EA/DICE respond to backlash
... https://www.reddit.com/r/battlefield2042/comments/rx0w9y/eadice_finally_responds_to_the_backlash/?utm_medium=android_app&utm_source=share
Forum Discussion
@Trokey66 wrote:
@AOD_moose004I read an example of how hard it can be to QA software and the example was....
There is a bug with the '9' on a phone where it will not work.
However, it only manifests itself if the two previous presses are '0' and '1'.
As you would appreciate, you can't test every combination of button presses so missing this bug is not impossible.
Unless some one presses '019', the bug may never manifest itself and on the rare occasion that it may occur, if it isn't indeed to the previous button presses, identifying the actual issue could be very difficult if not impossible.
button press combination bugs can easily be checked via automation so that isn't the best example
*edit* in my opinion the best way to do a final QA check is to actually have a sufficiently long beta with the launch build and actually analyze results and pay attention to feedback and do it close enough to launch to use the expected launch build but early enough to actually give yourself time to fix it and if you don't have time, push back the launch *as nintendo is famous for doing* until it's done
@Psubond With 10 digits, even automation may take too long and be cost prohibitive long before the 'magic' combination is hit.
At the end of the day though, it was a simplified example to highlight the difficulties in QAing complex software.
Even an extended Beta may not highlight some/all of the bugs so far encountered.
At the end of the day though, it was a simplified example to highlight the difficulties in QAing complex software.
Even an extended Beta may not highlight some/all of the bugs so far encountered.
@Trokey66 wrote:
@PsubondWith 10 digits, even automation may take too long and be cost prohibitive long before the 'magic' combination is hit.
At the end of the day though, it was a simplified example to highlight the difficulties in QAing complex software.
Even an extended Beta may not highlight some/all of the bugs so far encountered.i never said that it would find ALL of the bugs and i truly hope you are not trying to make the case that because it's hard to QA a game they shouldn't bother.
as for the button testing a human wouldn't be faster and not doing it is unacceptable.
multiple computers running the test to split up the work means it would not take that long. you aren't trying to break 256 bit encryption, you are testing button press combos and you don't need to build a robot that actually pushes buttons you just need to send the button presses to the xbox/ps/pc via an interface instead of a controller (yes, we don't have them but i'm sure the devs do. there is no way they have to have an actual person do everything)
- @Psubond When did I say the shouldn't bother?
Testing complex software is often carried out on a risk based cost/benefit basis depending on the criticality of the specific software.
All your examples come at a cost that may be deemed excessive for the development team.
Featured Places
