升职加薪 发表于 2020-10-19 11:24

YoumuChan 发表于 2020-10-19 11:57

跟是不是垂直同步,有没有gsync/freesync有关

好多大象啊 发表于 2020-10-19 12:06

显示器说的那个是 G2G 灰阶延迟,准确的说就是显示器已经知道这帧要显示什么了,但是从上一帧变化到这一帧需要4ms(其实不止,灰阶延迟一般比黑白小)。
但是实际上每一帧都在延迟这样,所以帧和帧之间的速度不怎么影响,除非帧刷新速度超过灰阶变化速度,那么可能有一些显示模糊,不过没怎么见过这种面板。
灰阶延迟不需要周期性变化的,随着每一帧刷新延迟而已,实际上就是电压的变化,下一帧来了,直接换成下一帧的电压,没有自己的频率的

zmw_831110 发表于 2020-10-19 15:40

本帖最后由 zmw_831110 于 2020-10-19 16:29 编辑

来一帧驱动一次,屏幕变化就是刷新率.
屏幕要及时响应灰阶的变化,应该只要考虑全响应速度均快于刷新率就可以确保.

可以看看TFT的评测
https://www.tftcentral.co.uk/reviews/lg_27gl850.htm
灰阶0-255,255-0
各种跳变情况,所需的响应时间是不一定的.
比如27GL850的nano ips fast mode 在144Hz刷新率下的各个主要灰阶变化所需时间
http://ww1.sinaimg.cn/large/46a42788gy1gjuowe9wuwj20om0hhjsq.jpg

We therefore stuck with the 'Fast' setting and tested that across the range of refresh rates. At all refresh rates the average response time was basically the same at 4.9 - 5.0ms G2G. We will test a wider range of transitions in the next section to get a more accurate final figure. The only difference in performance came with the levels of overshoot. At 144Hz there was none recorded at all, but this increased slightly as the refresh rate lowered. At 100Hz it was pretty slight, and not really noticeable in practice at all. At 60Hz it was a little higher but still pretty hard to spot in real life to be honest. There were some very slight pale halos behind moving objects but they were not overly distracting and hard to pick out. The good news is that the response times were very fast across all measured transitions, and for all refresh rates here. They were fast enough also to keep up with the frame rate demands, even at the maximum 144Hz refresh rate which was good news.

下面是使用K7B面板里,最经典的型号AD27QDhttps://www.tftcentral.co.uk/reviews/gigabyte_aorus_ad27qd.htm
http://ww1.sinaimg.cn/large/46a42788gy1gjupgbln1jj20mw0dy0to.jpg即使是最高加速的speed模式,在XXX-255,存在超过6.94ms的情况.

此文,在Balance响应模式里提及过1/144,6.94ms
As the refresh rate increased, the overdrive impulse seems to be turned down slightly, resulting in slightly slower response times - but the removal of any slight overshoot at least. At 100Hz we measured 9.6ms G2G average which was still on the whole fast enough to keep up with the frame rate here. At 144Hz the response times dropped slightly again, now measured at 10.2ms and not really being fast enough in many cases to keep up with the high 144Hz frame rate where <6.94ms response times are needed. This resulted in a little added blurring on moving images, although to be fair it was not too bad in visual tests.

虽然我没用过这款,只用过acer的那款XV272U P,即使开启加速模式,看弹幕横向飘逸,还是能发现和nano ips的明显差异.
(弹幕是偏高灰阶的字幕平移,类似XXX->255的这种情况)

两者有那么大体验的差别,应该就是nano ips恰巧站上了所有情况,响应时间均小于1/144S这个槛,而K7B,并不够,所导致的.

http://ww1.sinaimg.cn/large/46a42788gy1gjuq2g55t8j20hv0i8wfn.jpg


页: [1]
查看完整版本: 显示器帧率的倒数ms和显示器响应时间的ms是不是一个基础上的不同数据?