I was waiting for someone to eventually post the right answer but it seems ns2 is too unintuitive and convoluted for that to happen rofl
Basically this - If you are building an armoury
It doesn't matter If you are building an arms lab
use a welder If you are building a powernode blueprint
Do not use welder, it is slower If you are building a destroyed powernode
Do not even bother if you don't have a welder. Welder is twice the speed from something like 30 seconds base to 15. If you are building an IP
Use welder If you are building a robotics
It doesn't matter If you are bulding an RT
It doesn't matter
Just curious when did you test last? I showed this post to a dev and with some things he got same results while others he didn't (blueprint power node was faster with welder by 1.49 seconds). I'm not at home to test myself or I would .
I'll try to test this myself this weekend and report it as a bug with actual times for the devs to look at. It would make sense for all buildings to behave the same I would assume. If you ever find stuff like this in ns2 that seems to be a bug message it to the playtest leads (narfwak, explosif).
Many builds ago, not sure if this is accurate any more, Welders built structure armor faster than structure health. So the build rate differs based on the ratio of health to armor. I am not sure if structures even differentiate that anymore. I know they still have two health bars when attacking them, I just don't know if there is a functional difference anymore.
I was waiting for someone to eventually post the right answer but it seems ns2 is too unintuitive and convoluted for that to happen rofl
Basically this - If you are building an armoury
It doesn't matter If you are building an arms lab
use a welder If you are building a powernode blueprint
Do not use welder, it is slower If you are building a destroyed powernode
Do not even bother if you don't have a welder. Welder is twice the speed from something like 30 seconds base to 15. If you are building an IP
Use welder If you are building a robotics
It doesn't matter If you are bulding an RT
It doesn't matter
Just curious when did you test last? I showed this post to a dev and with some things he got same results while others he didn't (blueprint power node was faster with welder by 1.49 seconds). I'm not at home to test myself or I would .
I'll try to test this myself this weekend and report it as a bug with actual times for the devs to look at. It would make sense for all buildings to behave the same I would assume. If you ever find stuff like this in ns2 that seems to be a bug message it to the playtest leads (narfwak, explosif).
I would say the dev either forgot how to tell the difference between his welder and build tool, didn't measure the start and end frames properly, or that ns2 is just buggy as usual and highly variable when building.
I did a test just then to confirm i remembered right
Build time is measured as start: key press end: first frame powernode becomes solid from blueprint]
Giving me
1) 9.5 seconds without a welder
2) 12 seconds with a welder
Here is the video proof that i am not talking out of my ass
So you don't have to magic download it off youtube to see the build times (although you can if you want), note that building without welder starts at 0 sec frame, and building with welder starts at 10 sec frame. So you can see the welder build time easily as 22 - 10 straight off youtube.
I would say the dev either forgot how to tell the difference between his welder and build tool, didn't measure the start and end frames properly, or that ns2 is just buggy as usual and highly variable when building.
I had tested it on the internal build, which it turned out was slightly out of date. Having said that, I'm getting about 9 seconds for both the welder and build tool (on public build, with cheats).
Many builds ago, not sure if this is accurate any more, Welders built structure armor faster than structure health. So the build rate differs based on the ratio of health to armor. I am not sure if structures even differentiate that anymore. I know they still have two health bars when attacking them, I just don't know if there is a functional difference anymore.
I suggest taking a look at the Damage Types page for more information on Health & Armor interaction (yes, the page is up-to-date).
I had tested it on the internal build, which it turned out was slightly out of date. Having said that, I'm getting about 9 seconds for both the welder and build tool (on public build, with cheats).
I mean, elodea went to the effort of posting a video that shows the timings. What more can he say? Try measuring more accurately?
I had tested it on the internal build, which it turned out was slightly out of date. Having said that, I'm getting about 9 seconds for both the welder and build tool (on public build, with cheats).
I mean, elodea went to the effort of posting a video that shows the timings. What more can he say? Try measuring more accurately?
I'm not denying the evidence he presented, I'm just explaining what current1y said in his post (and what I saw). There must have been another factor in play, I wasn't 2 seconds out in my measurement. As I said, it might be because I used cheats, or because I tested it on a local server.
I always wondered if this is even true. People always scream at others, "Use the welders, they are faster!!!!111", but i never got the feeling they are any different. It took just the same (too) long.
I always wondered if this is even true. People always scream at others, "Use the welders, they are faster!!!!111", but i never got the feeling they are any different. It took just the same (too) long.
...Having said that, I'm getting about 9 seconds for both the welder and build tool (on public build, with cheats).
I can confirm the building time of 9 seconds with building tool and welder. I was testing this in a vanilla ns 2 with bots (0 marine, 2 alien), no cheats enabled.
I retried my measurements which confirmed my last statement, but i added Arms Lab, IP, Armory and PG. So here are my time savings for each of mentioned buildings:
How accurate is this? Consider a reaction time of plusminus 0.3 seconds for both start and stop events, so plusminus 0.6s for each value.
My conclusion:
PG, Armory, Observatory, RT, blueprint power-node, IP, doesn't make much of a difference. A Welder builds slightly faster, nonetheless.
Building an arms lab: Welder may save you about 2 seconds.
If anybody is doing testing on a local server, please also confirm the results from a real server. My last experience with the local server was over a year ago but it was massively different to the real thing.
Here are my tests in local without cheats, I got sensibly the same results on a distant server. It's a bit long, I wanted to make it faster but that would have ruined it for the people who want the real footage.
The original footage is recorded at 60FPS and I started/stopped the timer each time on the specific frame for consistent result, and to be the most precise possible:
- when the building appears,
- when the weapon switch,
- X frames before the deploying animation begin.
The time for the Build tool is on the left, while the ones for the Welder is on the right.
For those who don't want to wait, I have the direct links to the times in the description of the video:
Power nodes: 1:07
RTs: 2:27
Obs, Arms lab, CC: 4:32
Result: the Welder always win, sometimes it's not noticeable, but sometimes it is.
It also needs more iterations for the last buildings, but it already took me quite some time to do that so... I let that to the others
One thing that come to my mind is the framerate. What if the fire rate of the welder was limited by your framerate like it was for the rifle at one time?
If you have low FPS you build slowly with a welder, while it doesn't change anything if you use the build tool.
The lower I got while recording was around 75FPS so it is not really a problem for me. How much do you get elodea? I don't think the latency could have any effect, but who knows? Packet loss on the contrary could have a huge effect I think.
Comments
Just curious when did you test last? I showed this post to a dev and with some things he got same results while others he didn't (blueprint power node was faster with welder by 1.49 seconds). I'm not at home to test myself or I would
I'll try to test this myself this weekend and report it as a bug with actual times for the devs to look at. It would make sense for all buildings to behave the same I would assume. If you ever find stuff like this in ns2 that seems to be a bug message it to the playtest leads (narfwak, explosif).
I did a test just then to confirm i remembered right
Build time is measured as
start: key press
end: first frame powernode becomes solid from blueprint]
Giving me
1) 9.5 seconds without a welder
2) 12 seconds with a welder
Here is the video proof that i am not talking out of my ass
So you don't have to magic download it off youtube to see the build times (although you can if you want), note that building without welder starts at 0 sec frame, and building with welder starts at 10 sec frame. So you can see the welder build time easily as 22 - 10 straight off youtube.
I had tested it on the internal build, which it turned out was slightly out of date. Having said that, I'm getting about 9 seconds for both the welder and build tool (on public build, with cheats).
I suggest taking a look at the Damage Types page for more information on Health & Armor interaction (yes, the page is up-to-date).
I'm not denying the evidence he presented, I'm just explaining what current1y said in his post (and what I saw). There must have been another factor in play, I wasn't 2 seconds out in my measurement. As I said, it might be because I used cheats, or because I tested it on a local server.
I can confirm the building time of 9 seconds with building tool and welder. I was testing this in a vanilla ns 2 with bots (0 marine, 2 alien), no cheats enabled.
I retried my measurements which confirmed my last statement, but i added Arms Lab, IP, Armory and PG. So here are my time savings for each of mentioned buildings:
Arms Lab: 2 seconds
IP: 0.5 seconds
Armory, PG: 1 second
How accurate is this? Consider a reaction time of plusminus 0.3 seconds for both start and stop events, so plusminus 0.6s for each value.
My conclusion:
PG, Armory, Observatory, RT, blueprint power-node, IP, doesn't make much of a difference. A Welder builds slightly faster, nonetheless.
Building an arms lab: Welder may save you about 2 seconds.
EDIT: Though latency could some how affect it.
The original footage is recorded at 60FPS and I started/stopped the timer each time on the specific frame for consistent result, and to be the most precise possible:
- when the building appears,
- when the weapon switch,
- X frames before the deploying animation begin.
https://www.youtube.com/watch?feature=player_detailpage&v=YOzm24NTdfs#t=68
The time for the Build tool is on the left, while the ones for the Welder is on the right.
For those who don't want to wait, I have the direct links to the times in the description of the video:
Power nodes: 1:07
RTs: 2:27
Obs, Arms lab, CC: 4:32
Result: the Welder always win, sometimes it's not noticeable, but sometimes it is.
It also needs more iterations for the last buildings, but it already took me quite some time to do that so... I let that to the others
One thing that come to my mind is the framerate. What if the fire rate of the welder was limited by your framerate like it was for the rifle at one time?
If you have low FPS you build slowly with a welder, while it doesn't change anything if you use the build tool.
The lower I got while recording was around 75FPS so it is not really a problem for me. How much do you get elodea? I don't think the latency could have any effect, but who knows? Packet loss on the contrary could have a huge effect I think.