linux下提高nVidia显卡性能

1、使用新的 Nvidia 驱动(有多新用多新的),archlinux下可以使用aur自己打包安装最新的190.18,
官方修正了190.16的黑屏问题。

2、修改 xorg.conf
Option "PixmapCacheSize" "1000000"
Option "AllowSHMPixmaps" "0"

3、修改启动配置文件
# nvidia-settings -a InitialPixmapPlacement=2 -a GlyphCache=1

下面是另一个文章

上面也列出了一些 8000 之后显卡支持的 Option 选项，可以尝试打开后看看效果：

Options that are said to work well on 8xxx cards but are untested (by me)

Option "RenderAccel" "True"
enabled by default
Option "TripleBuffer" "True"
Enables triple buffering. "Decreases the time an application stalls while waiting for vblank events, but increases latency slightly" (NVIDIA Readme)
Option "DamageEvents" "True"
Recommended by NVIDIA if running composite glx, increases performance, enabled by default
Option "UseCompositeWrapper" "True"
Enables the X server's composite wrapper instead of the builtin one.
Option "AllowIndirectPixmaps" "True"
Could improve hardware rendering on G80 cards with more than 256 MB of video memory.
Option "BackingStore" "True"
Cache overlayed areas in case they get redisplayed later
Option "PixmapCacheSize" "200000"
allocate said number of pixels for pixmap caches

除此之外，177.67 还给我们带来了一个新的设置参数： AllowSHMPixmaps，通过在 Device Section 增加：

Option "AllowSHMPixmaps" "0"

可能会带来一些性能改善，具体的效果可以自己尝试一下，官方的说法：

Added an 'AllowSHMPixmaps' X configuration option, which can be used to prevent applications from using shared memory pixmaps; the latter may cause some optimizations in the NVIDIA X driver to be disabled.

之后 177.68 驱动，改善了 KDE4 下 OpenGL 的性能，官方的说法：

Improved GPU video memory management coordination between the NVIDIA X driver and OpenGL implementation; this should improve performance with e.g. the KDE4 OpenGL compositing manager.