Re: double dValue = 10.52 - 10.0; - C / C++ / MFC Discussion Boards

Vikash Dubey17-Nov-04 6:37

17-Nov-04 6:37

I am trying to implement the overlay blend mode as in photoshop. I have already implemented some blend modes(screen , multiply etc). Ovrlay blend mode is a combination of Multiply and screen blends depending on the base color this i found out. But when should i apply multiply and when should i apply Screen. What are the color values of the base, Can Anybody give me the info.

Thanks, Vikash

Re: Graphics blend modes

Jerry Evans17-Nov-04 7:39

Jerry Evans

17-Nov-04 7:39

have a look at the filters section in the SVG spec - iirc this is described quite well and may be of help. http://www.w3.org/Graphics/SVG/

Re: Graphics blend modes

Vikash Dubey17-Nov-04 8:24

Vikash Dubey

17-Nov-04 8:24

Thanks I will try that but i am open to any more suggestions

OLD TAPI CONTROL

John R. Shaw17-Nov-04 6:34

John R. Shaw

17-Nov-04 6:34

I just received a call to debug a TAPI problem.
The company that created the control no longer exists.

Problem: When the VB App finished making a call and downloading via modem, it closes the program without giving a reason. This only occurs on machines running Win2000 or WinXP. The program was origanaly written and tested on Win2000 (also tested on Win3.1, Win95 and Win98).

Could this have something to do with Windows security updates?

Any ideas would be helpfull.

NOTE: I already left this same message on the VB forum.

INTP

double dValue = 10.52 - 10.0;

Erick 'El Matador' Demers17-Nov-04 6:07

Erick 'El Matador' Demers

17-Nov-04 6:07

Try this in VC++

double dValue = 10.52 - 10.0;

Normally dValue should be 0.52 but the result is always: 0.51999999999999957

Why ?????

it cause me a lot of trouble, any idea on how to fix it ?

Re: double dValue = 10.52 - 10.0;

Maximilien17-Nov-04 6:20

Maximilien

17-Nov-04 6:20

because of precision computing.

it's standard "behaviour" when representing floating point numbers in binary.

Maximilien Lincourt
Your Head A Splode - Strong Bad

Re: double dValue = 10.52 - 10.0;

Chris Meech17-Nov-04 6:25

Chris Meech

17-Nov-04 6:25

Try this using pen and paper. It's called Long Division. Divide the value 1 by the value 3. When you get the exact answer and there is no remainder, please come back. Big Grin | :-D

If you do some Google searching for floating point approximations, you'll probably find your answer. Binary representations of a rational number requires approximations just as base ten representations of rational numbers requires approximations. Most times these approximations are very small and have only a minor or neglible effect on calculations, but as you are probably discovering, there are times that the effect can be appreciable.

Chris Meech

I am Canadian. [heard in a local bar]

Gently arching his fishing rod back he moves the tip forward in a gentle arch releasing the line.... kersplunk [Doug Goulden]

Re: double dValue = 10.52 - 10.0;

Phil J Pearson17-Nov-04 6:27

Phil J Pearson

17-Nov-04 6:27

It's inevitable. It is impossible to represent some real values exactly in floating-point format.

The best way to handle rounding errors like this is to decide what precision is necessary for your application and round values when necessary. Keep the values 'raw' as long as possible and round before displaying the result.

for example:

double round(double val, int ndp)
  {
  double mult = pow(10, ndp);

  val *= mult;
  return ((val - floor(val) < 0.5) ? floor(val) : ceil(val)) / mult;
  }

...

// dValue will be 0.5200
double dValue = round(10.52 - 10.0, 4);

The opinions expressed in this communication do not necessarily represent those of the author (especially if you find them impolite, discourteous or inflammatory).

Re: double dValue = 10.52 - 10.0;

Erick 'El Matador' Demers17-Nov-04 6:45

Erick 'El Matador' Demers

17-Nov-04 6:45

Thanks for your help.

With this simple Round function I was able to fix my problems.

Wink | ;-)

Re: double dValue = 10.52 - 10.0;

tom_dx17-Nov-04 13:35

tom_dx

17-Nov-04 13:35

why don't you just use int? or short int?

IM PROUD TO BE A GMAIL;

Re: double dValue = 10.52 - 10.0;

Michael Dunn17-Nov-04 15:49

Michael Dunn

17-Nov-04 15:49

As the other replies said, this is a hazard of floats/doubles. You should not test for equality, instead test that the difference is below some threshhold.

--Mike--
LINKS~! Ericahist | 1ClickPicGrabber | CP SearchBar v2.0.2 | C++ Forum FAQ | You Are Dumb

Control question

ricardo.padrela17-Nov-04 4:17

ricardo.padrela

17-Nov-04 4:17

What do i need to use this control: click here
Thanks!

Re: Control question

Phil J Pearson17-Nov-04 4:57

Phil J Pearson

17-Nov-04 4:57

A mouse.

The opinions expressed in this communication do not necessarily represent those of the author (especially if you find them impolite, discourteous or inflammatory).

Re: Control question

ricardo.padrela17-Nov-04 5:42

ricardo.padrela

17-Nov-04 5:42

Shame on you

Re: Control question

Chris Meech17-Nov-04 6:28

Chris Meech

17-Nov-04 6:28

LOL. When I saw the question, I was wondering about a smartass reply, but could come up with one. Yours is pretty good. Big Grin | :-D

Chris Meech

I am Canadian. [heard in a local bar]

Gently arching his fishing rod back he moves the tip forward in a gentle arch releasing the line.... kersplunk [Doug Goulden]

Re: Control question

Chris Meech17-Nov-04 6:30

Chris Meech

17-Nov-04 6:30

You have asked a question that is just so vague, it's impossible to answer. How have you tried to use it, whatever it is.

Chris Meech

I am Canadian. [heard in a local bar]

Gently arching his fishing rod back he moves the tip forward in a gentle arch releasing the line.... kersplunk [Doug Goulden]

Re: Control question

ricardo.padrela17-Nov-04 6:58

ricardo.padrela

17-Nov-04 6:58

What is the name of that control ? It is available through some MFC class or something ? I want to use that control but don't where to to start.

Thanks

Re: Control question

Chris Meech17-Nov-04 7:29

Chris Meech

17-Nov-04 7:29

Try using Spy++. That might help. What applications use it?

Chris Meech

I am Canadian. [heard in a local bar]

Gently arching his fishing rod back he moves the tip forward in a gentle arch releasing the line.... kersplunk [Doug Goulden]

Re: Control question

ricardo.padrela17-Nov-04 7:37

ricardo.padrela

17-Nov-04 7:37

I've seen it in visual studio, unreal editors and a couple more.

Spy++ gave me this class: WindowsForms10.Window.8.App1
I don't know what the hell is this.

Re: Control question

Chris Meech17-Nov-04 8:04

Chris Meech

17-Nov-04 8:04

PutaQuePariu wrote:
I've seen it in visual studio

Since I'm still using version 6, and don't see it, I'll assume it a new beast in 7. Smile | :)

I'm with Max on this one though. From the small pic you linked, it looks like a Tree Control that is ownerdrawn. Which basically means you have to 'reinvent' all that the owner draw implements.

Chris Meech

I am Canadian. [heard in a local bar]

Gently arching his fishing rod back he moves the tip forward in a gentle arch releasing the line.... kersplunk [Doug Goulden]

Re: Control question

Maximilien17-Nov-04 7:00

Maximilien

17-Nov-04 7:00

looks like an ownerdrawn tree control.

Maximilien Lincourt
Your Head A Splode - Strong Bad

Re: Control question

ricardo.padrela17-Nov-04 7:06

ricardo.padrela

17-Nov-04 7:06

I've seen it several times in different applications which makes me wonder that it is a standard one... or made by someone and freely distributed.

Re: Control question

John R. Shaw17-Nov-04 8:42

John R. Shaw

17-Nov-04 8:42

If you have seen it that much it is probably part of a frame work or suit of controls that several companies use (like Dundas or Prof-UIS).

It is simplier to buy interface componets the to creat them.

INTP

Re: Control question

Maximilien17-Nov-04 9:20

Maximilien

17-Nov-04 9:20

dundas offer a control like that in their Ultimate Toolkit package.

Maximilien Lincourt
Your Head A Splode - Strong Bad

Accused of function pointering

brianwelsch17-Nov-04 4:11

brianwelsch

17-Nov-04 4:11

My program is claiming the following runtime error:

Debug Error !<br />
Program : MyProgram.EXE<br />
Module :<br />
File : i386\chkesp.c<br />
Line : 42<br />
The value of ESP was not properly saved across a function call. This is usually a result of calling a function declared with one calling convention with a function pointer declared with a different calling convention.

I can ignore it and things appear OK, but that doesn't solve anything. The message is popping up when I return from my function. The thing is I don't use any function pointers so I'm not sure why I'm getting this. Any other funky reasons this error might show up?
The more frustrating part is haven't made changes to this section of code in months.
Thanks for any leads.

BW
The Biggest Loser

"Farm Donkey makes us laugh.
Farm Donkey hauls some ass."
-The Stoves

Use Ctrl+Left/Right to switch messages, Ctrl+Up/Down to switch threads, Ctrl+Shift+Left/Right to switch pages.