You know when you stare at some math stuff and you don't understand why it's working..

ah, I think I have it.

What'cha working on?

The correct width for a pixel buffer to send to windows via GDI.

It requires that the line be a multiple of LONG, but the code I came up with "works" but does not make sense.

  int width = updateRect.right - updateRect.left;
  int widthForBufferPadding = width % sizeof( LONG ); // Lines must be "multiple of LONG"
  int pixelBufferSize = ( 3 * width + widthForBufferPadding ) * height;
  std::_tout << "widthForBufferPadding " << widthForBufferPadding << std::endl;
  std::_tout << "sizeof( LONG ) " << sizeof( LONG ) << std::endl;
  std::_tout << "( 3 * width + widthForBufferPadding ) % sizeof( LONG ) " << ( 3 * width + widthForBufferPadding ) % sizeof( LONG ) << std::endl;

This always outputs ( 3 * width + widthForBufferPadding ) % sizeof( LONG ) 0.

But the line int widthForBufferPadding = width % sizeof( LONG ); seems wrong.

...yeah, I'm at least 1.5 glasses of wine too far in to understand that in any kind of depth...

But I think I get why.

Ahaha :P

Gin-tonic here.

But still on the first, and it has not hit yet.

Okay, it's fixed

The right line is:

int widthForBufferPadding = ((3*width) % sizeof( LONG )) == 0 ? 0 : sizeof( LONG ) - ((3*width) % sizeof( LONG )); // Lines must be "multiple of LONG"

I'll take your word for it. :)

I appreciate the vote of confidence :P

Here's the math that's breaking my brain atm. 😜

Yes, given the "too much wine", I would have expected that those curves would have been more... curvy? :P

I need to be more sober to teach Excel how to dance.

A friend of mine who has a PhD in math said he could not do math while drunk.

I can't say I can do math. But I can throw numerical eggs at a wall and copy-and-paste the graphs I get.

:P

Yeah, when I try to credit a formula to "Horizon Zero Darn", it's time to step away from the keyboard and go to bed.

Goodnight!

'Night!

@Tyyppi_77 and 'morning ;)

Funny to see people wake up that early on purpose :P

I totally understand that.

I used to leave for work around 5h45 in the morning.

Leaving work around 15h30.

Can't do that anymore because of kids.

Well... I could wake up at 5h45 but they would wake up too and I'd have to take care of them until we leave for daycare, so nothing gained there :P

Would save me from being waken up by a jolt of adrenaline.

@Vaillancourt perfect place NOT to use a ternary

ms teams is trash, fight me

@Tyyppi_77 what working hours do you have? At the last company I worked we weren't allowed to start before 0630, so leaving before 11530/1600 usually was difficult if you still wanted to get your hours

@Tyyppi_77 wow, no need to take lunchbreaks?

we're required by law to have at least 30min when working < 9hrs and at least 60min > 9 hrs

holy shit, what country do you live? i need to relocate

aaah, the dark north. with the language that shares roots with hungarian :D

@Almo yeah, this will be fixed during the code review phase. Meanwhile, it allows me to complain about how trash code this new feature I'm working on is :p

:D

If I had my way, ternaries wouldn't exist without a special compiler flag.

they strike me as an old (1970s) thing programmers wanted just to avoid typing.

Yep

Although I agree that ternaries could make the code harder to read, such as in the code avobe.

@Vaillancourt that line has so many issues other than the ternary operator....

// Lines must be multiple of LONG
int rest = ((3 * width) % sizeof( LONG ));
int widthForBufferPadding = rest == 0 ? 0 : sizeof( LONG ) - rest;

Indeed ;)

what does the 3 stand for?

int widthForBufferPadding = 0;
if ( rest > 0 )
  widthForBufferPadding = sizeof( LONG ) - rest;

3 bytes per pixel (R, G, B)

ah

and then you try fitting the bytesequence into a sequence of LONGs :)

int widthForPadding = sizeof( LONG) - ((3 * sizeof( BYTE ) * width) % sizeof( LONG );

ah no, that doesn't handle the case of 0 properly

ugh

still would probably write 3 * sizeof( BYTE) * width, thus making the number less magical; the sizeof instruction will be resolved at compiletime anyway

@dot_Sp0T it's only the padding at the end of the line as required by the Bitmap format!

I use 24 bit per pixel

well that is a lot of data

or not

it's 1am here^^

Why a lot of data? 255 values for r, g, and b. The classical values in windows?

see, it's not much at all, i just misread bit to byte

:P

either the font i'm using is completely broken, or I am using it wrong... considering that my code is working perfectly for all other fonts i've tried so far I like to think it's the first -.-