Timeline for Why are committed resources better than placed resources?
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 26, 2020 at 0:40 | vote | accept | Panda Pajama | ||
| Jul 17, 2020 at 18:22 | answer | added | macrod | timeline score: 5 | |
| Dec 7, 2018 at 10:53 | comment | added | Panda Pajama | @ChuckWalbourn: Thanks for the insight. I thought the entire point of Direct3D 12 was to leave less decisions to the driver so you can optimize for your use case, so I'm surprised that they would prefer to defer those decisions to the driver. | |
| Dec 4, 2018 at 18:17 | comment | added | Chuck Walbourn | The general answer is that the more flexibility you provide the driver, the better it can choose the right kind of memory based on the specific device. CreatePlacedResources is more proscriptive than CreateCommittedResource for technologies like nVidia TurboCache as well as dealing with video memory overcommittment. What actually happens is up to the driver writer here, which is basically what nVIDIA would prefer as noted in their article. In the end it's general advice, so you need to test on a variety of cards to see what actually works. | |
| Nov 18, 2018 at 15:29 | history | asked | Panda Pajama | CC BY-SA 4.0 |