Author Topic: o34 Faulty  (Read 2044 times)

0 Members and 1 Guest are viewing this topic.

Peter

  • Guest
o34 Faulty
« on: May 20, 2011, 10:21:59 AM »
Deleted
« Last Edit: April 11, 2015, 09:12:08 AM by Peter »

Charles Pegge

  • Guest
Re: o34 Faulty
« Reply #1 on: May 20, 2011, 11:06:17 AM »

Hi Peter,

I tried all combinations of your examples and Alpha032, Alpha034 and today's build. They all performed the same with no apparent anomalies. If you show me source code with the fault I should be able to isolate the problem.

Today I fixed a bug related to converting bytes in a UDT into floats. This may show up if you are dividing pixel colors by a scaling factor.


Charles


PS: I have just posted today's Oxygen-progress, if you want to try it.

Charles Pegge

  • Guest
Re: o34 Faulty
« Reply #2 on: May 20, 2011, 08:52:49 PM »
Excellent Peter!

If you were using low level binding, it does not know about byval double or anything wider than 32 bits (on a 32 bit system)

All my Opengl stuff uses the Khronos headers in their original form.

Code: [Select]

WINGDIAPI void APIENTRY glOrtho (GLdouble left, GLdouble right, GLdouble bottom, GLdouble top, GLdouble zNear, GLdouble zFar);


Anyway, I have been checking over the numeric types to see if there are any hidden anomalies. So far I have found a problem with the hex function converting values > 8000000h.

Charles