Solar wind - page 12

New comment
 
fxbs:
ok, just checking, to make sure there is no mistake...

again, SolarWind-ACS1.mq4 (3.8 KB, post#102) default Option (option=1; // option 1 "will make it NOT a repainter (except for last bar). " - a bit overstatement, don't you think?

still confusing?

I am sorry to say that your chart without any comments is still confusing. There are two solarwind indicators and they seem to show different values and different crossings and in one chart they both show the same thing.

I admit that my comment has a mistake in stating that "except for last bar". As per MT4 people, the last bar can include upto last three bars!!! So, the last three bars can be modified and a wrong histogram develops over time which will change upon reset (usually this MT4 "last bar" definition is not a problem for most indicators, but this Solarwind or Fisher method uses last value and when it goes backwards often, messes up by using future bar value as last value). But this difference is usually much smaller than the original version (mostly does not affect the signal crossover). It could be avoided completely, but in those days I was not aware of MT4 definition of "last bar" (some 9 months back). If anyone is interested, I can fix this problem as well. But I do not use this. I simply use Stochoistics(10,1,1) crossing 50 level (which is same as using this indi with a Smooth value of zero, but to reduce noise Stoch(10,3,3) is better; when Smooth value is set at zero, no version will repaint).

I am sorry for the wrong statement and any time waste I caused. Thank you fxbs for catching it.

Regards,

chandra

 

...

To avoid any "confusion" with any chart :

"option 1 will make it NOT a repainter. Other options leave it as a repainter!"

First picture :

Solar wind ACS with optionparameter set to 1

Solar wind ACS1 with optionparameter set to 1

Second picture :

Solar wind ACS with optionparameter set to 1 on a new chart

Solar wind ACS1 with optionparameter set to 1 on a new chart

Same thing will happen when you change time frames, restart metatrader, change symbol... The count of repainted bars I leave to others

For the nth time : what is known as "solar wind" is a wrongly coded Ehlers fisher transform indicator.

Someone tried to convert Tradestation version of Ehlers fisher transform and did a bad job doing it.

There is nothing to "fix" in that indicator. And there are some good Ehlers fisher transform indicators for Metatrader already

 
mladen:
To avoid any "confusion" with any chart :

First picture :

Solar wind ACS with optionparameter set to 1

Solar wind ACS1 with optionparameter set to 1

Second picture :

Solar wind ACS with optionparameter set to 1 on a new chart

Solar wind ACS1 with optionparameter set to 1 on a new chart

Same thing will happen when you change time frames, restart metatrader, change symbol... The count of repainted bars I leave to others

For the nth time : what is known as "solar wind" is a wrongly coded Ehlers fisher transform indicator.

Someone tried to convert Tradestation version of Ehlers fisher transform and did a bad job doing it.

There is nothing to "fix" in that indicator. And there are some good Ehlers fisher transform indicators for Metatrader already

Hi mladen,

SolarWind is exactly same as most Fisher indicator one can find for MT (but normally the solarwind had the Smooth parameter fixed and not available for change as an extern variable, that was the only difference). Most Fisher indies are repainters too (even when the authors claimed otherwise), except for one called Fisher-m11. I have discussed these nearly a year back in Factory and made fixed version before knowing the m11 version existed. So, the same thing I tried with SolarWind.

By the way, why you were testing the same thing twice (it is like tracking two rails seperately! they go to the same place). Both files are exactly same except that their default option is changed to be repainter or non-repainter (more accurately, calculating forward or backward). When you set the same option, they are identical and equal to attaching the same indi twice.

My whole purpose was making alerts for many indies and thats what I did with Fisher and then when someone asked, I added alerts for the fancy named same thing (with option to calculate backward as the original bad way or the proper forward calculation). Hope that clarifies some history!

Regards,

chandra

 

Of me comparing two differently named indicators : If someone names two indicators with different names I assume that there is a reason for it. If those indicators are the same, well, then I do not understand the "naming" logic of those indicators (but that is not a matter of this post).

The question was do they repaint, and I think that pictures itself made the point.

Most Fisher indies are repainters too (even when the authors claimed otherwise)

As of "Fisher indies" : this is what "Fisher" stands for in Ehlers Fisher transform Fisher transformation - Wikipedia, the free encyclopedia It is just a part of it (the line saying : 0.5*MathLog((1+Value)/(1-Value)))

Of repainting : you mean this one is repainting too : https://www.mql5.com/en/forum/174980/page5

achandrasekaran99:
Hi mladen,

SolarWind is exactly same as most Fisher indicator one can find for MT (but normally the solarwind had the Smooth parameter fixed and not available for change as an extern variable, that was the only difference). Most Fisher indies are repainters too (even when the authors claimed otherwise), except for one called Fisher-m11. I have discussed these nearly a year back in Factory and made fixed version before knowing the m11 version existed. So, the same thing I tried with SolarWind.

By the way, why you were testing the same thing twice (it is like tracking two rails seperately! they go to the same place). Both files are exactly same except that their default option is changed to be repainter or non-repainter (more accurately, calculating forward or backward). When you set the same option, they are identical and equal to attaching the same indi twice.

My whole purpose was making alerts for many indies and thats what I did with Fisher and then when someone asked, I added alerts for the fancy named same thing (with option to calculate backward as the original bad way or the proper forward calculation). Hope that clarifies some history!

Regards,

chandra

I do not care less what were, and are, your intentions and history with "Solar wind" indicator

The fact is that the posted versions are repainting ("even when the author is claiming otherwise") and that the "solar wind thing" is closed a long time ago on TSD forum (posts starting from this one : https://www.mql5.com/en/forum/178141/page2 are most educational regarding that issue) Please leave the dead things dead (even dead things deserve to rest in peace)

regards

mladen

 

for people, unfamiliar with simple indicator testing:

put indi on 1 min chart, in a few minutes put same indi on the chart again (together with first instance) - and see the difference

once in a while refresh one instance (just clik its properties)

see clear picture

 

indis been tested:

Solar_Wind_clean.mq4 (2.3 KB)

SolarWind-TK.mq4 (3.3 KB)

SolarWind-ACS1.mq4 (3.8 KB)

(shouldnt do that - but just 4 fun: SW repainted (rp) and nonrepainted(nrp) lines together (SolarWind-TKmx2.mq4 ) - check how rp. line slowly taking needed shape - basically repeats (redraws) nrp with shift to the left

 

Hi mladen,

The two indies have different default settings. You changed those settings to be same and used!

Non-repainting version of Fisher indi existed more than two years back and you have made yours just six months back (I had done working with these long before yours was made). May be you re-invented the wheel. Compared to m11 version, you have fixed smoothing parameters just like in SolarWind (Other Fisher versions I have seen have variable smoothing parameters). You don't seem to understand that Fisher and SolarWind are same, same and same (if a copy of yours is posted as corrected solarwind, no-one will know the difference, because there will be no difference!!!). Yours IS the real corrected solarwind (faaar better than my Solarwind-ACS1). There is nothing in the name, just compare the equations. You can see the Fisher versions with the same problem in the link below (thats where the m11 version made its point).

For those who like to use Fisher with changeable smoothing parameters, please check the link below:

Fisher-m11 (the first proper working Fisher, posted on 23-JUL-2006): https://www.mql5.com/en/forum/173169/page8

Regards,

chandra

 

...

Without hi

Of reinventing the wheel : compare pictures. Upper is m11, lower is made according to John Ehlers original tradestation code. Same?

Of reinventing the wheel again : if you download that document attached here (https://www.mql5.com/en/forum/174980/page5), it contains the original John Ehlers formula as well, so I do not have to reinvent the wheel (the process itself is called reading, but it seems that you have no knowledge of that process)

Of understanding what is the same and what is not : I have five fingers on my hand, all of them are different. Guess which one am I showing to you. See the difference?

bye, bye

Files:
fisher__compared.gif  16 kb
 

For a moment I was sad, thinking that you disappointed me (I really thought of you as of coder)

This post of you restores my confidence in you

Thank you. Keep up the good work

Regards

mladen

achandrasekaran99:
Hi All,

I am sorry about the problems with my old SolarWind-ACS versions. At least I would like to point to correct versions, though they do not have the alerts. Those who want to use corrected version of SolarWind can get proper versions from any of the following links (don't bother about the different names). They will be exactly same when used at default settings (in the release date order):

https://www.mql5.com/en/forum/173169/page4

https://www.mql5.com/en/forum/173169/page8

https://www.mql5.com/en/forum/174980/page5

Those who want to explore more options, please use the version from second link where two smoothing options can be changed (to make it smoother or reduce delay). The second IndexSmoothing should be at 0.5 as per Fisher. The first PriceSmoothing is not part of Fisher and it is just John Ehler's preference (one can explore different smoothing values).

At least this recent arguments will help those looking for SolarWind or Fisher in the future, I just hope!

Regards,

chandra
1...567891011121314
New comment
Reason: