Update 17 Release Notes

Status
Thread Closed: Not open for further replies.
Now if they could fix the 90 degree rotation of the docking port hologram when coming out of supercruise...

I'm very happy with the behavior of the std DC alignment
 
What about the engineered missile launchers randomly not firing? It happened to us as well... and a hauler kaboomed because we tried to kill drives with other weapons.
 
@sallymorganmoore how you doing? Is anyone else having problems with the transfer all to Carrier?
Not working at all.

Thing is I tried transferring Tritium yesterday and I'm sure I lost a full load (784 ton) but I don't have proof. Today, I can't do anything with it.

UPDATE: If I relog to Solo it worked, I was in Open.
 

Attachments

  • 20231027154726_1.jpg
    20231027154726_1.jpg
    769.5 KB · Views: 49
Last edited:
It shouldn't surprise anyone with user support experience that most people don't use things like the issue tracker, even if they complain in public. And I understand that this circumstance is very demotivating to write anything in the issue tracker at all, I've seen this effect already in corporate environments.
Even for very easily reproducible issues there would have to be regular advertising, which rarely happens here in the forum, on reddit or wherever. After all not everyone reads everywhere, so where to announce an issue to voters?
But as we're talking about such things here, here are two issues as examples that are easy to reproduce within minutes and still need confirmation:
Don't get me wrong, I totally understand the idea of confirming an issue on basis of user votes. But I don't think that a quick and easy reproducible issue needs to be confirmed by players. This unnecessarily leaves players alone with obvious bugs for a too long time.

Here are other examples of annoying but not that easy to reproduce issues (especially the second one):

@sallymorganmoore : Maybe it would be possible to pre-test issues @FDev? That would need a person with skills just to play the game.
The task would be something like try to confirm a new issue for not more than five minutes and if it's not confirmable within this time, leave the confirmation to the issue tracker.
There are by far not enough issues that would make an full-time job out of this! ;)
 
It shouldn't surprise anyone with user support experience that most people don't use things like the issue tracker, even if they complain in public. And I understand that this circumstance is very demotivating to write anything in the issue tracker at all, I've seen this effect already in corporate environments.
Even for very easily reproducible issues there would have to be regular advertising, which rarely happens here in the forum, on reddit or wherever. After all not everyone reads everywhere, so where to announce an issue to voters?
But as we're talking about such things here, here are two issues as examples that are easy to reproduce within minutes and still need confirmation:
Don't get me wrong, I totally understand the idea of confirming an issue on basis of user votes. But I don't think that a quick and easy reproducible issue needs to be confirmed by players. This unnecessarily leaves players alone with obvious bugs for a too long time.

Here are other examples of annoying but not that easy to reproduce issues (especially the second one):

@sallymorganmoore : Maybe it would be possible to pre-test issues @FDev? That would need a person with skills just to play the game.
The task would be something like try to confirm a new issue for not more than five minutes and if it's not confirmable within this time, leave the confirmation to the issue tracker.
There are by far not enough issues that would make an full-time job out of this! ;)
Thank you for this. I will look into them these evening and leave a comment.

Here's one about bad UI that might have you sell your fully engineered corvette instead of switching to or transferring it:
https://forums.frontier.co.uk/threads/accidentally-sold-ship.619193 / https://issues.frontierstore.net/issue-detail/61590

Important: "Confirming" does NOT mean FDev is looking into it already! 🥸 It means Fdev waits for more users 🤠👽😈🤡 to actively LOG IN, confirm and leave a comment. Otherwise the bug will expire and be ignored! 👻

The threshold is ten confirmations.

You can only confirm and comment when LOGGED IN with your frontier account on the issue tracker page. So dewit! 🚀
:coffee:
 
Another thing about the issue tracker is that it needs those 10 confirmations on a single bug report - spread out across duplicates doesn't work - and it doesn't make it easy to search for duplicates for anything which might be described in multiple ways (or needs a multi-word phrase to accurately describe). There are quite a few bugs which have been reported far more than 10 times, and the tracker has successfully hidden them from Frontier.

Fortunately, two of the issues you found have duplicates already past that stage on the tracker.
But as we're talking about such things here, here are two issues as examples that are easy to reproduce within minutes and still need confirmation:
Looks like a duplicate of the Acknowledged https://issues.frontierstore.net/issue-detail/62025 (with slightly more information about what's going on, and the right version number)
https://issues.frontierstore.net/issue-detail/60972

Duplicate of Acknowledged https://issues.frontierstore.net/issue-detail/62015
 
I didn't wanted to write mainly about the issues themselves, but rather make a suggestion for dividing issues into two classes which can be accepted differently. If one or the other issue gets some attention on the way, good! But that wasn't my point, that's just been examples.
I'm aware of different states, thresholds and problems within the issue tracker. And I think I know why FDev introduced this kind of filtering, as we all do: It's about protecting developers from work that tends to be much effort for a small gain. But it's also not helpful for developers to only be able to address bugs when the user base is already frustrated. That frustrates the devs, too.
I think pre-selecting issues that effect many players and are easy and fast reproducible can be a win for all sides. This doesn't even needs a different process, someone who looks at new bugs can pick them out of the issue tracker and - provided with the necessary rights - just confirm them there. This way, the State of the issue gets changed quickly and it is ready to be added... to some backlog, probably, but that's already a fast step further. And the issues that took or would take longer to test than the suggested 5 mins just stay the way they were in the issue tracker.
To be clear: By "confirm" I mean the administrative action to change the state of an issue from "Confirming" to "Acknowledged". Sorry if my choice of words caused confusion.
 
I gave up on ever reporting issues because they simply refuse to fix anything. To this very day, the colors on the Vintage Suits are wrong. They refuse to fix it. Also, the Salvage Ship Kit on the Python has been broken since day one. They refuse to fix that as well. These are two simple things that I paid money for, and I assume others did as well, that they simply refuse to fix. They know about it. They have for a year now. They simply don't care.

Exhibit A:
salvage ship kit flawed.png


Soon. 4 updates and 9 months later, it's still broken.

The broken colors on the Vintage suit kit was reported before that. It too remains unfixed.
 
I think pre-selecting issues that effect many players and are easy and fast reproducible can be a win for all sides.
They already do that - they just have an extremely strict definition of "many" compared with what players think it should be. That's why the confirmation threshold is 10 in the first place for Elite Dangerous, rather than the 3 required for all their other games. (And similarly, enough votes that would stick a bug right to 5 bars in most of their games won't even get you out of 1 bar in Elite Dangerous)

Anything actually critical or otherwise really obvious - like the scanning bug, like the mission board UI glitch, etc. - will get its ten confirmations on a single report (and fifty others scattered over the rest of the bug tracker) without trouble. And if it somehow didn't, the multitude of forum/reddit/etc. complaints would mean they found out anyway.

Something like "throttle gets zeroed if you go to the FC screen" - it's certainly trivial to reproduce, but it only affects the minority of players who own FCs, and not all of them will be managing their FC in flight regularly enough to really notice it. That's exactly the sort of "yeah, live with it" bug that the bug tracker is designed to discourage reports of and discard any reports people misguidedly submit.

Similarly "gravity too high for mission destination" - there aren't that many landable worlds above 2.75G, so it won't affect many people and it won't affect most of them more than once. Yes, it's easy to reproduce (if it even needs that: a quick confirmation that "no, the mission template doesn't check for that" would likely be quicker than trying to reproduce it in-game) - but it's not important enough to fix. So you shouldn't be reporting that sort of thing in the first place.

If they wanted more bugs to get to Confirmed (i.e. the point where someone on their team looks at them to try to reproduce and move to Acknowledged) they could just reduce the confirmation threshold, or improve the duplicate handling, or fix some of the bugs in the issue tracker itself. Is the current threshold perfect? No - occasionally official in-game events get derailed by a "well-known" bug which the higher threshold had hidden from them, and the whole process generates player discontent that a better-disguised triage mechanism could avoid. But it serves to keep the rate at which bugs reach Confirmed and the rate at which they can fix them somewhat in step (there's been roughly 150-175 open Confirmed bugs for years now) so there's no point in making it easier for people to submit them: either it's important enough to get there organically, or it's important enough to one player that they call in a favour from "Ten alts Tommy" to get it over the line, and that seems to keep things in balance.

(I used to be more annoyed about this situation, but nowadays I'm happy not reporting bugs, and Frontier are happy not reading the reports, so it works out fine.)
 
Something like "throttle gets zeroed if you go to the FC screen" - it's certainly trivial to reproduce, but it only affects the minority of players who own FCs, and not all of them will be managing their FC in flight regularly enough to really notice it. That's exactly the sort of "yeah, live with it" bug that the bug tracker is designed to discourage reports of and discard any reports people misguidedly submit.
Thank you for elaborating on how you don't want to help players nor developers to make the game better. :rolleyes: The issue tracker for a fact adds a layer of frustration. You have to put some effort into writing a bug report, then it just "expires" - how dumb is that?

Anyway, there should be ten carrier owners who are affected by this and can just confirm. Same for the other mentioned bugs or nuisances.
 
Thank you for elaborating on how you don't want to help players nor developers to make the game better
I spent quite some time in the Odyssey Alpha doing exactly the sort of organisation of confirmations you're trying to encourage. I spent quite some time before that filing bug reports and organising confirmation parties for various "important to some people including me" issues, many of which did get fixed. Sure, it's not that hard to get ten confirmations for a single bug you choose. But there are thousands of valid bug reports (especially if you count the expired ones) in the tracker, and it becomes a substantial amount of effort to track which ones need confirmations, which are getting close, which are on 11 confirmations and still haven't ticked over to Confirmed, etc.

Frontier could hire an extra QA person to do the triage instead; Frontier could use one of the many off-the-shelf trackers rather than writing their own with its own severe bugs included - they don't. If they're not going to put the effort in - after years of forum threads complaining about the obvious process problems - then nor am I. The bug tracker is not how the game is made better: this is now clear.

Still, far be it from me to discourage someone who hasn't been burnt out yet! If you want to encourage sharing of confirmations then I'd recommend resurrecting this thread with the issues you're interested in: https://forums.frontier.co.uk/threa...ing-due-to-lack-of-votes-confirmation.596654/ - it gets brought back every so often and plenty of people (some with lots of alts) do watch it and confirm bugs which are publicised there.
 
Soooo… I guess I should add the issue that’s been raised about AX reactivations that have been completed all failing, if you fail one that you started* to that bunch, then?

It sure is rather aggravating. But I only see one contribution added to it so far, which is mine(or I have, because this is apparently the design, missed any other reports that might be more ‘popular’…).

*Because you got caught up in the middle of Revenant hell, with a Banshee to sandwich you in good and proper on top.
 
Why is there a limit on the amount of bugs a player can confirm?

Surely once it's confirmed they should let your vote be freed up?
Nope, you have to vote to show that your issue is more deserving of being fixed... except for the times when an issue sits at most voted for months and then just gets closed as "won't fix".
 
Why is there a limit on the amount of bugs a player can confirm?

Surely once it's confirmed they should let your vote be freed up?
Votes are only used on bugs which have already been confirmed, to move them to "still confirmed, but with more bars". The number of bars only seems marginally useful in terms of chances of being addressed.

There's no limit to the number of Confirming bugs which you can add a "can reproduce" confirmation to.
 
Why is there a limit on the amount of bugs a player can confirm?

Surely once it's confirmed they should let your vote be freed up?
I think the voting limit is to discourage people supporting reports that they only know about from the forums.

Like me voting on a bug that only affects Power Play when I have never touched Power Play in seven years of play.

The thing is if they weren’t restricted people would just upvote everything which is fairly pointless.
 
Status
Thread Closed: Not open for further replies.
Top Bottom