.

Let's define what "trial period" means. So you are talking about the time needed for an EA to be backtested? If this is the case, I can expect EA created with fxDreema to be the slowest one of both, because fxDreema was made in a way to support various strategies and there are extra things in every block that can slow down testing. For example, block parameters - there are blocks with many input parameters and when such a block is executed it checks those parameters. Or if "on Trade" is used, because it is not native for MetaTrader 4. And it also depends on how the strategy was built, what blocks are used and how often are they executed. I beleive it's possible to make things faster, I am already aware of some things that slows down the process, but a well designed custom made EA will always be a bit faster.

EAs created with fxDreema does not have trial period normally. There is small option in "Project Options", but it is empty and inactive by default.

Е, точно и конкретно за това няма блок, но все пак става въпрос за сравнение на няколко нива, които са известни, та мога да предложа "Condition".

That looks bad. I don't know exactly why it happened, so I can recommend to relod the generator (or the project), to move one step back (before that paste) and to try on clear

I don't know how a Buy pending order can have Open price in the middle of the previous candles and SL at 10 pips above the High of the same candle, but anyway...

http://fxdreema.com/shared/oSr7HQEnb

I used Variables here and some calculations to get 3 levels - open price, SL and TP. All calculations can be found in block 1. Blocks 2, 3 and 4 are used to plot those levels on the chart - Open price is brown, SL is red and TP is blue. But I guess some conditions are needed before placing new order - you probably need to check the type of the previous candle and it's size. There are 2 blocks that can probably help - "Bull candle" and "Bear candle", but if they can't help - "Condition" is over there.

I recommend to use the same group of blocks instead of copy-paste them. I know it feels natural for beginner-programmer to copy-paste things, but this is not a good idea actually. Well, fxDreema is not exactly programming tool and it's not that flexible as the plain code, but the best is if any part of the program exists in one place and if needed it is used multiple times. Using Variables can help with this in general, not in every situation, but if it is possible use the same group of blocks with different parameters Even better is to program custom code with all the specific stuff, but this is for someone who knows how to do it

The way to save a title is to click on the block again so the message saying that parameters were updated will appear. At least I tried this now and it worked.

SL and TP are initially set in the blocks where you create trades - Buy now and Sell now. They can be placed at some distance (or example 10 pips) or they can be placed at some other level - see Dynamic options. If SL and TP needs to be changed after that, then other blocks should be used.

Objects are identified by name in MT, that's why they are sorted by name. But I was not added functionality to create fancy objects like fibonacci, so "For each Object" can help if there is already created object.

Don't you want to just check if some part of the candle is ~40% of some other part of the same candle?

Fibonacci is an object normally, some people use indicators as well. What do you use, or how do you know that level 38.2? And if this level is for a candle, is fibonacci really needed?
SL and TP are easy. Do you tried to do something at all?

I can't see partial close here, but otherwise yes - there is some order in blocks execution after all, it depends on block numbers.

Pass block does nothing actually, and that message has nothing to do with blocks. When EA is started it says "Starting" and when the first tick comes it is changed to "Working". In MQL4 language "Starting" shows with init() and "Working" shows with the first execution of start().

Click on "Expert Advisors" button on MetaTrader to enable them?

I can only guess, but one reason can be that the trade is closed just before reaching this block. Do you partially close trades somewhere above this block?

Blocks placed under "on Tick" are executed on every tick - again and again and again, maybe few times per second. At least first ones above others. A block, when executed, does what it has to do and it can pass or not depending on what it does (see it's description for information about that).

Some blocks are conditions, others are filters (like "No trade is running")... and filters are conditions on steroids.
Other blocks are actions, like "Buy now", "Sell now", "Close trades" and all the others that do changes somewhere. "Draw arrow" can be also called action block, but it is more for debugging (seeing what is going on).

Connect blocks like this: Conditions, Filters => Actions

If you don't know what some block does, there is an information about it. Hold your mouse over the block in the list, or if you have it as a block - right click and hit "Information".

Close losable trades, as it's name says, closes trades with negative profits when that block is executed. There are options to choose which trades to work with (filters) and what loss to wait for.

http://fxdreema.com/shared/9V6ldZNVc

In the EA, at the point where trades needs to be closed, have "Close each trade" and it will do exactly this - it will close each trade, as it's name says. Additionaly, it has options inside to choose which trades to close.

There is a block "Close trades" and another one - "Close losable trades". Additionaly, to close trades by more complex rules, a loop can be created (pink blocks) - read the description of "For each Trade" in this case.

При мен никога не е било различно Мога да настроя нещо да е печелившо за определен период от време, примерно с тоя блок за една година между 2011 и 2012 (така си го бях сложил) като най-добър резултат постигнах от 10000 до над 70000, ама с вероятно по някаква случайност и то с голям drawdown. И ако същото нещо го пусна на друг период от време... кофти работа, някъде умира.

Макар че аз не съм сериозен тестер и не бих сложил голяма тежест на думите ми, но си мисля че такива математически модели дето гонят някаква обикновено проста последователност от действия не могат да са винаги печеливши. Все си имат едно слабо място (последователност от ситуации водещи до фалит), което поради разнообразието на графиката ще се достигне. Което понякога ме води до едни други разсъждения...

Имам един блок за правене на някакви патерни, там в третата колонка от блокове. Та интересното, и нескопосаното при него е, че като си направиш един патерн, особено от 3, 4 или повече барчета, и става доста трудно да се намери съответстващ патерн на графиката. Понякога трябва да минеш на подходящ таймфрейм за да се намери нещо въобще. Ако приемем, че следваме някаква математическа стратегия (дето не се интересува от новини и фундамент) и тя има слабо място - патерн, при който се случва бедствие - то този патерн ще е най-вероятно да се случи в някой таймфрейм и най-невероятно да се случи извън някой таймфрейм. Та извода ми е, че за да се прескочи фаталното развитие на нещата трябва да се играе в мащаб или под М1 или над D1 Ама под М1 има разни проблеми с тестването, после с брокера... а над D1 нещата се случват толкова бавно, че като цяло печалбите накрая са по-малки (по-малък брой поръчки), а и докато ти висят поръчките може да понатрупат някой минус само от висене. То пък тестера не може да изчислява тия разходи.