@tankhang:

This indicator keep on ‘shouting’.. I guess that is the problem..

Not really – this is caused by the bug, but it’s not the reason for the bug.

@Lowphat:

all the loops seem odd.

I don’t think so!

prev_calculated == 0 still seems strange.

No, this if is intended to perform calculations that only should run when loading the indicator, and…[Read more]

@Lowphat: not quite, but you’re close!  

Found a nice and clear article about using and abusing inline comments in your code:

http://www.yacoset.com/Home/inline-documentation

Recommended reading!

s.

Hahaaaa: a true newbie bug in this version, which caused that high cpu resource consumption!

Who can find it?

Will be fixed in next release, to come soon.

s.

@gg53: thanks for testing and reporting!

array out of range: there’s obviously insufficient history in one of the symbols involved on your system. I’ll add a history check in a future version. Meanwhile you can reduce ‘maximum bar count’ to avoid this error.
lines ‘disappear’: this is caused by the session reset algorithm. You might consider…[Read more]

Ok, those mysterious symbol prefix and suffix again! I never understood why some brokers pretend to use them, or why some traders are choosing those brokers.

This is a quick patch, version 2.1 showing input parameters for symbol prefix and suffix.

Request for test: please would somebody who is connected to a broker with symbol prefix /…[Read more]

SPAM!

Ok guys!

Next public version of my CIX, including source code. I consider this one much more mature than the initial version 1, published several weeks ago. Especially the session reset algorithm is now working every day, not only once.

It also includes an alternate RSI normalization to choose from, alongside with my timeframe based…[Read more]

Ok guys!

Next public version of my CIX, including source code. I consider this one much more mature than the initial version 1, published several weeks ago. Especially the session reset algorithm is now working every day, not only once.

It also includes an alternate RSI normalization to choose from, alongside with my timeframe based…[Read more]

@gg53:

I’m happy that you changed your mind. This is what I proposed (and coded) several weeks ago, and you did not seem to like it.

See my next post for details and a new version.

s.

Ahhh, one more thing:

Good style does not mean an abundant number of comments in your code. These will only be disturbing, obfuscating the very structure of your algorithms.

Good style means that your code is so easy to understand, that it requires only a minimum of comments, and that you inserted exactly this minimum into your code.

s.

Hi Guys!

From code posted and code sent to me, I’ve got an idea now that most of the coders around have a sense about the basics of code formatting, indentation, and commenting. Yet it might be useful to review an article that deals with the very basics of these topics. The article uses C code for examples, which is very similar to MQL.…[Read more]

@tankhang: great, good luck!

s.

@smallcat:

That factor 100 will simply transform a simple relation (a-b) / (a + b) for easier reading, that’s all: for a = 4 and b = 1 you will get 60.0 instead of 0.6. Simple scaling for convenience.

Your code in general: avoid variable names that say nothing, like buffer1 or buffSix. If you’re using meaningful names like volumeBullish and…[Read more]

@smallcat:

That factor 100 will simply transform a simple relation (a-b) / (a + b) for easier reading, that’s all: for a = 4 and b = 1 you will get 60.0 instead of 0.6. Simple scaling for convenience.

Your code in general: avoid variable names that say nothing, like buffer1 or buffSix. If you’re using meaningful names like volumeBullish and…[Read more]

@smallcat:

That factor 100 will simply transform a simple relation (a-b) / (a + b) for easier reading, that’s all: for a = 4 and b = 1 you will get 60.0 instead of 0.6. Simple scaling for convenience.

Your code in general: avoid variable names that say nothing, like buffer1 or buffSix. If you’re using meaningful names like volumeBullish and…[Read more]

Hi guys!

A short advice for all unexperienced coders. When writing an indicator, you usually want to achieve 2 basic goals, no matter which calculation the indicator will perform:

When loading initially, the indicator should calculate and display its values for a certain number of bars or for all bars on chart, if possible.
After that, the…[Read more]

And this one seems to be interesting.

s.

Hi @smallcat,

I found this one. It’s a bit older and I don’t know whether it works on latest R / MT4 builds. Also some sample code in that thread.

s.

Thanks and yes:

1 second is possible, but much too ‘nervous’. 5 to 10 seconds seem to be a good choice to indicate where ultra short term momentum goes.

Have to try next week whether resetting the beginning of the time window to zero is a good idea.

s.