尝鲜试了下，似乎速度并没有什么提升 #61
Replies: 28 comments 5 replies
-
|
XTLS 无法突破线路瓶颈，建议了解这些协议的底层原理，测速建议 Speedtest |
Beta Was this translation helpful? Give feedback.
-
|
明白你的意思，目前对于大家都能跑满的情况下，其实用什么区别不大，特别是在家用软路由或者台式机软件的情况下。 |
Beta Was this translation helpful? Give feedback.
-
|
并不是都能跑满，之前很多人从硬路由到软路由，主要是因为前者性能不够，XTLS 可以改变这一状况 其中一些人是喜欢废物利用 DIY，但专门买软路由似乎没有什么必要了 |
Beta Was this translation helpful? Give feedback.
-
|
用两台VPS分别部署VLESS+XTLS和VMess+tls交叉测试，你就会有结果了 |
Beta Was this translation helpful? Give feedback.
-
|
我用的树莓派做单臂，600M联通可以跑满，不过似乎很多人不知道怎么玩树莓派 |
Beta Was this translation helpful? Give feedback.
-
|
XTLS在树莓派上对速度的提升是有目共睹的 尤其是在树莓派上运行服务端 |
Beta Was this translation helpful? Give feedback.
-
相同机房，相同配置，还要交叉测试，我觉得就是抬杠了。 |
Beta Was this translation helpful? Give feedback.
-
是的，对cpu占用要低。不过一般都是跑软路由，运行服务端干嘛 |
Beta Was this translation helpful? Give feedback.
-
|
对了，还有个比较麻烦的点，就是需要接管443，这样就没法caddy自动申请证书了。 |
Beta Was this translation helpful? Give feedback.
-
首先，测试时要排除其他因素干扰，从你的描述来看，你并没有对线路进行控制 或者说，被测试的两个对象都已经达到你的线路瓶颈 我仅仅是给你提供了一个相对更加合理的测试方案，并不是抬杠 |
Beta Was this translation helpful? Give feedback.
-
|
介绍里也是说性能上有提升，并没提到过会提升速度。速度快慢关键还是看线路 |
Beta Was this translation helpful? Give feedback.
-
|
youtube测速...... |
Beta Was this translation helpful? Give feedback.
-
|
youtube测试看不出来的，就算是这样，youtube也到不了20W，当然我的机器带不动8k加上2倍速。8k+1.25倍也就12W左右 |
Beta Was this translation helpful? Give feedback.
-
不知4b可否跑慢千兆？手头马上会有一个4b，近期搬家考虑新配网关，要是4b可以就不买新的了 |
Beta Was this translation helpful? Give feedback.
-
个人认为需要跑满的需求并不大，能流畅看4k就足够了。树莓派做单臂绰绰有余，我是3B+，你的4B应该更没问题了。 PS：youtube看4K的话，2W就够了 |
Beta Was this translation helpful? Give feedback.
-
我是个跑跑党-_-# |
Beta Was this translation helpful? Give feedback.
-
youtube的测速其实才是更为准确的，因为这才是你真正的使用场景。speedtest.net的原理是根据ip找到和你ip的地理位置最近的一个测试服务器进行连接测试 |
Beta Was this translation helpful? Give feedback.
-
以前我也是，后来发现毫无必要，现在主力vps都是便宜的，特别是俄罗斯那个，不限流量，一年才100出头，看4k足够足够。就网络只有200M |
Beta Was this translation helpful? Give feedback.
-
|
这里指的是当你要比较协议的客观表现时，Speedtest 是一个更好的选择，因为 Youtube 会受到加载及缓存策略的影响，浮动较大 |
Beta Was this translation helpful? Give feedback.
-
|
我只能说用youtube测速的都是辣鸡 |
Beta Was this translation helpful? Give feedback.
-
能透露下是哪家俄罗斯的吗，我的刚好快到期了，正犹豫买哪家的 |
Beta Was this translation helpful? Give feedback.
-
便宜、不限流量，带宽标的还可以
便宜、不限流量，带宽标的还可以 |
Beta Was this translation helpful? Give feedback.
-
它找的是vps附近的进行测试 |
Beta Was this translation helpful? Give feedback.
-
北方联通可以试试，其他地方我不清楚，带宽200，一般speedtest跑出来是180的样子 @shenlijun |
Beta Was this translation helpful? Give feedback.
-
|
这个要看高峰期的延迟带宽和丢包率 |
Beta Was this translation helpful? Give feedback.
-
是的，不过这家对联通较友好，一直在用。 |
Beta Was this translation helpful? Give feedback.
-
|
@RPRX 是的，对协议来说，应该采用speedtest，不过考虑到实际场景，就选择了看油管的速度 @2444989513 兄弟，我不想跟你杠，但是你在科学上网的时候是只去跑speedtest还是说去看视频？ |
Beta Was this translation helpful? Give feedback.
-
典型的“我用不到所以没用”，拿一个受一堆其他因素制约的方案得到一个没意义的所谓“差不多”的结论，不如看看人家是如何测试的 |
Beta Was this translation helpful? Give feedback.
-
Beta Was this translation helpful? Give feedback.
-
|
哈哈，和谐交流 （大雾 |
Beta Was this translation helpful? Give feedback.
-
|
他的原话：“相同机房，相同配置，还要交叉测试，我觉得就是抬杠了。” 我也是很无语 |
Beta Was this translation helpful? Give feedback.
-
|
还有“不过似乎很多人不知道怎么玩树莓派”，呃…… |
Beta Was this translation helpful? Give feedback.
-
xray+xtls+tcp的配置，同一机房，相同的两台vps
一台是docker跑的caddy+v2ray+ws，另一台是ProxySU安装的vless+xtls(direct)+tcp
看youtube的时候两者速度相差无几，甚至ws的还要快一点
个人感觉xtls的优势在于软路由cpu占用略低5%，但速度上并无太大差别
Beta Was this translation helpful? Give feedback.
All reactions