So, what _actually_ causes respecs to fail?

[thunderforce]

We had a discussion on Discord today where a GM said that respeccing with earned but untrained levels is "an automatic fail"; it "fails every time". What about VEATs, who always do an untrained respec at 24? Well, they are somehow special.

 

I went onto the beta server and started an energy/invul scrapper; did /levelupxp 11, levelled to 10, respecced. It worked. Reported back to Discord.

 

Manga (Community Helper) suggested /levelupxp was somehow special and this might not apply in normal play; another GM said it "fails with untrained levels in the vast majority of cases".

 

As it happens I have some characters on live with untrained levels. I've respecced an energy/fire scrapper (trained 12, earned 13); a fire/energy dominator (trained 22, earned 27); and an energy/energy blaster (trained 18, earned 19). All three worked first time. Obviously it doesn't fail "every time", but also, if it fails in the "vast majority" of cases, how lucky am I? (For example, if the "vast majority" is 3/4 of the time, the odds of this happening are 1/64).

 

(Also, GMs only know about the cases where it _does_ fail; without knowing how many try it and succeed, how can we know if it's a majority, vast or otherwise?)

 

I'm posting this because it seems increasingly likely to me that at least some of the things we think don't make respecs work are guesswork, perpetuated by confirmation bias and the way that anything that's likely to be true of a character respeccing is likely to be blamed. "Don't be in a busy zone" - most characters are in busy zones. "Don't be in an SG base" - that's where I keep my enhancement storage. "Don't have untrained levels" - why would you train a level when you planned a respec? But because respecs work most of the time, the player changes this thing and, hey presto, it looks like changing the thing fixed the problem.

 

So... it would be very useful if people reading this, when they're doing a respec _anyway_, would first try it in one of the not-recommended ways - busy zone, SG base, uncashed levels, in TF/Flashback mode - and report back here.

Edited January 22, 2021 by thunderforce

[SuperPlyx]

Altho I am guilty of repeating the "Don't be in SG base" multiple times. I always respec in the base and respec a fair amount. I did a respec this morning to move one slot on my farmer.

I have never had a respec fail.

Now I assume it was a "busy" base or something of the sort. I have a very simple base that only I use. 

[thunderforce]

15 minutes ago, SuperPlyx said:

Altho I am guilty of repeating the "Don't be in SG base" multiple times. I always respec in the base and respec a fair amount. I did a respec this morning to move one slot on my farmer.

I have never had a respec fail.

Apropos of which, now respecs don't enforce that the slotting pattern be possible with normal levelling, it would be nice if there was an alternative mode that was less of a pain to move two slots.

 

I wouldn't be surprised if "in SG base" turned out to be another one of these things that's just often where people try respecs because it's convenient.

[Greycat]

1 hour ago, thunderforce said:

"Don't have untrained levels" - why would you train a level when you planned a respec?

See, here's the thing I have with the "untrained levels" one - If this were the trigger, I think we'd see a *lot* more VEATs have failed respecs.

 

Why?

 

At level 24, they're *forced* to respec their powers while leveling. And I know I'm not the only one who's had higher level tiers sitting there waiting. While the forced respec is (to me) annoying and part of why I don't play VEATs often (or just to 10 to get the costume slot, then PL to 24+,) if I'd had even one respec fail at that point with extra levels, it would annoy me enough to remember it.

Edited January 22, 2021 by Greycat

[SwitchFade]

Lord Nemesis.

[kiramon]

Never happened to me and I’ve done it with extra levels not trained 

[thunderforce]

8 hours ago, Greycat said:

See, here's the thing I have with the "untrained levels" one - If this were the trigger, I think we'd see a *lot* more VEATs have failed respecs.

Well, exactly. That was my first objection to the idea that it was an "automatic fail"; it's in the first paragraph of the OP. But the response to that was that the VEAT forced-respecs are special somehow, which seemed unlikely but is not impossible.

[Greycat]

1 hour ago, thunderforce said:

Well, exactly. That was my first objection to the idea that it was an "automatic fail"; it's in the first paragraph of the OP. But the response to that was that the VEAT forced-respecs are special somehow, which seemed unlikely but is not impossible.

Yeah, it's th epoint that sticks in mind.

And other than being a "free" (and forced) respec, no, it's not all that special.

[GM Kaiju]

14 hours ago, thunderforce said:

 

(Also, GMs only know about the cases where it _does_ fail; without knowing how many try it and succeed, how can we know if it's a majority, vast or otherwise?)

I think you are getting hung up on the wording and missing the intent. You are correct, the GM team deals with the occasions Respecs fail, not when they succeed.

 

99% of the time when a Respec fails and a player requests help, the solution is one of the ones mentioned by the GMs you noted.

 

A developer would have to determine what exactly causes a failure, but IMO the common solutions make it a low impact problem.

[thunderforce]

34 minutes ago, GM Kaiju said:

I think you are getting hung up on the wording and missing the intent. You are correct, the GM team deals with the occasions Respecs fail, not when they succeed.

99% of the time when a Respec fails and a player requests help, the solution is one of the ones mentioned by the GMs you noted.

It seems increasingly likely to me that it's not the "solution" at all, it's just that most respecs succeed and so if the player stops whatever they were doing at the time of the first failure, it appears that that solved the problem when in fact it's pure coincidence. We could equally be in a position where when a respec fails, the answer is "turn off your aura", and hey presto, that works. We just happen to have fixated on things which naturally one would expect to be likely to true of a player trying a respec, because that appears to be the common element. In that case the advice would be no better than "wait a bit and try again", even if the intent was good.

 

(That said, I might try the same process in busy zones, in SG bases, in Flashback mode...)

38 minutes ago, GM Kaiju said:

A developer would have to determine what exactly causes a failure

Until they do so, it doesn't seem harmful to collect what facts we can as players. Already the idea that you must train your untrained levels, or even that it is very likely to fail if you don't, has been debunked. This will let better advice be given to other players.

42 minutes ago, GM Kaiju said:

I think you are getting hung up on the wording

I don't think it's unreasonable to suggest the wording should not include things now definitely known to be false.

[Conviction]

4 minutes ago, thunderforce said:

It seems increasingly likely to me that it's not the "solution" at all, it's just that most respecs succeed and so if the player stops whatever they were doing at the time of the first failure, it appears that that solved the problem when in fact it's pure coincidence. We could equally be in a position where when a respec fails, the answer is "turn off your aura", and hey presto, that works. We just happen to have fixated on things which naturally one would expect to be likely to true of a player trying a respec, because that appears to be the common element. In that case the advice would be no better than "wait a bit and try again", even if the intent was good.

 

(That said, I might try the same process in busy zones, in SG bases, in Flashback mode...)

The first thing I would say here is that the two of us (the GM team and yourself) are looking at this issue from two different perspectives. The GM team is not looking for the cause which your tests seem to be focused on. As Kaiju mentioned, that is more of a dev team focus. If we stumble across a solution for the cause or you (generic you) find it, we are more than happy to bring that to their attention. We are looking to solve the immediate issue of a player having a failed respec. Is it possible that just waiting and trying again resolves the issue? Sure, I personally haven't seen that occur. I'm not saying that it hasn't but, I have not had a player comeback to me saying that is what they did and it worked. I have however had them report it worked when they levelled up, or changed zones from a busy one, or left a base. Is that confirmation bias. Absolutely, I don't argue that it is not. But, it is a resolution to the issue at hand which has been confirmed by those impacted to work. Why would we not offer that as the solution based on the information we have at hand?

 

If you wish to test that scenario of just waiting to resolve the failure, feel free. If it is discovered that something is not the cause of the fix we will be happy to no longer suggest it.
 

4 minutes ago, thunderforce said:

Until they do so, it doesn't seem harmful to collect what facts we can as players. Already the idea that you must train your untrained levels, or even that it is very likely to fail if you don't, has been debunked. This will let better advice be given to other players.

I don't think it's unreasonable to suggest the wording should not include things now definitely known to be false.

Has it being debunked? Your two tests and some more confirmations from others proves that? I know what your going to say. It is not always a failure so, yes it has already been debunked. I will say that in the strictest sense that is true. But, some people use hyperbole and it is not always (and I mean that literally) necessary  to point out that always means every time. I don't think it is unreasonable to recognize that fact and accept that as a possibility. I won't speak for the GM that touched that conversation or this thread. I do not know if they meant it as always or as hyperbole. But, to be honest it is irrelevant. As I said before, we will take the feedback into consideration.

 

What I think would be more interesting is to see is that of those who do have failure how many of them are because they have untrained levels, or because they were in a busy area, or they were in an SG. The answer may not be that because you have untrained levels it will always fail. It may be that when it fails it is always because of one of these three possibilities. Or it may not and the devs would probably like to know that either way. But from my perspective as a GM, I'd love to find one solution to always fixes that failure. I haven't found it yet though.

[thunderforce]

14 minutes ago, GM Conviction said:

Is it possible that just waiting and trying again resolves the issue? Sure, I personally haven't seen that occur. I'm not saying that it hasn't but, I have not had a player comeback to me saying that is what they did and it worked.

That seems unremarkable if that's not a thing you advise players to do. Of course, players only come back and say "I did XYZ and it worked" if XYZ is one of the things you advise them to do.

 

Also, it's possible that some of the current remedies actually work and some don't. Perhaps busy zones do matter. I hope to shed some light on the other cases when I have time.

18 minutes ago, GM Conviction said:

If you wish to test that scenario of just waiting to resolve the failure, feel free.

The only thing I can test is assertions of the form "respecs are likely to fail under these circumstances". I don't have a convenient way to produce failed respecs for testing purposes.

 

However, the GM team here is in a position to actually test this on a grand scale. When a player has exactly one of the supposed common problems, have half of all GMs advise people to use the supposed remedy; the other half, advise them to wait a bit. Document the results, being careful not just to rely on impressions of the outcomes. The benefit would be that in the long run players could be given better advice.

20 minutes ago, GM Conviction said:

Has it being debunked?

Yes, not just "always" (although I submit that GMs should not make "hyperbolic" statements that an uninformed player would take as simple statements of fact; with no prior knowledge, anyone would have taken the statement that it fails "every time" as literally meaning just that) but, as I said, that it is "very likely" seems to be debunked. If the odds of a failure with uncashed levels are 90% - which sounds like "very likely" to me - the odds of my three consecutive respecs all working are one in a thousand (even if we suppose for the sake of argument that using /levelupxp is special and that VEAT forced respecs are special, so the successes there don't count against the idea that uncashed levels matter). The simpler explanation is that in fact respecs with untrained levels are not "very likely" to fail.

24 minutes ago, GM Conviction said:

What I think would be more interesting is to see is that of those who do have failure how many of them are because they have untrained levels, or because they were in a busy area, or they were in an SG.

"Because" is an awkward word here because it is very hard for either of us to demonstrate a causal relationship. All we can do is try and demonstrate a correlation, or the lack thereof.

 

I think it is instructive to consider this: if my hypothesis is correct, if untrained levels have nothing to do with it, but because most respecs work, if you tell a player to train their levels they usually come back and say it works - how would your observations be different from reality? I submit that in fact they would not be different.

[golstat2003]

12 minutes ago, thunderforce said:

 

 

However, the GM team here is in a position to actually test this on a grand scale. When a player has exactly one of the supposed common problems, have half of all GMs advise people to use the supposed remedy; the other half, advise them to wait a bit. Document the results, being careful not just to rely on impressions of the outcomes. The benefit would be that in the long run players could be given better advice.

 

This . ..  actually doesn't seem realistic for all the GMS to drop whatever they are doing and try this. It would be more useful for them to have their QA/BA team try this (do they even have one)? or the devs who could look at the underlying code while trying this.

 

EDIT: Might be a good test case for the next large scale beta or closed beta test group, whenever the next Page is ready for testing.

 

I say all this as someone who works in QA in IT for a living.

Edited January 22, 2021 by golstat2003

[tidge]

A quick on-point comment about 'failed respecs', and then an observation about a different type of failure (that is given a similar 'avoidance solution')

 

The quickie observation: I've only ever had failed respecs messages when I had levels I had not yet trained into.

 

A completely unrelated 'problem' I've encountered is the 'sometimes the catalyst doesn't apply to an enhancement'. Perhaps the most common 'avoidance solution' I've seen on this topic is 'Don't apply catalysts in your base'. (sound familiar to the respec issue yet?) Anyhow:

 

EVERY time I've had a catalyst fail to 'take' it is because I was in a SG base and I had storage device window open (AFAIK, it has always been a storage rack) and the Catalyst went back into the storage rack instead of being applied to whatever Enhancement. The UI is resolving things, just not in the way that was expected. (I believe this is why sometimes you get another error/warning in a SG base when messing around with open Enhancement tables, even though you don't intend to be using the table.)

 

Is it possible that (aside from the untrained levels) that some of the failures being observed are due to having an unclosed/unresolved UI window?

[ForeverLaxx]

I just look at it from the angle one does when giving/receiving tech support. How many times are you asked to make sure the device is plugged in? How many times are you asked to just reset the device? How many times are asked to close the program and re-launch it? 85% of the problems brought up to tech support are solved by those answers even if the underlying problem that caused the error was never addressed.

 

"Train your levels" and "don't be in an SG base" are the equivalent to "turn it off and on again". Whether it's a true solution or not, it fixes, at least temporarily, whatever problem was occurring. So why shouldn't they be suggested as solutions to get players back in the game having fun?

[golstat2003]

24 minutes ago, ForeverLaxx said:

I just look at it from the angle one does when giving/receiving tech support. How many times are you asked to make sure the device is plugged in? How many times are you asked to just reset the device? How many times are asked to close the program and re-launch it? 85% of the problems brought up to tech support are solved by those answers even if the underlying problem that caused the error was never addressed.

 

"Train your levels" and "don't be in an SG base" are the equivalent to "turn it off and on again". Whether it's a true solution or not, it fixes, at least temporarily, whatever problem was occurring. So why shouldn't they be suggested as solutions to get players back in the game having fun?

Exactly, it's not really the GMs job to solve bugs, there is a dev and QA team for that. Their primary focus should be getting players back into the game and having fun. They can pass on what they learn to another team that can do that actual SQL or code research to track down the underlying issue.

[Conviction]

2 hours ago, thunderforce said:

That seems unremarkable if that's not a thing you advise players to do. Of course, players only come back and say "I did XYZ and it worked" if XYZ is one of the things you advise them to do.

I straight up said that there is confirmation bias at work here. Why are you pushing this point? Again, we are providing solutions we know to have worked. That will not change until we find out something does not.

 

Quote

 

Also, it's possible that some of the current remedies actually work and some don't. Perhaps busy zones do matter. I hope to shed some light on the other cases when I have time.

The only thing I can test is assertions of the form "respecs are likely to fail under these circumstances". I don't have a convenient way to produce failed respecs for testing purposes.

Sure it is possible that some work and some don't. We do know that some combination of these three do work. As for testing busy zone, thank you for doing so. I'm sure that will further help the devs in identifying the cause. As for testing it under failed respec circumstances, neither do we. If we did, we would know the root cause and this conversation would be moot. We would simply say don't do X until it was addressed by the devs.

Quote

 

However, the GM team here is in a position to actually test this on a grand scale. When a player has exactly one of the supposed common problems, have half of all GMs advise people to use the supposed remedy; the other half, advise them to wait a bit. Document the results, being careful not just to rely on impressions of the outcomes. The benefit would be that in the long run players could be given better advice.

Ideally, sure. In reality, we are not in the position to do this. We do not have the resources as a fully volunteer team to do so even if that would benefit all of us in the long run. Especially since we do have a workable, if not ideal solution, to fix the circumstances.

Quote

Yes, not just "always" (although I submit that GMs should not make "hyperbolic" statements that an uninformed player would take as simple statements of fact; with no prior knowledge, anyone would have taken the statement that it fails "every time" as literally meaning just that) but, as I said, that it is "very likely" seems to be debunked. If the odds of a failure with uncashed levels are 90% - which sounds like "very likely" to me - the odds of my three consecutive respecs all working are one in a thousand (even if we suppose for the sake of argument that using /levelupxp is special and that VEAT forced respecs are special, so the successes there don't count against the idea that uncashed levels matter). The simpler explanation is that in fact respecs with untrained levels are not "very likely" to fail.

Not going to address this again.

Quote

"Because" is an awkward word here because it is very hard for either of us to demonstrate a causal relationship. All we can do is try and demonstrate a correlation, or the lack thereof.

You are far too tied up in word usage to be honest. That statement was a hypothetical with a "how many" prior to the because. I'd like to know the answer to what fixes it. If it is any of those or not. The answer to the situation may well be zero for all and it does not change the statement that I would find that interesting. 

Quote

 

I think it is instructive to consider this: if my hypothesis is correct, if untrained levels have nothing to do with it, but because most respecs work, if you tell a player to train their levels they usually come back and say it works - how would your observations be different from reality? I submit that in fact they would not be different.

If your hypothesis is correct and untrained levels has nothing to do with it, then we would stop suggesting it; until that time where that is proven to be the case, there is no harm in it being part of our presented solution.

Edited January 22, 2021 by GM Conviction

[DoctorDitko]

For what it's worth, quitting the game ("turning it off and back on") has fixed every failed respec I've encountered.

[BelleSorciere]

I attempted and failed a respec with untrained levels the other day. This was the only time I've ever had one fail, and that includes in busy zones.

 

I've never tried in a base.

Edited January 22, 2021 by BelleSorciere

[Patti]

What if players used the test server to test this for the GMs?

[thunderforce]

58 minutes ago, Patti said:

What if players used the test server to test this for the GMs?

I did. The response was largely a suggestion that /levelupxp might work differently, and also the realisation that respecs are cheap and I might as well do it on live.

[thunderforce]

5 hours ago, GM Conviction said:

I straight up said that there is confirmation bias at work here. Why are you pushing this point? Again, we are providing solutions we know to have worked.

Very probably you are not. If you have a cold, and I tell you to rub honey on your nose for a week, and you get better, the honey isn't a "solution" to the cold. The cold went away anyway.

5 hours ago, GM Conviction said:

Sure it is possible that some work and some don't. We do know that some combination of these three do work.

You don't know that. It is quite possible that all three of those achieve nothing, but the second attempt just tends to work anyway.

5 hours ago, GM Conviction said:

In reality, we are not in the position to do this. We do not have the resources as a fully volunteer team to do so even if that would benefit all of us in the long run.

It requires GMs to record one bit of information per failed respec ticket at the time they deal with it in the normal way (not to "drop whatever they are doing", as someone else said upthread). This is not an onerous task.

5 hours ago, GM Conviction said:

Not going to address this again.

"This" is the point that the idea that untrained levels are likely to cause a failed respec has largely been debunked. But "this" is the point of this thread.

5 hours ago, GM Conviction said:

f your hypothesis is correct and untrained levels has nothing to do with it, then we would stop suggesting it; until that time where that is proven to be the case, there is no harm in it being part of our presented solution.

The harm is that players are given a small amount of pointless makework after a failed respec, but more importantly that we're less likely to get to the bottom of what (if anything) _does_ cause respecs to fail if we allow a mythology to build up supported by confirmation bias.

Edited January 23, 2021 by thunderforce

[thunderforce]

6 hours ago, tidge said:

Is it possible that (aside from the untrained levels) that some of the failures being observed are due to having an unclosed/unresolved UI window?

Let me add that to my list. (That's also an excellent example of the sort of thing that would naturally tend not to be the case on a second attempt, giving the impression that leaving the base - or whatever - cured the problem).

6 hours ago, ForeverLaxx said:

"Train your levels" and "don't be in an SG base" are the equivalent to "turn it off and on again". Whether it's a true solution or not, it fixes, at least temporarily, whatever problem was occurring. So why shouldn't they be suggested as solutions to get players back in the game having fun?

I feel that this is largely a response to something I'm not actually trying to talk about in this thread. Some responses read as if I'm making an extended criticism of GM behaviour, where the only thing I'm saying about what I think the GMs should do right now is "don't say things that are known to be false", which seems uncontroversial. (ETA: maybe I'm also saying they should test it? IDK that I am. I'm saying they could test it, in response to a suggestion that I test something which manifestly I can't.)

 

What I want to talk about is - what do we actually know about what causes respecs to fail, and what is just accreted guesswork? I'm getting tangled up in the other discussion because in the Discord conversation, the GMs suggested they were giving this advice because it is well known that respecs fail if it's not taken, and it's useful to ask if there's any actual reason to think that's so.

Edited January 23, 2021 by thunderforce

[thunderforce]

Well, it's been a while, but today I tried "in a busy area" - ie, Excelsior, next to Ms Liberty, in the evening on Saturday. Four respecs in a row all succeeded. Hence, I suspect "busy area" is as much of a red herring as "untrained levels".

[Mr. Vee]

I'll buy that untrained levels doesn't cause a failure every time, but it's not going to stop that from being the first thing I ask when someone in help says their respec fails. Mainly because every single time I've asked they did in fact have untrained levels. Sure it'd be nice to know for sure the exact failure conditions but for my purposes I just pick random stuff until I'm fully trained before respeccing.