Site statistics

Skeltom

Engaged Member
Oct 9, 2017
2,416
2,823
I just had a look at the bottom of the latest update page and if the site stats are correct it doesn't get much more well rounded.

Stat.png
 

anne O'nymous

I'm not grumpy, I'm just coded that way.
Modder
Donor
Respected User
Jun 10, 2017
10,409
15,316
I see. Seems pretty dumb to put a temp statistic for no reason.
Well, as the page link say itself, by default you get the alpha version. Therefore yes, it make sense to put a temp statistic. It permit to have a better view of the change in the execution time of the page, by removing as much unrelated processing as possible.
 

Skeltom

Engaged Member
Oct 9, 2017
2,416
2,823
Well, as the page link say itself, by default you get the alpha version. Therefore yes, it make sense to put a temp statistic. It permit to have a better view of the change in the execution time of the page, by removing as much unrelated processing as possible.
Yeah, I guess it makes sense. But we are talking about a difference so negligible it hardly matters. Unless you are viewing the page, or any other page for that matter, on a cellphone from the 90s with a network connection to match its probably only going to be 3 seconds max and that's being generous. They already have it implemented into rest of the site as it is. But apparently the old page for new releases was having performance issues? Something I never even noticed personally. At the end of the day it's just a counter at the bottom of the page that few here even bother to look at I'd wager.
 

anne O'nymous

I'm not grumpy, I'm just coded that way.
Modder
Donor
Respected User
Jun 10, 2017
10,409
15,316
But we are talking about a difference so negligible it hardly matters. Unless you are viewing the page, or any other page for that matter, on a cellphone from the 90s with a network connection to match its probably only going to be 3 seconds max and that's being generous.
I'm pretty sure that I talked about the "execution time", not the "loading time".

It's not that they don't care how much time it will need for someone to see the page, but what matter is the charge implied by the page. There's more than 4 millions members, for an average around 15.000 simultaneous connections, more than 6 millions posts, without counting the private messages, and, more relevant to this page, more than 10 000 threads that can possibly be updated. Of course, processing the stats is near to atomic, and don't imply a high rise in charge for the servers. But when you works on an architecture with such size, you want to know the exact numbers, not extrapolate what they could be "without this". And you also want to be sure that a sudden slowness effectively come from the process, and not for something external.
Even with the caches, there's a high charge on the servers, and I totally understand why Sam and co want to know precisely how a page, that is probably one of the most requested one, insert itself in all this, and what level of stress it can add.