IronHorseDeveloper, QA Manager, Technical Support & contributorJoin Date: 2010-05-08Member: 71669Members, Super Administrators, Forum Admins, Forum Moderators, NS2 Developer, NS2 Playtester, Squad Five Blue, Subnautica Playtester, Subnautica PT Lead, Pistachionauts
@mattji104
While it is fixed in the current internal build through applying the hitreg build that was already being worked on for months prior... we still have no idea why.
And yes that's a concern to many of us. We have the tools now though, and are looking into it (impromptu sunday group test in 10 minutes, actually)
The heightened degree of issues were not present in our testing[/i]. Again... even after we knew there was a problem, reproducing those symptoms required certain individuals because it is not present on everyone, and it is not present in every network condition.
Pardon me... The reaction was so like everybody see / have it. Everybody spoke like one man. I don't believe you. Even if it wasn't a item on the test list (if one exists); it should have changed the decision making; leading NOT to make b278 ready to go until more tests are done. Eventually with other people. Eventually excluding from the new build the forked branch where the new code is (probably a GIT stuff).
B277 was kind of OK considering hit reg. I've seen better, I've seen worst. But what i never seen before is that difference between the two. Also the "everybody syndrome". Even if i could only see "a weird" thing on local server. It clearly can't be unnoticed and ignored.
You can test this yourself - 278 is still on the Beta branch, fire up a vanilla server and fill it with at least 12. You'll notice you can reproduce it on some people but not others. Latency and packet loss seem to impact it in some way, but from our testing I can tell you that high latency vs high latency, or low vs low makes no difference. Conversely, high versus low will show NO ISSUES for the high latency player, but the low latency player will experience them.
What ? you have no more playtesters ? Or are they all in the same cyber coffee ? You're kidding right ?
That's what we call edge case, something that is not apparent through our previously typical means of testing. Again... as a result we've adjusted our testing procedures to try to account for future bugs that may be reproduced the same way. So yes, again, it is true that we messed up and this slipped by.. I have no qualms with blame being taken (I dont think anyone does?) but it was a new parameter for us to account for that we've never had to before!
Please... If it's not on the list; one adult would just put it on and give it a priority. This would act as a veto until things are ready. As it can be modular (code , art etc) nothing prevents the code/script changes to be postponed. Or anything else btw.
If you have magic powers and can predict every single future parameter and scenario that should be tested- while still completing everything in a reasonable time - please by all means, we would love to have a psychic on board our team, it would save us a lot of head aches
Yeah i got magic powers. It's called common sense.
I wouldn't make a new build available until I'm sure all questions are answered. Maybe it takes more time. But it wouldn't be the mess it is today. It's a first for me to see a rollback on a game. Yes it's difficult to do so, but it's safer. Right now it looks like always : Things are done in a hurry.
Having a set of "danger flag" for things that may be hurting the game. A prop not aligned on a wall is nothing compared to a bug in the netcode (this term has never been accurate but it's for simplicity sake). So including something in a build would require different stages of validation that I think are missing today. There's probably a guy that say "it's working" and he cross fingers. The usual. Tip: It's always better to postpone instead of what happened.
Be a little more forthcoming and honest in the way the devs would tell us what they are working on, announcing a schedule and a target. If it fails to be delivered on time, the community will be far more comprehensive than what UWE just did. Right now it's "hey a new build came out guys!!!" And everybody has an heart attack. Slower, better. It's ok to have difficulties and postpone when you go in the right direction.
Not working with Trello would be a start... Trello is such a joke in terms of project management
And for the groupies (i got to keep then warm) and my collection of disagree : Fire Hugh!!! Nah joking... Keep him.
All this puts in perspective the notion of "dead line". Why the hell would UWE and CDT try to "launch a build" every month while they probably can't do it considering all those things (time, technical and such) and past experiences ? This is just nonsense.
So, please UWE and CDT stop trying to make something out as soon as you have/make it, take the time to do the right thing.
(Psychic power on) Change guys, chaaaannnge... don't do like always... chaaaannnge, you liiiiiiiike change... This is the word chaaaannnge, and you like it... .... chaaaannnge... no more of the same... chaaaannnge for the greater good. chaaaannnge... chaaaannnge...
(Psychic power off)
It seems to me that @UncleCrunch is saying the CDT team easily noticed this and let it go anyway? I don't think people that are donating their time to make this game better would just let something like this go.
DC_DarklingJoin Date: 2003-07-10Member: 18068Members, Constellation, Squad Five Blue, Squad Five Silver
If you do not believe the answers the CDT/Devs give you, why asking in the first place? It serves no purpose.
Either you believe what they tell you, or you dont. If you do not feel you can believe the answers the 'source' gives you.. well.. just dont ask. really.
I never felt in the past with any CDT patch that they were rushing it. Remember how long players waited on certain fixes or patches in the past? Its clearly shown if they believe its not ready, that they do not want to release.
And lets look at it from the other side of the coin. Lets assume you would be right.
"Hey guys, we have the worst hitreg probably since alpha. Lets release this build. It will make everyone mad and there will come absolute nothing useful out of it."
Does that make sense? No. If they DID notice this insane hitreg in testing I can not fathom anyone would remotely want to release it. It would destroy more, build nothing, and make them in essence waste all the time spend in improving the game.
So if its obvious they do not want to release this willingly, then it has to be done unwillingly. And in such a scenario the most simply and obvious thing is usually the truth. Aka 'It could not be reproduced in playtests'.
You can argue all you want about how protocols should have been used so they would have picked it up, but did Ironhorse not state that they also think so & the protocols have been adjusted?
NarfwakJoin Date: 2002-11-02Member: 5258Members, Super Administrators, Forum Admins, NS1 Playtester, Playtest Lead, Forum Moderators, Constellation, NS2 Playtester, Squad Five Blue, Reinforced - Supporter, Reinforced - Silver, Reinforced - Gold, Reinforced - Diamond, Reinforced - Shadow, Subnautica PT Lead, NS2 Community Developer
@UncleCrunch
I don't know what to tell you that you haven't been told already. It literally did not appear in testing at all prior to pushing the build to live. If you don't want to believe us then that's your own prerogative. As I've said in other threads we would have liked to have gotten more test games in for this release but given that it was considered fairly low-risk at this point by everyone involved we saw no issues in pushing the build as it was. In hindsight that was a mistake, but as we had never come across such a widespread new hitreg issue in over a month of testing prior to last week this all came as quite a surprise to both the programmers and the testers.
As for your suggestions:
We already have a very thorough process that we follow prior to approving a Release Candidate build to be pushed live.
We already have a priority system for addressing bugs and fixes.
I challenge you to find a more transparent development team working on a commercially published multiplayer team shooter. You won't.
Obvious non-starter. We use Trello because it works. We're always open to using new platforms and look at many alternatives as they become available... but given that you haven't even suggested any I can't take this "suggestion" seriously.
Comments
While it is fixed in the current internal build through applying the hitreg build that was already being worked on for months prior... we still have no idea why.
And yes that's a concern to many of us. We have the tools now though, and are looking into it (impromptu sunday group test in 10 minutes, actually)
No, there is no way that could ever happen, I know because I made that change
B277 was kind of OK considering hit reg. I've seen better, I've seen worst. But what i never seen before is that difference between the two. Also the "everybody syndrome". Even if i could only see "a weird" thing on local server. It clearly can't be unnoticed and ignored.
What ? you have no more playtesters ? Or are they all in the same cyber coffee ? You're kidding right ?
Please... If it's not on the list; one adult would just put it on and give it a priority. This would act as a veto until things are ready. As it can be modular (code , art etc) nothing prevents the code/script changes to be postponed. Or anything else btw.
Yeah i got magic powers. It's called common sense.
All this puts in perspective the notion of "dead line". Why the hell would UWE and CDT try to "launch a build" every month while they probably can't do it considering all those things (time, technical and such) and past experiences ? This is just nonsense.
So, please UWE and CDT stop trying to make something out as soon as you have/make it, take the time to do the right thing.
(Psychic power on)
Change guys, chaaaannnge... don't do like always... chaaaannnge, you liiiiiiiike change... This is the word chaaaannnge, and you like it... .... chaaaannnge... no more of the same... chaaaannnge for the greater good. chaaaannnge... chaaaannnge...
(Psychic power off)
Either you believe what they tell you, or you dont. If you do not feel you can believe the answers the 'source' gives you.. well.. just dont ask. really.
I never felt in the past with any CDT patch that they were rushing it. Remember how long players waited on certain fixes or patches in the past? Its clearly shown if they believe its not ready, that they do not want to release.
And lets look at it from the other side of the coin. Lets assume you would be right.
"Hey guys, we have the worst hitreg probably since alpha. Lets release this build. It will make everyone mad and there will come absolute nothing useful out of it."
Does that make sense? No. If they DID notice this insane hitreg in testing I can not fathom anyone would remotely want to release it. It would destroy more, build nothing, and make them in essence waste all the time spend in improving the game.
So if its obvious they do not want to release this willingly, then it has to be done unwillingly. And in such a scenario the most simply and obvious thing is usually the truth. Aka 'It could not be reproduced in playtests'.
You can argue all you want about how protocols should have been used so they would have picked it up, but did Ironhorse not state that they also think so & the protocols have been adjusted?
Will use.
I don't know what to tell you that you haven't been told already. It literally did not appear in testing at all prior to pushing the build to live. If you don't want to believe us then that's your own prerogative. As I've said in other threads we would have liked to have gotten more test games in for this release but given that it was considered fairly low-risk at this point by everyone involved we saw no issues in pushing the build as it was. In hindsight that was a mistake, but as we had never come across such a widespread new hitreg issue in over a month of testing prior to last week this all came as quite a surprise to both the programmers and the testers.
As for your suggestions: