|
Author |
Thread Statistics | Show CCP posts - 9 post(s) |
|
CCP Zymurgist
Gallente C C P
|
Posted - 2010.11.16 16:54:00 -
[1]
CCP Veritas has been investigating CPU usage for Tranquility and has some great info to share. Read all about CPU utilization here.
Zymurgist Community Representative CCP Hf, EVE Online Contact Us |
|
|
CCP Veritas
|
Posted - 2010.11.16 17:09:00 -
[2]
Originally by: Ifly Uwalk Is the green line the number of Drakes killed, and the blue spikes CPU usage?
Blue is the raw data, green is a trendline.
|
|
|
CCP Explorer
|
Posted - 2010.11.16 17:19:00 -
[3]
Originally by: Ifly Uwalk Yay for tech-****. I don't understand most of it but like to pretend I do!
Second graph returns a 404 error btw. Is the green line the number of Drakes killed, and the blue spikes CPU usage?
Thanks a lot for the dev blog!
Large graphs have been fixed. Click each graph for awesome larger versions.
Erlendur S. Thorsteinsson Software Director EVE Online, CCP Games |
|
|
CCP Veritas
|
Posted - 2010.11.16 17:31:00 -
[4]
Originally by: Athea G'man What's the huge spike on 1 Oct 2009?
Haven't got a clue. Maybe someone was running SETI@Home on TQ that day.
|
|
|
CCP Veritas
|
Posted - 2010.11.16 17:51:00 -
[5]
Originally by: Erebus Adrastos I'm a little confused on how the number of drakes kill in null-sec is represented as a percentage. Percentage of what?
Atlas can correct me if I'm wrong, but I believe it's a percentage of nullsec ships killed.
|
|
|
CCP Atlas
|
Posted - 2010.11.16 18:04:00 -
[6]
Originally by: CCP Veritas
Originally by: Erebus Adrastos I'm a little confused on how the number of drakes kill in null-sec is represented as a percentage. Percentage of what?
Atlas can correct me if I'm wrong, but I believe it's a percentage of nullsec ships killed.
Yes, percentage of total number of ships killed in nullsec PvP.
|
|
|
CCP Explorer
|
Posted - 2010.11.16 18:33:00 -
[7]
Originally by: Meiyang Lee
Originally by: CCP Veritas
Originally by: Athea G'man What's the huge spike on 1 Oct 2009?
Haven't got a clue. Maybe someone was running SETI@Home on TQ that day.
Does that spike happen to coincide with Fanfest? I know that sounds odd, but FF started on October 1st in 2009.
This CPU spike was "TRANQUILITY HOTFIX v.6.14.101786_1". This code was deployed 29 September 2009 but reverted 30 September 2009 for further analysis. It was a problem with a memory allocation change in the Stackless Python interpreter.
Erlendur S. Thorsteinsson Software Director EVE Online, CCP Games |
|
|
CCP Explorer
|
Posted - 2010.11.16 18:38:00 -
[8]
Originally by: Hentes Zsemle Am i the only one to notice that CCP's biggest accomplishment with reducing server load was banning macros?
Yes, you are, and the arrow on the graph pointing to "Unholy Rage" is purely coincidental.
Erlendur S. Thorsteinsson Software Director EVE Online, CCP Games |
|
|
CCP Veritas
|
Posted - 2010.11.16 20:38:00 -
[9]
Originally by: Bartholomeus Crane The underlying problem is not necessarily the ship used and it's CPU utilisation.
I agree with you, but my point in showing that graph isn't that fixing missiles will fix lag, it's that making them performant should help reverse some of the degradation we've seen in large fleet engagements around Dominion-launch-time. They really are quite exceptionally worse than guns, in ways that I'm quite certain would offend your sensibilities.
I'm well aware that all the optimizations we're doing at this point are, in the limit, futile, since we always allow/encourage players to bring one more ship. Since I don't know when/if that's going to change, I'll keep aiming down the path of supporting the current design as best we can.
|
|
|
|
|