These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

Test Server Feedback

 
  • Topic is locked indefinitely.
 

Duality Open Period - 19/10 to 23/10

First post First post
Author
Previous Topic Next Topic
CCP Goliath
C C P
C C P Alliance
#21 - 2012-10-18 16:50:09 UTC
Dinsdale Pirannha wrote:
CCP Goliath wrote:


For the purposes of AI testing, the only difference between Duo of Death and Buzz-Kill is that Duo of Death has Cruise Missile Batteries. Mordus Headhunters has some interesting trigger interactions that could throw up some unexpected stuff (please see above "a few exceptions") but any other mission with odd triggers would provide basically the same information (for instance, Worlds Collide). BTW, you should try just deleting every sweeping generalisation or assumption from the end of your posts - it would make you seem a lot more reasonable.


Sorry, you are dead wrong.

Buzzkill has multiple triggers and includes a huge number of frigs and cruisers, which orbit at ranges varying from 20km to 2500m.
Duo of Death has a single spawn of 2 Spider Drones, 4 Cruise Towers at extreme range, and 2 Battleships that orbit at 49km.

To equate the 2 missions is ridiculous.

I could see using Sentry Drones on the Cruise Missile towers and Centus Tyrants in Duo of Death, and possibly a smartbomb to take out the spider drones.
Precisely how would said Sentry Drones fare against 6 Buzzkill assault frigs orbiting at 12km from your ship? Or said smartbomb work against the same assault frigs?

To equate the methods of completing them is equally ridiculous.
To suggest that testing the AI in one of the missions allows for an extrapolation of the effectiveness of the AI for the other is ludicrous.


And as for being reasonable, I have seen how "being reasonable" works on these forums.

And I ask again, by the same logic that one mission, with few exceptions, is the same as another, does that mean if one null sec plex is tested, that is good enough? Or does null sec get a different treatment than high sec?

You know what, I have to go do some RL stuff now, and stop posting on this disaster. It is utterly pointless anyway.
The only thing that can reverse your view is a massive hue and cry on the forums post Dec 4th and if subs start dropping.

Since the sub rate may get conflated with the Dust release, no one will know for sure the true impact until March/ April.

Enjoy yet another triumph over high sec. Whether it lasts remains to be seen.


Ugh. OK. The triggers point I have already covered, and the triggers on buzzkill aren't interesting from a testing perspective when looking at AI. You seem to be talking about the efficiency and technique of completing missions, which frankly I couldn't care less about - that's not the goal of the tests. You don't know our AI - I do. I also know how to effectively test it, which again, would be something you clearly do not know how to do. Kindly desist arguing with me on this topic. I understand you are entitled to your opinion, but this is my profession, I am extremely capable in it, and you are out of your element.

To quote the internet, Dinsdale, wat r u doin, Dinsdale, stahp.

CCP Goliath | QA Director | EVE Illuminati | @CCP_Goliath
Ghazu
#22 - 2012-10-18 16:51:21 UTC
Dinsdale you make hisecers look bad.

http://www.minerbumping.com/ lol what the christ https://forums.eveonline.com/default.aspx?g=posts&m=2299984#post2299984
Xercodo
Cruor Angelicus
#23 - 2012-10-18 16:52:15 UTC
I think the point is that all the missions and PLEXs mostly boil down to NPCs of 5 or 6 different sizes (frig, dessy, cruiser, BC, BS, and occasional industrial) spawning at different ranges shooting you with a few sentries thrown in.

If you're a drone boat then you're gonna use lights on frigs and dessies, mediums on cruisers and BCs, and use sentries and heavies on BSes and sentries depending on range.

Not much else changes between missions. There are more of one ship class, there are fewer of others, sometimes some elite ships are mixed in like loyal sansha and elder blood raiders and dire guristas, but that's about it.

The Drake is a Lie
Tippia
Sunshine and Lollipops
#24 - 2012-10-18 16:58:49 UTC  |  Edited by: Tippia
Dinsdale Pirannha wrote:
Sorry, you are dead wrong.

Buzzkill has multiple triggers and includes a huge number of frigs and cruisers, which orbit at ranges varying from 20km to 2500m.
Duo of Death has a single spawn of 2 Spider Drones, 4 Cruise Towers at extreme range, and 2 Battleships that orbit at 49km.

To equate the 2 missions is ridiculous.
Since neither mission has any special scripting beyond “spawn X when Y dies”, no, it really isn't. They're essentially the same, except that, presumably, those towers in Duo follow a slightly different logic than the ships that fly around and their presence is the only thing that separates the two missions.

What you're talking about isn't testing — it's filling out eve-survival mission reports on how to complete a mission, and that is not part of the bug hunting and ironing out oddities in the AI. Basically, you are completely confused about what releasing and testing the code is about. What you intend to do doesn't need any Duality testing. It can (and will) be done on TQ as it has always been done.

edit: Bah, ninja'd by Goliath.
Petrus Blackshell
Rifterlings
#25 - 2012-10-18 17:05:18 UTC
I thought I'd point out for the sake of completeness of this thread that the "AI testing" is not "testing what you need to do to run these same missions with the AI changes". It is "testing if the new AI works as expected, free of glitches, etc."

Problems you want to spot are things like "Rats don't fire on me after this particular trigger... wtf?" or "Battleship rats are primarying frigates"

Problems that are beyond the scope of the test are things like : "The Dominix sucks at mission X now" or "Stealth Noctis nerf! I can't salvage while mission-running anymore!"

Accidentally The Whole Frigate - For-newbies blog (currently on pause)
MeBiatch
GRR GOONS
#26 - 2012-10-18 18:58:54 UTC  |  Edited by: MeBiatch
you have to love goliath... he reminds me of a new teacher who was really nice back in the day but got walked all over by her students and is now jadded...


though most of the time its not what you have to say but how you go about saying it...

so in fact he is right... but just posts it in a way that he is really saying htfu... lolz... one of me new fav devs...

There are no stupid Questions... just stupid people... CCP Goliath wrote:

Ugh ti-di pooping makes me sad.
DJ P0N-3
Table Flippendeavors
#27 - 2012-10-18 19:09:02 UTC
CCP Goliath wrote:
To quote the internet, Dinsdale, wat r u doin, Dinsdale, stahp.


I'd have to disagree with you on this point. All Spiny Norman ever says is "DINSDALE".
Dinsdale Pirannha
Pirannha Corp
#28 - 2012-10-18 20:07:59 UTC
CCP Goliath wrote:

Ugh. OK. The triggers point I have already covered, and the triggers on buzzkill aren't interesting from a testing perspective when looking at AI. You seem to be talking about the efficiency and technique of completing missions, which frankly I couldn't care less about - that's not the goal of the tests. You don't know our AI - I do. I also know how to effectively test it, which again, would be something you clearly do not know how to do. Kindly desist arguing with me on this topic. I understand you are entitled to your opinion, but this is my profession, I am extremely capable in it, and you are out of your element.

To quote the internet, Dinsdale, wat r u doin, Dinsdale, stahp.


You can ridicule me all you want. I am used to it. It means that the null sec zealots are annoyed. Good.

If that is all you are testing, the base mechanics, then why open up the server at all to the players?
Fox Four in the 40 plus page thread has stated in numerous instances of him/her testing the mechanism already.
He / she refers to a slider which allows for the AI switching aggressiveness to be altered, and kept asking us to be patient about stating this is doom for drones until we can test it on Duality. So Fox Four has already tested the actual game mechanic.

So now you are reversing what Fox Four has said.
You are saying, "Oh, it was never about testing the fairness and viability of mission or plex running with a Duality test, but rather a test whether the game mechanics work at all."

So precisely when are we to test whether this AI allows many many missions and plexes to still be viable?
When it hits TQ?

And for the 3rd time I ask, are you suggesting that testing ONE exploration plex is good enough, since the AI works the same in all plexes? Or are there going to be tweaks based on null sec player input, since it is clear that any comments about high sec mission viability is going to be ignored and/or ridiculed.
Salpun
Global Telstar Federation Offices
Masters of Flying Objects
#29 - 2012-10-18 20:14:16 UTC
Dinsdale Pirannha wrote:
CCP Goliath wrote:

Ugh. OK. The triggers point I have already covered, and the triggers on buzzkill aren't interesting from a testing perspective when looking at AI. You seem to be talking about the efficiency and technique of completing missions, which frankly I couldn't care less about - that's not the goal of the tests. You don't know our AI - I do. I also know how to effectively test it, which again, would be something you clearly do not know how to do. Kindly desist arguing with me on this topic. I understand you are entitled to your opinion, but this is my profession, I am extremely capable in it, and you are out of your element.

To quote the internet, Dinsdale, wat r u doin, Dinsdale, stahp.


You can ridicule me all you want. I am used to it. It means that the null sec zealots are annoyed. Good.

If that is all you are testing, the base mechanics, then why open up the server at all to the players?
Fox Four in the 40 plus page thread has stated in numerous instances of him/her testing the mechanism already.
He / she refers to a slider which allows for the AI switching aggressiveness to be altered, and kept asking us to be patient about stating this is doom for drones until we can test it on Duality. So Fox Four has already tested the actual game mechanic.

So now you are reversing what Fox Four has said.
You are saying, "Oh, it was never about testing the fairness and viability of mission or plex running with a Duality test, but rather a test whether the game mechanics work at all."

So precisely when are we to test whether this AI allows many many missions and plexes to still be viable?
When it hits TQ?

And for the 3rd time I ask, are you suggesting that testing ONE exploration plex is good enough, since the AI works the same in all plexes? Or are there going to be tweaks based on null sec player input, since it is clear that any comments about high sec mission viability is going to be ignored and/or ridiculed.

They are interested in how the server handles the new code in this build. While testing the playablity will happen as a result of this they need real numbers when 500 people run missions all at once, as an indication if the server load is going to be higher or lower with this new code. This is the reason why they are testing this code this early.

If i dont know something about EVE. I check https://wiki.eveonline.com/en/wiki/ISK_The_Guide

See you around the universe.
MeBiatch
GRR GOONS
#30 - 2012-10-18 20:21:06 UTC
Salpun wrote:
Dinsdale Pirannha wrote:
CCP Goliath wrote:

Ugh. OK. The triggers point I have already covered, and the triggers on buzzkill aren't interesting from a testing perspective when looking at AI. You seem to be talking about the efficiency and technique of completing missions, which frankly I couldn't care less about - that's not the goal of the tests. You don't know our AI - I do. I also know how to effectively test it, which again, would be something you clearly do not know how to do. Kindly desist arguing with me on this topic. I understand you are entitled to your opinion, but this is my profession, I am extremely capable in it, and you are out of your element.

To quote the internet, Dinsdale, wat r u doin, Dinsdale, stahp.


You can ridicule me all you want. I am used to it. It means that the null sec zealots are annoyed. Good.

If that is all you are testing, the base mechanics, then why open up the server at all to the players?
Fox Four in the 40 plus page thread has stated in numerous instances of him/her testing the mechanism already.
He / she refers to a slider which allows for the AI switching aggressiveness to be altered, and kept asking us to be patient about stating this is doom for drones until we can test it on Duality. So Fox Four has already tested the actual game mechanic.

So now you are reversing what Fox Four has said.
You are saying, "Oh, it was never about testing the fairness and viability of mission or plex running with a Duality test, but rather a test whether the game mechanics work at all."

So precisely when are we to test whether this AI allows many many missions and plexes to still be viable?
When it hits TQ?

And for the 3rd time I ask, are you suggesting that testing ONE exploration plex is good enough, since the AI works the same in all plexes? Or are there going to be tweaks based on null sec player input, since it is clear that any comments about high sec mission viability is going to be ignored and/or ridiculed.

They are interested in how the server handles the new code in this build. While testing the playablity will happen as a result of this they need real numbers when 500 people run missions all at once, as an indication if the server load is going to be higher or lower with this new code. This is the reason why they are testing this code this early.


thats the problem with QA they are looking for specifics about one thing... but they get responces about everything and it must be annoying as hell to have to sift threw crap just to get proper responces...

somone needs to get the QA team a large joint... otherwise they will all be jadded...

There are no stupid Questions... just stupid people... CCP Goliath wrote:

Ugh ti-di pooping makes me sad.
Tippia
Sunshine and Lollipops
#31 - 2012-10-18 20:25:49 UTC
Dinsdale Pirannha wrote:
If that is all you are testing, the base mechanics, then why open up the server at all to the players?
Because it's the players who are going to go up against this AI, and they will always behave in ways that are different to how the devs have come to go up against them.

We always do things they haven't predicted, and they want to know what happens when our unpredictability and the new code collide.

Quote:
And for the 3rd time I ask, are you suggesting that testing ONE exploration plex is good enough, since the AI works the same in all plexes?
No. He's suggesting that testing one class of rat encounters will give good data for all encounters of the same class. Not all plexes belong to the same class (or, more accurately, every complex or mission will cover a number of cases, and data from one part of one mission will answer questions about a different part of a different mission).

Quote:
So precisely when are we to test whether this AI allows many many missions and plexes to still be viable?
When it hits TQ?
Whether or not there are parts of the AI that simply makes some missions impossible can be tested now; exactly how to overcome the missions that can be overcome (which, unless something has gone wrong, will be all of them) is something that can be done on TQ.
Ranger 1
Ranger Corp
Vae. Victis.
#32 - 2012-10-18 21:42:47 UTC  |  Edited by: Ranger 1
In addition, let us not forget that during that testing window several hundred players will be running thousands of missions. What is important is that they are being run (and any problems reported), not that YOU have time to run them all personally. Blink

Quote:
You can ridicule me all you want. I am used to it.


We know.

View the latest EVE Online developments and other game related news and gameplay by visiting Ranger 1 Presents: Virtual Realms.
KIller Wabbit
MEME Thoughts
#33 - 2012-10-19 06:48:55 UTC
CCP Goliath wrote:

... but this is my profession, I am extremely capable in it, and you are out of your element.

To quote the internet, Dinsdale, wat r u doin, Dinsdale, stahp.


Helluva attitude. Yeah, other professional software developers never play a game.

Obviously, EVE is known for having thorough testing. Roll

You'd be a lot better off going with the attitude that "if it hasn't been tested it is broken." That way you think about how to get more test coverage in, not if you've successfully hit a random number of high points to adequately fill out a weekly box.
War Kitten
Panda McLegion
#34 - 2012-10-19 11:30:42 UTC
KIller Wabbit wrote:
CCP Goliath wrote:

... but this is my profession, I am extremely capable in it, and you are out of your element.

To quote the internet, Dinsdale, wat r u doin, Dinsdale, stahp.


Helluva attitude. Yeah, other professional software developers never play a game.

Obviously, EVE is known for having thorough testing. Roll

You'd be a lot better off going with the attitude that "if it hasn't been tested it is broken." That way you think about how to get more test coverage in, not if you've successfully hit a random number of high points to adequately fill out a weekly box.




Dinsdale isn't a professional software developer. He's raving lunatic who thinks "nullsec zealots" and CCP are all out to get him and ruin highsec. He's a former nullbear, now apparently living in highsec, and he whines and moans about how every change being even thought about is obviously going to ruin his whole highsec drone mission-running existence. I suspect he got kicked out of nullsec and has a huge axe to grind now.

He generates that kind of attitude in anyone who tries to remotely address his concerns with his illogical paranoia.

I don't judge people by their race, religion, color, size, age, gender, or ethnicity. I judge them by their grammar, spelling, syntax, punctuation, clarity of expression, and logical consistency.
Allandri
Liandri Industrial
#35 - 2012-10-19 11:56:43 UTC
Are the AI changes going to affect FW NPC's? If so, I might have to try and run a few of the missions/complexes
CCP Goliath
C C P
C C P Alliance
#36 - 2012-10-19 12:23:12 UTC
Allandri wrote:
Are the AI changes going to affect FW NPC's? If so, I might have to try and run a few of the missions/complexes


Yup, all NPCs except CONCORD and maybe a few other exceptions will be affected.

CCP Goliath | QA Director | EVE Illuminati | @CCP_Goliath
Rengerel en Distel
#37 - 2012-10-19 13:19:07 UTC
Any updates on when it's going to open?

With the increase in shiptoasting, the Report timer needs to be shortened.
Rengerel en Distel
#38 - 2012-10-19 14:08:41 UTC
So did the guy that was supposed to flip the switch go to the movies too?

With the increase in shiptoasting, the Report timer needs to be shortened.
nat longshot
The Scope
Gallente Federation
#39 - 2012-10-19 14:17:18 UTC
Rengerel en Distel wrote:
So did the guy that was supposed to flip the switch go to the movies too?


Hope CCP Goliath is enjoying the film given i have stay up to test on the damn thing 14:16 and still nothing.

YO CCP YOUR OVER A HOUR LATE...oh wait thats what there good at getting it wrong.

 [13:12:18] CCP Punkturis nat longshot you're a cutie.. OH YAH I WIN!!
Petrus Blackshell
Rifterlings
#40 - 2012-10-19 15:21:23 UTC
nat longshot wrote:
Rengerel en Distel wrote:
So did the guy that was supposed to flip the switch go to the movies too?


Hope CCP Goliath is enjoying the film given i have stay up to test on the damn thing 14:16 and still nothing.

YO CCP YOUR OVER A HOUR LATE...oh wait thats what there good at getting it wrong.

Deployment of code, particularly development-stage code, is a very inexact art, requiring just the right combination of skill, black magic, and alcohol on the part of the developers, and just a little bit of patience on the part of the people the deployment is meant for.

Accidentally The Whole Frigate - For-newbies blog (currently on pause)