benchmark testing for addons?
(Didn't know where to put this so I put it here)
One thing I'm always interested before purchasing a script is performance impact. Be it small, be it large I feel that it is very important to consider. When a person comes here they usually are here for more then 1 script. They are probably going to buy a hud, maybe an f4 menu, maybe more depending on what they want. say they buy 5 things, those things have relatively small impact to performance on their own but together it adds up. lets say they keep buying but they hit 1 addon that just drives their server to the ground. This has happened to me before.
Wouldn't it be nice if those who create a script were to do things such as netgraph 3 and fprofiler? maybe even more benchmark tool data being shown aswell such as idle CPU usage of the server with and without the addon and similar stuff. knowing these things before purchase can better prepare a server for dealing with them and help make a more informed decision.
It could also be seen as a goal for script makers. Make the impact lower and lower with each update to the script.
So this is basically what I would love to see more of in the future on scrips. I'd love for it to be a requirement actually. There are a good amount of scripts that I've ended up buying but never use because of this issue, and I feel that I'm not the only one who has this issue. Tell me what you think on this
Quite a while ago I had the idea of introducing a new measurement unit which tells you the performance impact on your server for certian addons.
But I have no clue if it's even possible to create and measure such a unit because there are too many initial conditions (player count, gamemode, server specs) which would matter.
Maybe someone knows if sth like that is possible.
hmmm, I had been thinking something like the people in charge of reviewing a script put it on a test server after code review and then having 5 other reviewers join and test it. would simulate 6 people total on a server, in a server environment. private so you know nobody would be using it without permission and could get a consistent result across all tests on scripts
By[RN] Vaas Kahn Grimat 30th of December 2017 16:38:24
hmmm, I had been thinking something like the people in charge of reviewing a script put it on a test server after code review and then having 5 other reviewers join and test it. would simulate 6 people total on a server, in a server environment. private so you know nobody would be using it without permission and could get a consistent result across all tests on scripts
Not only would this require way too much effort for such little payoff you're not keeping in mind all the variables. People have different addons they use that we don't even sell here, as well as custom content. Add in their different player counts and server hardware, as well as everything else under the sun - and two experiences are never the same.
aye, tho people may want some form of benchmark results before purchase. I've bought alot of things already that really caused more problems then benefits given by using. I have the $300 spent to prove it. I'm just loosly thinking here tho on the testing idea. personally I feel it should be required from the author but not real sure how to go about that honestly. kinda spit balling ideas while working