Page 1 of 1
Active users
Posted: Sun Aug 04, 2013 11:38 am
by billford
When I visit my stats page I get something like:
Donor Rank: xxxxx of 1699341 (as of posting this)
But according to the F@H front page there are only 282,750 computers in use (same qualifier), and I would assume that the number of active usernames is significantly less than this.
(The reasons for the discrepancy seems obvious enough- many (most?) have run a couple of work units then given up for whatever reason.)
Would it not make more sense for the donor rank to be given in terms of the number of active users, perhaps those who have returned a work unit within the last 50 days (to parallel the number of active clients on a username)?
Re: Active users
Posted: Sun Aug 04, 2013 1:18 pm
by ChristianVirtual
On the other side there are a number of now inactive user achieved quite some points in the past. Those can't and should not be ignored in a ranking system. Myself I'm looking only for ranking within my team; I don't care too much about the overall ranking.
Re: Active users
Posted: Sun Aug 04, 2013 2:08 pm
by billford
ChristianVirtual wrote:On the other side there are a number of now inactive user achieved quite some points in the past. Those can't and should not be ignored in a ranking system.
I think that's arguable, though I can see your point. But I think something should be done to exclude those who, judging by some of the usernames, just joined for a laugh one night after too many beers.
Myself I'm looking only for ranking within my team; I don't care too much about the overall ranking.
I'm solo not in a team, I do it partly out of interest and partly because it pricked my conscience (bent and battered though it is) to have a reasonably powerful machine with 90% of it's power lying around doing nothing.
Re: Active users
Posted: Sun Aug 04, 2013 2:20 pm
by Joe_H
In addition to inactive users having earned points in the past, not all inactive users stay that way. Some fold only during cooler weather or while at school. Others take a break from F@H to contribute to other distributed computing projects they are interested in and come back after a bit. Some persons have folded under a variety of names and continue to fold under a current one. So I can see there is some sense in what you suggest, but it would make for a complicated change in the way stats are reported by PG.
This feature though is something one of the third-party F@H stats sites could or already provide. For instance EOC does have filtering on team stats to just show active folders. Kakao stats has its filters, though I don't use that site often enough to know all of the available ones. There are other site that follow the folding stats, perhaps one of them does ranking that way.
Re: Active users
Posted: Sun Aug 04, 2013 4:02 pm
by mdk777
But I think something should be done to exclude those who, judging by some of the usernames, just joined for a laugh one night after too many beers.
But look at it this way; those are people you will blow by
Nothing more satisfying than moving up the rankings at a fast clip.
Once you get to the top 10K in the rankings, your progress will have slow down considerably...but if you are addicted to competitive progression in the rankings...you just have to increase your effort.
Easy and gratuitous confirmation early, more realistic after you have settled in; sounds like the perfect progression.
Re: Active users
Posted: Sun Aug 04, 2013 9:11 pm
by bruce
The official stats system is a database that reports names/teams/points. It's not directly connected to the activity state of each Donor. Stats systems developed by third parties keeps track of previous stats updates, compares where people are today (points-wise) with where they were some time back and computes whether to call you active or not.
How much of Stanford's compute processing should be devoted to science and how much should be diverted to duplicate the processing which has already been programmed by 3rd parties? (HINT: The government grants which fund FAH are generally restricted to doing science.)
I'm not suggesting that your it is a bad idea...just that it's probably not going to be something Stanford is going to do.
Re: Active users
Posted: Mon Aug 05, 2013 5:57 am
by billford
mdk777 wrote:
Nothing more satisfying than moving up the rankings at a fast clip.
Yes, I can't argue with that. Once I'd been folding for a while I started keeping track of how many places I gained for each 1,000 points. It started at well over 300, now it's down to about 4
And occasionally going negative as people overtake me
Once you get to the top 10K in the rankings, your progress will have slow down considerably...but if you are addicted to competitive progression in the rankings...you just have to increase your effort.
See above
At some point I realise I'll catch up with those who are gaining points at the same rate as me and I'll just bob up and down a bit, going nowhere. Then I'll have to find a way to increase my available computing power without spending too much money, retirement has its downsides
Re: Active users
Posted: Mon Aug 05, 2013 6:05 am
by billford
Joe_H wrote:There are other site that follow the folding stats, perhaps one of them does ranking that way.
That's a very fair point, I haven't investigated those. An oversight I need to correct.
Thanks to all for the replies, they've given me some extra perspectives. Bruce's second paragraph in particular is something I hadn't considered- budgetary constraints in the US and the UK differ markedly!