I always use Frontline's own dropship, never my own.
Proposal Discussion Bugs & Issue Tracker
- Thread starter rootsrat
- Start date
Wasn't that over Christmas though? No-one expects them to cancel their Christmas holidays.
Well, it was reported and acknowledged by FD quite some time beforehand. I linked the issue above. I do appreciate the Christmas period though.
Who says that there is a skeleton crew working on Elite? The (frankly, several years old by now) number given was ~100 Devs for Elite, if I recall correctly. Okay, I think they said something like "about 100 people work on Elite" and folks in the forums were immediately joking that the janitor and the cleaning ladies are included in this number.
Still, considering that even the CM team already consists of four members, I don't believe that there is only a "skeleton crew" working on Elite right now.
Personally, i think the problem is that it's a bug AND issue tracker and it's muddied the waters a fair bit.
A bug is generally a mechanics thing but when you add issues, you open the door to complaints which takes the focus away from the bugs.
A recent example would be when AWS had a fair few problems in December, which went away, came back, interfered with many things... etc.(issue)
These aren't in-game bugs though so having these two things together is detracting from actual bugs being fixed. It works for neither.
I think they need a dedicated bug tracker and an issue tracker(would need work and rules tbh).
Just my thoughts on it.
It seems a bit messy as it stands.
A bug is generally a mechanics thing but when you add issues, you open the door to complaints which takes the focus away from the bugs.
A recent example would be when AWS had a fair few problems in December, which went away, came back, interfered with many things... etc.(issue)
These aren't in-game bugs though so having these two things together is detracting from actual bugs being fixed. It works for neither.
I think they need a dedicated bug tracker and an issue tracker(would need work and rules tbh).
Just my thoughts on it.
It seems a bit messy as it stands.
As a coder, i disagree. That bug is almost certainly caused by some C memory safety issue(s). Those might A) cause malfunctions during gameplay already, and B) are quite detectable by using a decent compiler with warnings turned up (and/or dedicated tools). The typical reason for not fixing them is that your code base is so horrible that getting it warning/tool-clean is seen as impossible and abandoned. And then any new code you write will also contain new bugs that the compiler would've found, except you turned the warnings off, because there's so many of them. And then you spend days debugging one of those avoidable bugs, because that one happens to cause some critical issue. Say, mission boards no longer loading.2) The game sometimes hangs on exit. Not only is this probably hard to track down (there are some things which seem to guarantee it, but they're such bizarre things that I can't begin to guess why they should) it's also a pretty minor inconvenience in practice with an easy workaround, and doesn't technically affect gameplay at all. And it's the second-highest non-recent bugPredates Odyssey but it'd be really hard to argue - despite its votes - that this is what Frontier should drop everything to fix.
So you absolutely must take some time to clean up your code base every now and then, or you'll be wasting far more time putting out fires. And that hang-on-exit is a fairly good indication that the ED coders don't get to do that.
It''s usually management that doesn't understand this relation, and demands "work on features, we'll fix the bugs later." But the only way to stay on top of bugs is to not introduce most of them in the first place. Thus I voted for that bug as a way to give the ED coders a lever to tell management "we need to get our memory safety issues under control", and hopefully get some time to improve their code base. Well, one can dream...
Last comment was four days ago, neither I nor anyone I play with has experienced that problem since then and the forum thread has died... is it not resolved for you?Jumping on the thread to give this as an example:
Issue Tracker
issues.frontierstore.net
69 pages, acknowledged by FD but no resolution.
The old forum used to have that, basically it was used as a toilet / rant den / salt spreader, place to diss the Devs etc.
Literally anything but to report actual bugs.
The issue tracker isn't without flaws but it does provide a structured (sort of) method of getting something reported it. It needs work though and perhaps an input mechanism from FDev to say if the bug is being worked on/not being worked on and/or if a fix is incoming.
Generally it means that the traffic is exceeding the capacity of the Amazon servers the game does it's background stuff on in fact, more often than not many of the bugs occur over weekends & holidays due to it. They then disappear through the week only to return again the next weekend when traffic ramps up once more.
The thing that grinds me the most is, the most common way to get attention on a bug seems to be complaining on the forums and tagging a CM (usually Sally) in them... which shows there's something horribly broken with the system.
Reason I simply cannot be bothered putting the effort in to report bugs anymore is this:
For those playing at home... that's three distinct bug reports for the same issue flagged as "expired", followed by one triggered by this, marked as "FIXED"... no comms as to what was fixed, how, or anything.
Surprise surprise, it WASN'T FIXED, thus, the next report above it about Orthrus sampling (which is exactly the same bug, which was never fixed). And no, before anyone familiar with the bug in question asks, this isn't an indiciation it's "Working as intended"... there's a status update for that called "By Design", and if this was by design, it wouldn't have caused the cancellation of two CGs. And I don't care how "minor" or "niche" tissue sampling might be in the eyes of some, getting two CGs cancelled means this is a major issue as far as I'm concerned.
This issue has been extant for over five years now, dating back to the 5th Aegis Initiative. If issues which cause significant disruption like this cannot be resolved in 5 years due to, among other things, a god-awful issue report system... reporting issues is just an utter waste of time.
Reason I simply cannot be bothered putting the effort in to report bugs anymore is this:
For those playing at home... that's three distinct bug reports for the same issue flagged as "expired", followed by one triggered by this, marked as "FIXED"... no comms as to what was fixed, how, or anything.
Surprise surprise, it WASN'T FIXED, thus, the next report above it about Orthrus sampling (which is exactly the same bug, which was never fixed). And no, before anyone familiar with the bug in question asks, this isn't an indiciation it's "Working as intended"... there's a status update for that called "By Design", and if this was by design, it wouldn't have caused the cancellation of two CGs. And I don't care how "minor" or "niche" tissue sampling might be in the eyes of some, getting two CGs cancelled means this is a major issue as far as I'm concerned.
This issue has been extant for over five years now, dating back to the 5th Aegis Initiative. If issues which cause significant disruption like this cannot be resolved in 5 years due to, among other things, a god-awful issue report system... reporting issues is just an utter waste of time.
It is working fine fior me... now.
I'm not sure about other players but I'm thinking it may well have been/is a server side issue that is now resolved. I have no idea.
It's a possibility ... but I'm not sure it's the most likely one.
Neither of the two things known to trigger the shutdown hang nowadays [1, 2] seem to cause any problems whatsoever in gameplay before that point - I've tripped both repeatedly, and never noticed anything out of the ordinary until it failed to exit cleanly. So if it is a memory corruption [3], it's one localised so much as to be apparently harmless. And the game itself seems not prone to memory corruption in general - I don't see more issues after playing for a whole evening compared with the first ten minutes
[1] When first reported, it seemed more common and likely to occur randomly. Seemed to happen on every exit for a month, once.
[2] I say "nowadays". I haven't seen it personally for a long time, and the last comment on the bug report was in October after comments being previously added to the bug most months. So it's entirely possible that it's been fixed anyway. Can't be bothered to try deliberately causing it...
[3] Some unresolvable thread deadlock which only shows up when it's time to clean everything up is my preferred
To me it's always felt like a socket closing bug. Windows has some unique behaviour if you close a socket without doing a manual shutdown or adjusting the linger options, that can block the close call until the remote end responds (with ACK/FIN or RST)... which isn't always guaranteed to happen.
Ozric
Volunteer Moderator
Not sure I really agree with the first bullet point, but the rest yes
Part of the problem is that a huge chunk of the top voted issues that were "fixed" were either not bugs, or are still actually broken and just marked as fixed. The Pulse Wave Analyzer for example is still broken.
As I said in my thread I created at the start of 2021. We were initially told that the issue tracker wasn't going to be used to fix issues that had been pushed by certain groups. Then we were later told that they would only fix the issues we voted on.
I proved this was the case by taking a bug that was close to my heart, smuggling. Since December 2018 goods sold on the Black market had been getting an incorrect modifier applied to the sale price. Many attempts had been tried to get it fixed, especially as it's one of the advertised roles for selling the game, but all failed. The issue had sat confirmed and languishing on the tracker for about 18 months, until I decided to prove a point and force the issue. I basically got everyone I knew, and loads of people they knew to sacrifice one vote (per account they have, cause the tracker rewards you for having multiple game accounts) and push this issue up https://issues.frontierstore.net/issue-detail/12357
It ended up being the fifth highest issue on the tracker, and the next point update it was fixed. So I had taken a 3 year old issue, and got it fixed within 5 weeks, just because I managed to drum up enough support.
That is not right.
*obviously this was all in the golden hey day of the Issue Tracker, which was part of 2021. The issue tracker hasn't been discussed by the team since the start of 2022.
Do you have a copy of that? - it's certainly not how they ever operated in any actual release, so if it was policy it must have lasted about a week, and I don't recall them ever saying anything like that myself.
"Votes are an input to determinations of priority, but not the only one or even the most important one" seems to have been fairly consistent in terms of what has actually been fixed over the four years since they introduced it.
Your experiment I think is actually a pretty good one for showing importance: if a 3-year-old bug that didn't appear that severe gets a massive boost in votes suddenly, well, maybe something else has changed elsewhere which means more people are experiencing it and it should get a higher priority for fixing. Certainly it's gameable [1], but the numbers involved make it more effort than most people actually go to, so it's still mostly usable as a measure.
[1] The top-20 bugs articles probably encouraged more of that, which combined with them not actually providing much useful information to players is probably why they stopped doing them in that form.