I can think of a few ways something like that could happen, and I've actually gone and explained them in various forums, which (I believe) includes the old CO forums circa 2009. I could attempt to explain them here, but I'm not going to because:
1) Not enough information, so I don't know what would be relevant.
2) The explanation is long, dry, technical, and boring.
3) Nobody reads them, either because #2 or because they'd rather be angry than learn.
The hideously oversimplified, short-n-dirty version is:
Fixes to a problem art asset can be applied to the base object, or to various layers of information attached to that object. Changing the base object can result in unexpected errors because of adjustment layers.
For example, in Skyrim, a normal map can be inverted by changing the normal map itself, or by choosing Invert Normals in the .nif object that the engine actually reads. If a change is made to a base normal map to fix an inversion issue, but the Invert Normals option is not changed, the Invert Normals option will re-invert the fixed normal map so that it displays incorrectly.
Given a few behaviors I've noticed in-game myself, I suspect that the issue in CO is significantly more complex than my example.
-
"Whoever is careless with the truth in small matters cannot be trusted with important matters." - Albert Einstein
What concerns me is a lack of impact analysis. If I change a base class in my code, I know for certain that I have to regression test every derived class and every piece of code that invokes those derived classes. Cryptic needed to do the same thing here. Changing a base mesh means regression testing every texture and attachment mesh. That appears to been either not done at all, or simply shoveled over the wall into production before it was done on the assumption that players either wouldn't notice or wouldn't care.
And yes, communication would have gone a long way toward resolving things here. "Hey, everybody. We're making some fixes and improvements to the base female character model. You may notice some clipping, poke-through, or texture glitches over the next n of weeks as we work through the impacts. Thank you for your patience." Would that really have been so hard?
Choose your enemies carefully, because they will define you / Make them interesting, because in some ways they will mind you
They're not there in the beginning, but when your story ends / Gonna last with you longer than your friends
It's getting a bit tu quoque fallacy in here, with a sprinkling of ad hominem. "Don't complain about them not fixing it, you can't either so you have no room to talk."... The hell? Since when did I have to apply to a mechanic in order to be allowed to get annoyed that they were taking forever to fix my car? Do I have to apply to a restaurant to send back my steak? If my doctor messes up some stitches, do I have to go to medical school? What about if a building falls over in an earthquake, can I not blame the engineer without a master's degree? I don't have a license, can I not sue someone if they hit me with their car? Never flown an airplane, so if the one I'm in crashes I can't curse the pilot as I go down? See where I'm going with this?
Here's the truth. We mentioned a bunch of times on pts "this costume is severely jacked up, fix it before going live" and were ignored. They obviously didn't do all of the testing needed on the new mesh fixes, either, like *every programmer on the planet knows to do*. It's development 101, as has been mentioned: If you change something, you check what it's connected to. In fact, if you change something, it's almost like there's a platform to see how it behaves. I think we call it "the public test server"?
There's no need to be venomous, here, but let's not pretend these issues just magicked themselves into existence completely devoid of any developer culpability. And ffs, can we maybe not yell at the people who are trying to post *constructive* feedback on what is broken? "They're probably wrong", sure, but what if they're not? At the very least, knowing what parts are impacted can only help.
lol fyi? I'd just like to point out that this is now the third time Cryptic has tried to push a patch into the game that broke just a significant chunk of stuff that was only partially related to what was patched. I'm beginning to think the name's cursed or something. On the bright side, at least the chat server works this time around.
In game, I am @EvilTaco. Happily killing purple gang members since May 2008. RIP Caine
Two questions: Where have they said this? And why is the new mesh obviously faulty? I mean, you can just *look* at the shoulder blades on the back and see sections of the mesh bulging out in impossible ways - it's not even just that it clips through tight armor pieces, it's just flat-out improperly shaped to start with.
I was inclined to think that maybe Cryptic had someone not experienced with their models who made a mistake of slipping a seemingly identical mesh instead of an old mesh.
Especially since one of the things pretty arbitrary for 3d engines is which way normals are pointed, and if you are new to the engine... It's not something obvious at all if Y is + or - on pieces you are working with.
But this would be only once.
If it was the thing, it would be corrected long time ago.
Nobody would be going through other meshes and textures file after file opening and changing them.
And what if old CO mesh has some issues like doubled vertexes or split edges? Those aren't uncommon things at all, professionally made models may have them.
And that would be enough in the long term to go through changing every mesh to a new one. For an example, if mesh is dirty, then it makes rigging harder and you can't (for an example) speed up your work with weighting it automatically through whatever tool or script your 3d program has and then only finish it by hand.
And it means that dirty mesh slows your work on every other costume using it as a base... Because rigging it takes longer.
Could be a reason enough for CO, where models aren't very complicated, but they are multiparted and I suppose that rigging it to skeleton may be the hardest part of making a costume.
Also, a slew of newer costume pieces like Holoforce, Cosmic Knight and Tactical Armor are also using different mesh with slightly different geometry. They also clip with neck cowls (on back).
I was wondering why someone was using a different mesh for them, but maybe it was the case? They found an issue with standard Cryptic meshes and they went with not using it as a base?
Dunno. But it might be reason enough to go through the hassle of tuning everything else after changing base meshes.
What concerns me is a lack of impact analysis. If I change a base class in my code, I know for certain that I have to regression test every derived class and every piece of code that invokes those derived classes. Cryptic needed to do the same thing here. Changing a base mesh means regression testing every texture and attachment mesh. That appears to been either not done at all, or simply shoveled over the wall into production before it was done on the assumption that players either wouldn't notice or wouldn't care.
And yes, communication would have gone a long way toward resolving things here. "Hey, everybody. We're making some fixes and improvements to the base female character model. You may notice some clipping, poke-through, or texture glitches over the next n of weeks as we work through the impacts. Thank you for your patience." Would that really have been so hard?
... oh, thank god, someone I don't have to explain this stuff to.
You're right, more or less... but given some of the shuffling around, I think they're also dealing with undocumented legacy modifiers on top of everything else, and fixing bad practice can be... complicated. I've had to deal with some pretty nightmarish stuff, things that I looked at and went "Who in their right mind would do something like this?" There's a reason I used Skyrim as an example... a lot of the modding community wouldn't know best practice if it bit them.
(No, I'm not trying to excuse the problem. Explaining why it might have happened does not automatically mean attempting to get out of the consequences.)
@meedacthunist:
I noticed that, too. Maybe they're in the process of standardizing the meshes? Any of the things you mention would be a good reason to start using a cleaner mesh as a base, but... that'd also require replacing the old ones, then testing them all with existing modifiers, etc. Dunno. I don't work there, so I can't say for sure.
-
"Whoever is careless with the truth in small matters cannot be trusted with important matters." - Albert Einstein
"They're probably wrong", sure, but what if they're not?
Its the "their probably right" part that these players don't care about. They don't care that they are attacking the people who want it fixed instead directing their issues toward the people who broke it. Or they are just trying to cause chaos and distract from the problem.
Then there's the people who assume that there is more to the problem and claiming that there is possibly no way to understand the issue despite proof that a fix is simple and has been dealt with.
When Gentleman Crush first made himself publicly available to players on CO, long standing issues that have been made to his attention by players (who people claim don't know what they are talking about) were fixed all but immediately. Issues that devs have claimed were impossible to fix.
Something about each issue seems fishy, but most of the problems the game has is Cryptic being greedy and not offering enough development time and giving that development time to devs who know the game. They don't want to give CO players the time of day anymore and that's how this all ends.
Flourishing games require a developer and player base relationship which we don't even meet the bare minimum of and have lacked since the conception of On Alert. Cryptic North and Gentleman Crush's first months being the only exception.
... oh, thank god, someone I don't have to explain this stuff to.
You're right, more or less... but given some of the shuffling around, I think they're also dealing with undocumented legacy modifiers on top of everything else, and fixing bad practice can be... complicated. I've had to deal with some pretty nightmarish stuff, things that I looked at and went "Who in their right mind would do something like this?"
Reminds me of one of my less pleasant experiences in the Air Force. Some folks think it's odd that nuclear-weapons planning was being done in COBOL - but that was significantly easier and less clunky than the original, FORTRAN. And one of my assignments was to go through all the old FORTAN modules and convert them to COBOL.
Many times I would trace the evolution of a variable, find how it was treated, and start ranting about what kind of half-educated lunatic would do things like that to a poor inoffensive dataset. Things that should have been variables were hard-coded, and when the data changed they would hard-code another piece. (For instance, weapon yields - they changed significantly over the decades, and the module I have in mind referenced seven different hardcoded yields instead of a single variable that could be filled from the database.)
tl;dr - legacy code can really suck hard.
"Science teaches us to expect -- demand -- more than just eerie mysteries. What use is a puzzle that can't be solved? Patience is fine, but I'm not going to stop asking the universe to make sense!"
Blah blah blah brilliant stuff that wins the thread. Hurray.
As they say...winner...winner...chicken...dinner? Sure, let's go with that. :biggrin:
Gradii: Honestly, they should fix that twice then. Once to clean it up and once to slap some Goku Hair on it for an actual "Bad Hair Day Emoticon." Then everyone wins.
Silicon: It has been said multiple times, likely too much by myself as well, but I agree that one of the major issues here is impractical timetables for PTS and how that data is actually handled. Communication, Testing, Civil Exchanges of Information/Suggestions/Whatever, More Communication, and enough time to make it right so you "Release" onto Live and not "Inflict" upon Live should be the baseline goal in this process. IMO, from current observations, "NEW" Old Cryptic has shown to be better in this department than Old Cryptic but not as good as Cryptic North. I, personally, see them trying though. I see Gadfly making many more posts than Old Cryptic did. I see the efforts but, on the other hand, we all see the current results as well. I see the potential for a proper PTS -> Live Setup...the question is, and always kinda has been from one degree to another, whether potential becomes reality.
Jon: Sure we don't have access to all the files and data....but what about those rare few who can piece together the problem and the actual solution just from what we DO have?
Buxom: Buxom, in your case it's not so much whether you're right or wrong alot of the time that apears to draw so much ire from the playerbase and such on these forums. It's how you present the information. It's not like the actual issues and such are a super hard secret to get from those who know. I would suspect there is little to nothing you know about these particular issues that I don't know, for example...the difference is purely a matter of tact and presentation. Your word choices come accross as much more accusational, brash, and tin foil hatty than pretty much anything I've ever posted on these forums. Consider that for a momment. Try presenting the information as information and not with insulting adjectives, name calling, or malicious plots. A is Broke. I believe A + B will fix the problem. The End.
Also, I only really know of two people on these forums who try to use chaos to distract from the actual problems. One of them is usually half oblivious that they're doing it and "The King" doesn't show up unless he smells a certain level of player sorrow on which to feed. So...yeah...that's 2 people. Not exactly an epidemic.
Crush wasn't the only person to make themselves available ingame for conversations and fix things. Lordgar, Gadfly, Robo, and even Tumerboy(who was an artist but went well beyond that trying to help out) have also performed that task. Also, there's TT, who was/is the single best CR by a country mile that this game has ever seen in the history of ever. When he "passed things up the chain", to borrow a phrase from a lesser CR, stuff actually came back. If I dug through the mountains of screenshots I have on my HD between the office and home office PCs I'm fairly certain I could find at least 1 pic of all 6 of those people standing in the Ren Center and just listening to the players....and even some of the old GMs(Drannic, IndigoFyre, Aand, Hopkins, .....) too, back when we used to have those.
But, as has been said time and time again, the communication levels need ramped up. So...8 words time.
STATE OF THE GAME
ASK CRYPTIC
PRETTY PLEASE*
Silicon: It has been said multiple times, likely too much by myself as well, but I agree that one of the major issues here is impractical timetables for PTS and how that data is actually handled. Communication, Testing, Civil Exchanges of Information/Suggestions/Whatever, More Communication, and enough time to make it right so you "Release" onto Live and not "Inflict" upon Live should be the baseline goal in this process. IMO, from current observations, "NEW" Old Cryptic has shown to be better in this department than Old Cryptic but not as good as Cryptic North. I, personally, see them trying though. I see Gadfly making many more posts than Old Cryptic did. I see the efforts but, on the other hand, we all see the current results as well. I see the potential for a proper PTS -> Live Setup...the question is, and always kinda has been from one degree to another, whether potential becomes reality.
Oh, yes. There have been a few too many "push to PTS on Wednesday, push live on Thursday" cycles lately. Listen to player feedback? Often there's no time for player feedback.
But, as has been said time and time again, the communication levels need ramped up. So...8 words time.
STATE OF THE GAME
ASK CRYPTIC
PRETTY PLEASE*
*Nope, not gonna let this one go.
You know what? I'm back on that bandwagon. I had given up hope with CNorth in charge, but how about a new broom sweeping clean and all that?
Choose your enemies carefully, because they will define you / Make them interesting, because in some ways they will mind you
They're not there in the beginning, but when your story ends / Gonna last with you longer than your friends
It's getting a bit tu quoque fallacy in here, with a sprinkling of ad hominem. "Don't complain about them not fixing it, you can't either so you have no room to talk."... The hell? ... See where I'm going with this?
Yeah, I do. I sure hope you're not directing that at me, because it'd be a straw man argument, seeing as how I certainly haven't denied anyone the right to complain. Pulling "facts" and "solutions" from where the sun shineth not is an entirely different thing.
Then there's the people who assume that there is more to the problem and claiming that there is possibly no way to understand the issue despite proof that a fix is simple and has been dealt with.
I'd love to see that proof. I've yet to see any in this thread.
Yeah. I don't envy programmers that aspect of the job... worst I usually have to deal with is bad tessellation or broken geometry, or just having to bug the problem.
Sure we don't have access to all the files and data....but what about those rare few who can piece together the problem and the actual solution just from what we DO have?
Those people would be making what is called an "educated guess." Guessing, even educated guessing, isn't fact... and unless Cryptic becomes less cryptic *cough* about information, there's no way to vet those guesses against the truth.
On a side note: this is pretty much exactly what was going on three years ago when I stopped posting the last time. Justifiably angry players making unjustifiably abusive comments, with a liberal sprinkling of the confidence of ignorance as to how easy fixing software is and absolutely no interest in anyone else's point of view, or in actually reading anything except what they want to see. In other words, internet as usual.
-
"Whoever is careless with the truth in small matters cannot be trusted with important matters." - Albert Einstein
Comments
I believe what people are referring to is ladygadfly's post (post #18) in this thread......
http://co-forum.perfectworld.com/showthread.php?t=311431
Click here to check out my costumes/milleniumguardian (MG) in-game/We need more tights, stances and moods
1) Not enough information, so I don't know what would be relevant.
2) The explanation is long, dry, technical, and boring.
3) Nobody reads them, either because #2 or because they'd rather be angry than learn.
The hideously oversimplified, short-n-dirty version is:
Fixes to a problem art asset can be applied to the base object, or to various layers of information attached to that object. Changing the base object can result in unexpected errors because of adjustment layers.
For example, in Skyrim, a normal map can be inverted by changing the normal map itself, or by choosing Invert Normals in the .nif object that the engine actually reads. If a change is made to a base normal map to fix an inversion issue, but the Invert Normals option is not changed, the Invert Normals option will re-invert the fixed normal map so that it displays incorrectly.
Given a few behaviors I've noticed in-game myself, I suspect that the issue in CO is significantly more complex than my example.
"Whoever is careless with the truth in small matters cannot be trusted with important matters." - Albert Einstein
And yes, communication would have gone a long way toward resolving things here. "Hey, everybody. We're making some fixes and improvements to the base female character model. You may notice some clipping, poke-through, or texture glitches over the next n of weeks as we work through the impacts. Thank you for your patience." Would that really have been so hard?
They're not there in the beginning, but when your story ends / Gonna last with you longer than your friends
Here's the truth. We mentioned a bunch of times on pts "this costume is severely jacked up, fix it before going live" and were ignored. They obviously didn't do all of the testing needed on the new mesh fixes, either, like *every programmer on the planet knows to do*. It's development 101, as has been mentioned: If you change something, you check what it's connected to. In fact, if you change something, it's almost like there's a platform to see how it behaves. I think we call it "the public test server"?
There's no need to be venomous, here, but let's not pretend these issues just magicked themselves into existence completely devoid of any developer culpability. And ffs, can we maybe not yell at the people who are trying to post *constructive* feedback on what is broken? "They're probably wrong", sure, but what if they're not? At the very least, knowing what parts are impacted can only help.
lol fyi? I'd just like to point out that this is now the third time Cryptic has tried to push a patch into the game that broke just a significant chunk of stuff that was only partially related to what was patched. I'm beginning to think the name's cursed or something. On the bright side, at least the chat server works this time around.
RIP Caine
I was inclined to think that maybe Cryptic had someone not experienced with their models who made a mistake of slipping a seemingly identical mesh instead of an old mesh.
Especially since one of the things pretty arbitrary for 3d engines is which way normals are pointed, and if you are new to the engine... It's not something obvious at all if Y is + or - on pieces you are working with.
But this would be only once.
If it was the thing, it would be corrected long time ago.
Nobody would be going through other meshes and textures file after file opening and changing them.
And what if old CO mesh has some issues like doubled vertexes or split edges? Those aren't uncommon things at all, professionally made models may have them.
And that would be enough in the long term to go through changing every mesh to a new one. For an example, if mesh is dirty, then it makes rigging harder and you can't (for an example) speed up your work with weighting it automatically through whatever tool or script your 3d program has and then only finish it by hand.
And it means that dirty mesh slows your work on every other costume using it as a base... Because rigging it takes longer.
Could be a reason enough for CO, where models aren't very complicated, but they are multiparted and I suppose that rigging it to skeleton may be the hardest part of making a costume.
Also, a slew of newer costume pieces like Holoforce, Cosmic Knight and Tactical Armor are also using different mesh with slightly different geometry. They also clip with neck cowls (on back).
I was wondering why someone was using a different mesh for them, but maybe it was the case? They found an issue with standard Cryptic meshes and they went with not using it as a base?
Dunno. But it might be reason enough to go through the hassle of tuning everything else after changing base meshes.
You're right, more or less... but given some of the shuffling around, I think they're also dealing with undocumented legacy modifiers on top of everything else, and fixing bad practice can be... complicated. I've had to deal with some pretty nightmarish stuff, things that I looked at and went "Who in their right mind would do something like this?" There's a reason I used Skyrim as an example... a lot of the modding community wouldn't know best practice if it bit them.
(No, I'm not trying to excuse the problem. Explaining why it might have happened does not automatically mean attempting to get out of the consequences.)
@meedacthunist:
I noticed that, too. Maybe they're in the process of standardizing the meshes? Any of the things you mention would be a good reason to start using a cleaner mesh as a base, but... that'd also require replacing the old ones, then testing them all with existing modifiers, etc. Dunno. I don't work there, so I can't say for sure.
"Whoever is careless with the truth in small matters cannot be trusted with important matters." - Albert Einstein
Its the "their probably right" part that these players don't care about. They don't care that they are attacking the people who want it fixed instead directing their issues toward the people who broke it. Or they are just trying to cause chaos and distract from the problem.
Then there's the people who assume that there is more to the problem and claiming that there is possibly no way to understand the issue despite proof that a fix is simple and has been dealt with.
When Gentleman Crush first made himself publicly available to players on CO, long standing issues that have been made to his attention by players (who people claim don't know what they are talking about) were fixed all but immediately. Issues that devs have claimed were impossible to fix.
Something about each issue seems fishy, but most of the problems the game has is Cryptic being greedy and not offering enough development time and giving that development time to devs who know the game. They don't want to give CO players the time of day anymore and that's how this all ends.
Flourishing games require a developer and player base relationship which we don't even meet the bare minimum of and have lacked since the conception of On Alert. Cryptic North and Gentleman Crush's first months being the only exception.
Many times I would trace the evolution of a variable, find how it was treated, and start ranting about what kind of half-educated lunatic would do things like that to a poor inoffensive dataset. Things that should have been variables were hard-coded, and when the data changed they would hard-code another piece. (For instance, weapon yields - they changed significantly over the decades, and the module I have in mind referenced seven different hardcoded yields instead of a single variable that could be filled from the database.)
tl;dr - legacy code can really suck hard.
- David Brin, "Those Eyes"
Get the Forums Enhancement Extension!
As they say...winner...winner...chicken...dinner? Sure, let's go with that. :biggrin:
Gradii: Honestly, they should fix that twice then. Once to clean it up and once to slap some Goku Hair on it for an actual "Bad Hair Day Emoticon." Then everyone wins.
Silicon: It has been said multiple times, likely too much by myself as well, but I agree that one of the major issues here is impractical timetables for PTS and how that data is actually handled. Communication, Testing, Civil Exchanges of Information/Suggestions/Whatever, More Communication, and enough time to make it right so you "Release" onto Live and not "Inflict" upon Live should be the baseline goal in this process. IMO, from current observations, "NEW" Old Cryptic has shown to be better in this department than Old Cryptic but not as good as Cryptic North. I, personally, see them trying though. I see Gadfly making many more posts than Old Cryptic did. I see the efforts but, on the other hand, we all see the current results as well. I see the potential for a proper PTS -> Live Setup...the question is, and always kinda has been from one degree to another, whether potential becomes reality.
Jon: Sure we don't have access to all the files and data....but what about those rare few who can piece together the problem and the actual solution just from what we DO have?
Buxom: Buxom, in your case it's not so much whether you're right or wrong alot of the time that apears to draw so much ire from the playerbase and such on these forums. It's how you present the information. It's not like the actual issues and such are a super hard secret to get from those who know. I would suspect there is little to nothing you know about these particular issues that I don't know, for example...the difference is purely a matter of tact and presentation. Your word choices come accross as much more accusational, brash, and tin foil hatty than pretty much anything I've ever posted on these forums. Consider that for a momment. Try presenting the information as information and not with insulting adjectives, name calling, or malicious plots. A is Broke. I believe A + B will fix the problem. The End.
Also, I only really know of two people on these forums who try to use chaos to distract from the actual problems. One of them is usually half oblivious that they're doing it and "The King" doesn't show up unless he smells a certain level of player sorrow on which to feed. So...yeah...that's 2 people. Not exactly an epidemic.
Crush wasn't the only person to make themselves available ingame for conversations and fix things. Lordgar, Gadfly, Robo, and even Tumerboy(who was an artist but went well beyond that trying to help out) have also performed that task. Also, there's TT, who was/is the single best CR by a country mile that this game has ever seen in the history of ever. When he "passed things up the chain", to borrow a phrase from a lesser CR, stuff actually came back. If I dug through the mountains of screenshots I have on my HD between the office and home office PCs I'm fairly certain I could find at least 1 pic of all 6 of those people standing in the Ren Center and just listening to the players....and even some of the old GMs(Drannic, IndigoFyre, Aand, Hopkins, .....) too, back when we used to have those.
But, as has been said time and time again, the communication levels need ramped up. So...8 words time.
STATE OF THE GAME
ASK CRYPTIC
PRETTY PLEASE*
*Nope, not gonna let this one go.
Join Date: Aug 2009 | Title: Devslayer
Oh, yes. There have been a few too many "push to PTS on Wednesday, push live on Thursday" cycles lately. Listen to player feedback? Often there's no time for player feedback.
You know what? I'm back on that bandwagon. I had given up hope with CNorth in charge, but how about a new broom sweeping clean and all that?
They're not there in the beginning, but when your story ends / Gonna last with you longer than your friends
On a side note: this is pretty much exactly what was going on three years ago when I stopped posting the last time. Justifiably angry players making unjustifiably abusive comments, with a liberal sprinkling of the confidence of ignorance as to how easy fixing software is and absolutely no interest in anyone else's point of view, or in actually reading anything except what they want to see. In other words, internet as usual.
"Whoever is careless with the truth in small matters cannot be trusted with important matters." - Albert Einstein