open All Channels
seplocked EVE Information Portal
blankseplocked New Dev Blog: Fixing Lag: Picking Up Low Hanging Fruit
 
This thread is older than 90 days and has been locked due to inactivity.


 
Pages: 1 [2] 3

Author Topic

Charles37
Posted - 2010.09.13 20:22:00 - [31]
 

I'll read the rest of the blog after I've stripped my Maelstrom of it's current setup in favor of one the CCP sanctioned setup that uses all Civilian modules.

CCP Veritas

Posted - 2010.09.13 20:36:00 - [32]
 

Originally by: Charles37
I'll read the rest of the blog after I've stripped my Maelstrom of it's current setup in favor of one the CCP sanctioned setup that uses all Civilian modules.


I'm tellin' ya, it can kill a Titan! Well, so long as the Titan isn't piloted and you have a lot of time.

Sedilis
Lead Farmers
Kill It With Fire
Posted - 2010.09.13 20:56:00 - [33]
 

Another nice blog.

Can we have more graphs showing the "Awesomeness rating" please Laughing

Ulair Memmet
ORIGIN SYSTEMS
Posted - 2010.09.13 20:56:00 - [34]
 

Originally by: CCP Veritas
Originally by: Charles37
I'll read the rest of the blog after I've stripped my Maelstrom of it's current setup in favor of one the CCP sanctioned setup that uses all Civilian modules.


I'm tellin' ya, it can kill a Titan! Well, so long as the Titan isn't piloted and you have a lot of time.


... and a hack to disable the titan's passive shield recharging ^^

Nye Jaran
Posted - 2010.09.13 21:59:00 - [35]
 

Good blog, but your math is wrong here:

Quote:

A drop from approx 82% CPU to approx 35%, a 57% improvement



Should be a 47% improvement, or the pre /post number(s) are wrong.

ReddSky
Posted - 2010.09.13 22:16:00 - [36]
 

Great Job CCP. This kind of investigation / improvement on module lag is what players have been asking for for ages. Please keep up this kind of work, much appreciated.
Very Happy

StonerPhReaK
Nasgul Collective
Cascade Imminent
Posted - 2010.09.13 22:26:00 - [37]
 

Good Job CCP.Keep up the good work.

The mouths of many are heard far and wide when errything is borked,Fix it and hear nothing back in the way of thank you's.I guess silence is indeed golden.

CCP Explorer

Posted - 2010.09.13 22:50:00 - [38]
 

Originally by: Nye Jaran
Good blog, but your math is wrong here:
Quote:
A drop from approx 82% CPU to approx 35%, a 57% improvement
Should be a 47% improvement, or the pre /post number(s) are wrong.
82% to 35% is an improvement of 47 percentage points but 1-35/82 is 0.57 or 57%.

CCP Veritas

Posted - 2010.09.13 22:52:00 - [39]
 

Damn it Explorer, beat me to it!

But yeah, what he said. Relative difference is what matters, not absolute.

R4 D2
Posted - 2010.09.13 23:02:00 - [40]
 

  • Corporate buzzword [Check]

  • Mission runners used as guinea pigs [Check]

  • Lag reduced by a significant percent [Check]

  • DevBlog aproved!

Elzon1
Caldari
Shadow Boys Corp
Bloodbound.
Posted - 2010.09.13 23:38:00 - [41]
 

Edited by: Elzon1 on 13/09/2010 23:38:41
Good blog and using missioners as guinea pigs very nice Cool

So you guys are still looking for lag it seems. Perhaps a map would help? Maybe an interactive coding display of all the previous expansions compared against each other would help? Perhaps with such a display you could plug various data from player tests or thin client tests to test for errors or ineffeciencies. Maybe make lag appear as red through the map comparisons so you can "see" where the lag resides in the coding itself YARRRR!!

You guys already have something like this? Neutral

Manfred Rickenbocker
Pan Galactic Gargle Blasters
Important Internet Spaceship League
Posted - 2010.09.14 00:06:00 - [42]
 

Edited by: Manfred Rickenbocker on 14/09/2010 00:07:23
Sooo, Ive gotta ask: Why wasnt something this simple investigated sooner?

Regardless, good job! BTW, since this was such an easy catch, might I suggest searching for ALL instances of the server creating dummy instances of a ship (edit: or any object for that matter!) and querying them to see if those might be fruitful as well?

Xailia
Unsteady Corporation
Posted - 2010.09.14 01:48:00 - [43]
 

\o/ yay blogs!

Hack Harrison
Caldari
Posted - 2010.09.14 02:30:00 - [44]
 

Originally by: Manfred Rickenbocker
Edited by: Manfred Rickenbocker on 14/09/2010 00:07:23
Sooo, Ive gotta ask: Why wasnt something this simple investigated sooner?

Regardless, good job! BTW, since this was such an easy catch, might I suggest searching for ALL instances of the server creating dummy instances of a ship (edit: or any object for that matter!) and querying them to see if those might be fruitful as well?


Because they didn't have the thin client infrastructure available earlier Shocked

Rakshasa Taisab
Caldari
Sane Industries Inc.
Posted - 2010.09.14 05:33:00 - [45]
 

Originally by: Amy Garzan
5% is better than nothing. Keep up the good work.

Better than nothing?

I think you are failing to understand how much of a ****le in the pants moment finding a easy-to-apply 5% optimization in a mature high-performance system really is...

Ban Doga
Posted - 2010.09.14 06:22:00 - [46]
 

Great work!
A 5% drop in overall load is quite remarkable for a single optimization.

Do you use any profiling tools to find those bottlenecks/hot spots or do you have code reviews to find them?

Louis deGuerre
Gallente
Malevolence.
Posted - 2010.09.14 07:56:00 - [47]
 

Nice work. Did you get same results using missiles and guns or is that just an assumption ?

CCP Explorer

Posted - 2010.09.14 08:29:00 - [48]
 

Originally by: Louis deGuerre
Nice work. Did you get same results using missiles and guns or is that just an assumption ?
All repeating modules.

Nye Jaran
Posted - 2010.09.14 12:55:00 - [49]
 

Originally by: CCP Explorer
Originally by: Nye Jaran
Good blog, but your math is wrong here:
Quote:
A drop from approx 82% CPU to approx 35%, a 57% improvement
Should be a 47% improvement, or the pre /post number(s) are wrong.
82% to 35% is an improvement of 47 percentage points but 1-35/82 is 0.57 or 57%.


Ah, didn't think of it like that. Suppose that's why you people all run the game and I just play the thing. Very Happy

Trebor Daehdoow
Gallente
Sane Industries Inc.
Posted - 2010.09.14 13:05:00 - [50]
 

Originally by: CCP Explorer
Originally by: Louis deGuerre
Nice work. Did you get same results using missiles and guns or is that just an assumption ?
All repeating modules.

Since this is the case, would it not be a good idea, as several people have suggested, to consider making some common modules passive -- such as Damage/Drone Control Units?

Or is the current overhead for repeating modules low enough now that ~1 less repeating module per pew-pewing ship (and one less command per jumpin) is not enough of a win?

Marconus Orion
D00M.
Northern Coalition.
Posted - 2010.09.14 14:08:00 - [51]
 

Originally by: Trebor Daehdoow
Originally by: CCP Explorer
Originally by: Louis deGuerre
Nice work. Did you get same results using missiles and guns or is that just an assumption ?
All repeating modules.

Since this is the case, would it not be a good idea, as several people have suggested, to consider making some common modules passive -- such as Damage/Drone Control Units?

Or is the current overhead for repeating modules low enough now that ~1 less repeating module per pew-pewing ship (and one less command per jumpin) is not enough of a win?


Damn good question. CCP?

UVPhoenix2
Rim Collection RC
Test Alliance Please Ignore
Posted - 2010.09.14 14:28:00 - [52]
 

Good work CCP. Dev updates like this are always fun to read. Keep 'em coming!

EdgeOf Insanity
Posted - 2010.09.14 15:10:00 - [53]
 

Edited by: EdgeOf Insanity on 14/09/2010 15:10:45
Nvm, someone already pwnt it.

CCP Veritas

Posted - 2010.09.14 17:42:00 - [54]
 

Originally by: Trebor Daehdoow
Since this is the case, would it not be a good idea, as several people have suggested, to consider making some common modules passive -- such as Damage/Drone Control Units?


I don't know how much of a win it would be offhand, but I have been keeping an eye on your thread over in the Assembly about it. Very interesting discussion, that.

Trebor Daehdoow
Gallente
Sane Industries Inc.
Posted - 2010.09.14 20:59:00 - [55]
 

Originally by: CCP Veritas
Originally by: Trebor Daehdoow
Since this is the case, would it not be a good idea, as several people have suggested, to consider making some common modules passive -- such as Damage/Drone Control Units?


I don't know how much of a win it would be offhand, but I have been keeping an eye on your thread over in the Assembly about it. Very interesting discussion, that.

Zagdul has suggested that you test this during an upcoming mass test; it would seem to me that it would be pretty easy to do a thin-client test and follow it up with a live test if it is promising. Even if it shaves 1/2% off the load, that's a nice little win, and best of all, you can dump all the work on another team! Twisted Evil

Noun Verber
Gallente
Posted - 2010.09.15 03:43:00 - [56]
 

You claim that "players using modules is a very common thing", but offer no proof!

Shenanigans!

Melchior Grimm
Caldari
Posted - 2010.09.15 05:08:00 - [57]
 

Thank you devs. I appreciate the blogs as well as the lag fixes.

Disgusting Carebear
Posted - 2010.09.15 07:45:00 - [58]
 

Originally by: Noun Verber
You claim that "players using modules is a very common thing", but offer no proof!

Shenanigans!


Can we have some clarification on this please?

Joey's Mom
Posted - 2010.09.15 22:28:00 - [59]
 

"low hanging *********", that is all!

Andrevv
ANZAC ALLIANCE
IT Alliance
Posted - 2010.09.16 01:46:00 - [60]
 

Seeing as 0.0 combat involves manual cycling anyway, I don't expect this particular optimization to have any impact on fleet fights. Good job regardless :)

Originally by: CCP Explorer
Originally by: Louis deGuerre
Nice work. Did you get same results using missiles and guns or is that just an assumption ?
All repeating modules.


Originally by: Blog
Analysis showed that most of this CPU time was spent determining if the module was in the process of reloading whenever it was repeated.


Am I understanding this correctly, that any repeating module, regardless of whether it takes a charge, does a query to check if its charge group is empty?? because that is what the blog is implying... which seems kind of pointless.


Pages: 1 [2] 3

This thread is older than 90 days and has been locked due to inactivity.


 


The new forums are live

Please adjust your bookmarks to https://forums.eveonline.com

These forums are archived and read-only