Old and defunct Holy Graily Bob

[betterskin]

did you ever fix the problem of the EA not closing trades on the wavy ?

[trendinator]

just a thought. in the history check (HistoryOK) it looks for 100 bars. but HGI needs 550 to function correctly as pointed out by tommaso. perhaps worth changing? will test out the new version 3

[SteveHopwood]

just a thought. in the history check (HistoryOK) it looks for 100 bars. but HGI needs 550 to function correctly as pointed out by tommaso. perhaps worth changing? will test out the new version 3

Good point.

The upside of having shells from which to quickly create ea's is that the basics are already there.

The downside is redundant code. HGI ensures there are sufficient bars on the chart. The EA 100 bars check is redundant - at least it is until I finally bite the bullet and incorporate Tommaso's HGI library into HGB. The stumbling block there is not a coding one but an emotional one. I remember only too well the last time I incorporated a library that users needed to download into an ea.

Ask yourself this; given that a lot of people have a lot of trouble getting a basic EA to work, what happens when they then have to incorporate a library? I will do this at some point - but need to be feeling strong when I do so.

[SWG123]

Good point.

The upside of having shells from which to quickly create ea's is that the basics are already there.

The downside is redundant code. HGI ensures there are sufficient bars on the chart. The EA 100 bars check is redundant - at least it is until I finally bite the bullet and incorporate Tommaso's HGI library into HGB. The stumbling block there is not a coding one but an emotional one. I remember only too well the last time I incorporated a library that users needed to download into an ea.

Ask yourself this; given that a lot of people have a lot of trouble getting a basic EA to work, what happens when they then have to incorporate a library? I will do this at some point - but need to be feeling strong when I do so.

Please forgive me if this is a stupid suggestion but HGI v16.04 was designed by Tommaso to circumvent the 550 bar history requirement, so ....

[droland]

just a thought. in the history check (HistoryOK) it looks for 100 bars. but HGI needs 550 to function correctly as pointed out by tommaso. perhaps worth changing? will test out the new version 3

Good point.

The upside of having shells from which to quickly create ea's is that the basics are already there.

The downside is redundant code. HGI ensures there are sufficient bars on the chart. The EA 100 bars check is redundant - at least it is until I finally bite the bullet and incorporate Tommaso's HGI library into HGB. The stumbling block there is not a coding one but an emotional one. I remember only too well the last time I incorporated a library that users needed to download into an ea.

Ask yourself this; given that a lot of people have a lot of trouble getting a basic EA to work, what happens when they then have to incorporate a library? I will do this at some point - but need to be feeling strong when I do so.

wow, you are so right there. Maybe slowly modify your own code to whatever the library has to offer over your own realistic time frame. I have figured out (to some degree) your coding path logic and adding outside libraries really does complicate matters.

Just me gut thoughts, not being a high end coder.

Dave R.

[SteveHopwood]

just a thought. in the history check (HistoryOK) it looks for 100 bars. but HGI needs 550 to function correctly as pointed out by tommaso. perhaps worth changing? will test out the new version 3

Good point.

The upside of having shells from which to quickly create ea's is that the basics are already there.

The downside is redundant code. HGI ensures there are sufficient bars on the chart. The EA 100 bars check is redundant - at least it is until I finally bite the bullet and incorporate Tommaso's HGI library into HGB. The stumbling block there is not a coding one but an emotional one. I remember only too well the last time I incorporated a library that users needed to download into an ea.

Ask yourself this; given that a lot of people have a lot of trouble getting a basic EA to work, what happens when they then have to incorporate a library? I will do this at some point - but need to be feeling strong when I do so.

wow, you are so right there. Maybe slowly modify your own code to whatever the library has to offer over your own realistic time frame. I have figured out (to some degree) your coding path logic and adding outside libraries really does complicate matters.

Just me gut thoughts, not being a high end coder.

Dave R.

The next upside is this: when I decide to start incorporating the HGI library into HGB, I will have Tommaso to hold my hand and mop up my tears.

[milanese]

The next upside is this: when I decide to start incorporating the HGI library into HGB, I will have Tommaso to hold my hand and mop up my tears.

I will do so

[SteveHopwood]

did you ever fix the problem of the EA not closing trades on the wavy ?

No, and it sure looks as though there is one.

[trendinator]

CloseOnOppositeTma is still broken for yellow waves if WaveTradeBlueToYellowOnly=true

i really think the TmaStatus needs to be broken up into 4 different signals, which is really what it is - YellowAbove, YellowBelow, BlueAbove, BlueBelow.

honestly its a real mess at the moment, theres TmaStatus, WaveStatus, BuyWaveSignal, SellWaveSignal, WaveIsBuyTrend, WaveIsSellTrend.

most of them are redundant, it really needs to be simplified in my opinion.

also...
i think ShouldPendingsBeDeleted isn't quite working as I thought it might. say we have Rad trades, then the price retraces away from the pendings, then we get a different kind of signal, say Trend Arrow signal in the same direction. the pendings won't be deleted because it is a different trade type due to:

if (OrderComment() != comment) continue;

what we really want is to delete any pendings if they are more than XX pips (20?) away from the current signal, no? then make new trades at the correct level to the new signal.

correct me if im wrong here.

[acostafulano]

Steve, I think both the functions

- Open only one trade per candle
and
- Min Distance between trades

Are still broken (See screenshots, which corresponds to the attached setfile)

7212 HGIG - M15 - 45SL-35TP.set

To sum it up:
Multiple trades within less than the min distance to each other, and all of them taken within the same M15 candle (if you see, maybe 1 each minute, as long as the signal is still there, I suppose).

USDJPY.PNG

EURCAD.PNG

Now, I cannot code for shit, but I've been looking at codes of other robots with a similar working function and I found these snippets:

if (MyRealOrdersTotal(Magic)==1 && Allow_Second_Trade && Time[0]!=Last_Time && LastOpenType()==0 && Ask<=(LastBuyPrice()-Distance*PointValue()))

I have tried to adapt these piece of code to the way HGB is written and came up with this (just for the sake of trying something rather than just ask you).

//Minimum distance between trades
//Trend
if (TrendMinDistanceBetweenTradesPips > 0)

if ( Ask<=( TrendMostRecentBuyPrice - (TrendMinDistanceBetweenTrades / factor) ) )
(THIS IS WHAT I REPLACED WITH THE PREVIOUS THINGY INTENDED TO DO THIS)
return(false);

This will probably not work (I'm about to try it), since as I said, I can't code, but I'd appreciate if you could point me in the right direction or perhaps use this idea to attempt a new fix of this functions.

Diego