T3 - page 12

 

...

The modification by Fulks and Matulich that actually makes T3 different from the original Tim Tillson formula - it makes it faster but it is not the original (red line bellow - line 169 in the source)

if (!t3Original)

n = 1 + 0.5*(n-1);

w1 = 2 / (n + 1);

w2 = 1 - w1;

You can see that it is avoided if the t3Original is set to true

fxbs:
demit, i been searching my computer like 4 an halfanhour like crasy, found ind. you deleted, but you posted new version alreay! thank you Mladen! ( error we discovered only much later )- what error you mean?
 

i missed the school when they teached math, but you and Linux knew it all along!

 

little tricks (not trix) with bands

Files:
 

32000 shades of red/blue

Just jet an other T3 version. I wrote this some time ago, just felt like sharing. May be more show then goo however, but looks pretty IMHO.

Also, this one demonstrates uses of Arraycopy to emulate Indexbuffers (I have not seen this before). Same technic can be used in EA's, very handy sometimes.

Enjoy.

 

Bugfix

There was a small bug in Pretty_t3_LK_V1.mq4 realated to PriceShift -> Fixed. (Also, as someone pointed out by e-mail: Infact only 254 shades of red/blue are used. I guess I just can't count today.) Sorry about the mess.

Files:
 
LastViking:
There was a small bug in Pretty_t3_LK_V1.mq4 realated to PriceShift -> Fixed. (Also, as someone pointed out by e-mail: Infact only 254 shades of red/blue are used. I guess I just can't count today.) Sorry about the mess.

LV, you can just edit your posts -no time limit

thanks 4 indi

 
LastViking:
There was a small bug in Pretty_t3_LK_V1.mq4 realated to PriceShift -> Fixed. (Also, as someone pointed out by e-mail: Infact only 254 shades of red/blue are used. I guess I just can't count today.) Sorry about the mess.

so only 508 colors? - anyway you right - very pretty

(there also been Plot_ ind. series (like shaff plot, etc - plots obj. lines) - yes, they are beautiful)

LV, people would ask - how to call it's main t3value from outside (no buffers) - may be put one just in case (draw none)?

 
fxbs:
so only 508 colors? - anyway you right - very pretty

No coment.

fxbs:
(there also been Plot_ ind. series (like shaff plot, etc - plots obj. lines) - yes, they are beautiful)

Thanks for pointing out, I hade not seen them. Nice!

fxbs:
LV, people would ask - how to call it's main t3value from outside (no buffers) - may be put one just in case (draw none)?

I can do that. But the main t3 value vill be the same as the usal t3ma so maybe not mush point? (Just use some other t3 indi/lib/whatever for EA etc.)

On the other hand, maybe the value used for selecting color (t3 rate of change / ATR), (g variable in source code, a primitive trend/notrend filter) can be of interest to some? I could put that in one buffer & t3 value in an other. If someone need this just say, and Il fix it.

BTW: It seems that I can't edit my posts (can't find the edit button described in the FAQ: I guess because Im a new member Im not allowed, or Im just blind), so haven't fixed the mess with the buggy version i my first post.

 

edit posts, post attacements, links - could be antispam protection (new members <5?posts)

t3 = c1*p6 + c2*p5 + c3*p4 + c4*p3;

double g = (t3-t3)/iATR(NULL, 0, T3Period, i);

t3 and g - both of interest

+more reliable2 and convenient use same indi values in eas and trailors

 

Done. (Main t3 & Trend strength now in separate IndexBuffers) + RGBMode is no optional. (Also fixed a bug there the line would get shorter as time passed... And Improved the trend strength filter a bit.)

Hopefully final ver... (And no more bugs i hope )

Enjoy.

Files: