Jump to content

  • Log In with Google      Sign In   
  • Create Account


Member Since 06 Oct 2006
Offline Last Active Yesterday, 02:18 PM

Posts I've Made

In Topic: Problems reading texture data back

11 October 2016 - 10:15 PM

BYTE* bytes = (BYTE*)locked.pBits;
BYTE test = bytes[0];

Oh Man! Works perfectly.

As much as I try to understand pointers, they have a habit of biting me. Hopefully one day I'll get there. I really do try.

Thanks again for your help, it is truly appreciated.

In Topic: Problems reading texture data back

11 October 2016 - 10:02 PM

More importantly though, locked.pBits is the address the pixel data... and you're printing out the address of locked.pBits, which is the address of the address of the pixel data.

I think this is more the case of what is the source of the issue.

How would I get the actual data as opposed to the address of the data?

With and with out & gives the same symptoms (different value each time). If it were a simple format issue, i'd expect the results to remain constant.

In Topic: Basic multithread question

18 September 2016 - 02:53 AM

Nice! That looks like exactly what I am looking for!

Thanks heaps :)

In Topic: 8x FSAA vs 8x MSAA

29 August 2016 - 04:16 AM

Oddly enough, this might be a hardware/driver thing. I pushed the samples back to 4x and the image improved drastically.

In Topic: Depth buffer not working

16 August 2016 - 03:51 PM

hmmm...that's a trip.you're right. I don't like the way that looks, even though it's correct. Doesn't read well. 

HRESULT test_okay = S_OK; // == 0L
if (test_okay)
   MessageBox(0, L"not okay", L"fail", 0);
   MessageBox(0, L"okay", L"good", 0); // <---- this path is traveled 
Clearly I don't understand what I'm talking about.+1 for getting it working.Probably still should be using the FAILED or SUCCEEDED macro because we're looking for negative fail values,and as you said, zero on success.but if it's not broke... :)

Yeah, I see your point. I just looked up the macro and anything negative is a fail, so positive values will possibly give a false fail.