Custom HUD Colour UI - resolution suggestion

Hi Frontier and Elite Community,

This relates to the long standing issue where custom HUD colours impact the UI.

I use very similar colours to ObsidianAnt, as it looks better in the Vive than red. Plus, I just prefer blue than red. Whenever I see video's on youtube, everyone's got their own preferences - some like purple, some like green etc. I know there's a vast amount of people with custom HUD colours but...

The problem is that the graphics config GUI is tied into various aspects - such as the HUD, Radar, game Menus like Livery, Station, Holo-Me and probably a few others. This means all colours are in essence, reading from the same code which causes everything else to be wrong colours.

A fix suggestion is to have a second GUIColour module in the Graphics configuration xml for only HUD (minus radar and menus) which would probably be easiest on dev; and fix all occurrences. Plus it'd allow people to use negative numbers in the GUI without it changing NPC pictures (as it still does with negative numbers).

This is a base game feature suggestion which would also fix the new Holo-me Menu colours. Hope this helps.

Fly safe
 
The problem is that this is not how the GUI and HUD work. If that stuff was easy to separate, the character pictures would never have been an issue.
 
The problem is that this is not how the GUI and HUD work. If that stuff was easy to separate, the character pictures would never have been an issue.
Yes, that was why I suggested a different approach. If Frontier were to change the HUD to be independent then none of the other functions would be affected. A new HUD module alone would by-pass GUIColour in the XML completely.

If that stuff was easy to separate, the character pictures would never have been an issue.
I appreciate what you're saying but as a QA Engineer (Agile Software tester) I have to work with solutions all day and sometimes they can be easier than you might think :). Neither of us can say for sure but it's worth raising the idea.
 
Last edited:
The thing is, customisation was discussed very early on, and I don't think anyone at Frontier thinks it'd be a bad thing. The current implementation, however, is not conducive to that and is pretty much a tolerated hack with little to no direct support (TBH I was surprised to see the workaround put in place to try and keep character pictures intact, but that was probably a matter of two minutes since it's just putting those elements through the inverse of the user-supplied transform). This is not something they'd want to spend any significant time on, since it has a whole bunch of downsides like no guarantee that target colours stay distinguishable.

(I work in software development, and keeping dirty hacks from evolving into even dirtier "features" is kinda my job :p)
 
The thing is, customisation was discussed very early on, and I don't think anyone at Frontier thinks it'd be a bad thing. The current implementation, however, is not conducive to that and is pretty much a tolerated hack with little to no direct support (TBH I was surprised to see the workaround put in place to try and keep character pictures intact, but that was probably a matter of two minutes since it's just putting those elements through the inverse of the user-supplied transform). This is not something they'd want to spend any significant time on, since it has a whole bunch of downsides like no guarantee that target colours stay distinguishable.

(I work in software development, and keeping dirty hacks from evolving into even dirtier "features" is kinda my job :p)

Cool. Thanks again for taking the time to reply mate. Yeah hopefully there's already something planned with full risk analysis is already considered. If you're an agile Dev, I presume your Dev team chat with PMs, BAs and QAs as part of your sprint planning and daily work - unless your a waterfall Dev?

Say if you were the chap at Frontier who'd built it, you'd flag those risks to me, the BA would write the acceptance criteria and risks, then it'd be up to me to risk based and regression test it all - feed back a recommendation to the PM... Then it'd be decided on whether to rebuild / retest / release etc. Fingers crossed, it'd be a non-hacky bolt-on type solution, if any
 
Back
Top Bottom