• Content count

  • Joined

  • Last visited

Community Reputation

122 Neutral

About BlueWanderer

  • Rank
  1. I think one resource cannot be input and output at the same time. [quote]If an overlapping resource view is already bound to an output slot, such as a render target, then the method will fill the destination shader resource slot with NULL.[/quote] Guess you will need put the result in the shared memory and do the filtering yourself.
  2. You can render to the back-buffer. It is the back-buffer be stretched to the front-buffer if the back-buffer and the front-buffer are not compatible. If they are compatible, the back-buffer becomes the front-buffer, and the ex-front-buffer is put into the swapchain.
  3. What will happen when shaders meet textures?

    Let's put it another way. My card has 16 x 9 SIMD units and 4 x 9 texture mapping units (and I wonder if the 2 "x 9"s have same meaning). Can I say that one SIMD group have 4 texture mapping units? If so, when the group is executing one sampling instruction, how the texture mapping units are used? If not, how will they do the job?
  4. As it seems to me, when there are around one thousand shader units, we can only expect around one hundred or less texture units. So when the shaders are sampling textures, will they be stalled?
  5. Can I call the Present method of different swapchains of a single adapter each in a separate thread? I have 2 monitors attached to one card and 2 swapchains for each monitor. Without triple buffering, call Present sequentially would be silly. And with triple buffering, I will have to add one more staging buffer for each stream content (and I have a lot of them). ---- I found that I didn't know when Present would wait... Will Present be blocked when there is no free buffer in swapchain, or will it be blocked when number of queued frames have reached the maximum?