Chinese :
Subspace 为了保证获取奖励的公平性和足够的随机性，会通过一系列延迟算法(目前的算法时长在 900-1100 ms) 以内完成，作为一个时隙，时隙与时隙之间是串行实现的，所以你没法提前获取时隙，通过初始计时器引导网络内是所有成员，当然这个过程有一系列保证，防止因为超前的计算利用算力破解达到不需要存储即可完成证明，当计时员迭代完成后会像网络内广播时隙，也就是1秒钟广播一次，按照正常来说 1s 足以传播至全球任何一个地方了，但是国内可能是阻断的原因或者ISP 的封锁导致 即使过去 4秒你甚至都无法接收到来自其他计时器的时隙，这就导致了你收到的插槽延迟太大了，当你收到的时候区块已经归档了。你自然就没法在这个区块内获取奖励了。
很多时候其实国内发生的不只是网络延迟问题，比如国内经常会出现Node 不同步了(通常发生一会之后他又自动同步上了，有时候甚至都没感觉但是确实是不同步了慢了1个区块或者更多)、甚至是出现节点永远追不上块，这种情况下你运行计时器意义是不大的。
不知道为什么官方取消了关于节点同步状态的事件通知，我相信早期用户经常会看到P盘的时候就是P一会报错说节点不同步然后又开始过一会又停止了然后又开始了，有时候可能很快过几秒就同步上了，有时候也可能很慢过几分钟甚至更长才能同步上，虽然现在取消了这个的显示，但是问题仍然是存在的。
插槽超时不一定就是Node的问题!!
插槽超时不一定就是Node的问题!!
插槽超时不一定就是Node的问题!!
个人认为开启计时器不如把Node运行在非中国境内 
运行计时员的方法:
在node cli 的启动命令中追加
示例 :
./Node
...
--timekeeper #开启计时员
--timekeeper-cpu-cores #讲计时器绑定到指定的线程上
English :
Subspace, to ensure fairness and sufficient randomness in reward acquisition, uses a series of delay algorithms (currently, the algorithm’s duration is within 900-1100 ms) to create time slots. These slots are implemented serially, meaning you can’t access a slot in advance. The network guides all members with an initial timer. This process includes several safeguards to prevent using computing power to crack the system and prove without needing to store data. After the timer iteration is complete, it broadcasts the time slot to the network, effectively every second. Normally, 1 second is enough to transmit anywhere globally, but due to blockages or ISP restrictions in some countries, you might not receive the time slot from other timers even after 4 seconds. This results in receiving the slot too late, by which time the block has already been archived, and you naturally can’t earn rewards within that block.
Often, the issues in certain countries are not just network delays. For instance, nodes often become unsynchronized (usually, they resynchronize after a while, sometimes without even being noticed but indeed they lag by one block or more), or nodes may never catch up with the block, rendering running a timer somewhat pointless.
It’s unclear why official notifications about node synchronization status were discontinued. Early users might remember seeing errors about unsynchronization on the Plotting disk, then it would stop and start again after a while. Sometimes synchronization happens quickly, in just a few seconds, and sometimes it can take several minutes or longer. Even though this notification feature was removed, the problem still exists.