For those of us whose rigs have less than perfect reliability, and who aren't always present to monitor them, the 128/256 second stats can tell us at a glance that our rigs are up and running. It's super handy to just check it on my phone when I'm at work or wherever.
This +1million.
My rigs are a bit wobbly at best, so knowing if they're dead or not is pretty important to me.
Changing the vardiff to double the diff of the share would be a good idea IMO.
If you are mining at diff 10 after it is settled in, you would now be mining at diff 20.
To keep the resolution intact on the stats- they time frame would need to be doubled also- so 128 seconds would now be 256 seconds.
This would reduce the demand on the DB server by half ~, cut bandwith usage and position the pool for future growth before this becomes a problem again.
This would not affect anyone, as long as they are not staring at a screen 24 hours a day every 128 seconds- now you would need to stare every 256 seconds to get the same result.
WK might have another reason, but I can not think of it- unless some devices time out that I don't know of.
The current setup, which has worked well for so long, just doesn't seem to be able to handle catching up to real time with such a large backlog (~48 hours).
