First Quarter 2019 Roadmap

Future of T3D, are we heading the right way, what's next and what's coming.

Moderator: Steering Committee

4 posts Page 1 of 1
Steering Committee
Steering Committee
Posts: 900
Joined: Tue Feb 03, 2015 9:49 pm
by JeffR » Tue Jan 29, 2019 7:46 am
Hey Guys,

Figure rather than trying to edit the other roadmap thread, can just segment the Roadmap-as-planned into relevent checkpoint threads like this.

So, where are we at?

Well, a keen observer may note that 4.0 did not, in fact, make it out by the end of the new year. Between holidays, sickness and that whole meddlesome "real life" thing, targets were missed or delayed. I did intentionally push a somewhat optimistic roadmap to try and keep nose-to-grindstone, so I'll take the failings of that into account going forward on new roadmaps.

In light of that - and the intended plan to shift to a more rapid update schedule once these big, meaty, foundational changes are rolled in, the idea is to have multiple releases per year. We've gotten better at testing stuff, and have a pretty good RC process at this point, as well as a number of people able to test on various platforms, OS configs and the like.

In the interest of better transparency(and categorization) of goals, figured a more localized breakdown of the roadmap by intended quarter would help focus things more and facilitate the more regular release cadence.

Works getting done, but if people don't know it, then it's still an issue. So I'll definitely work to be better on top of updating the Roadmap.


What got done?
A good bit, actually even if it didn't necessarily feel like it. From the old list
  • Core directory of BaseGame was moduleified
  • The gigantic majority of the PBR work has been locked in, it's down to some hanging-on issues preventing the rollin
  • The RenderPipeline R&D showed solid results, and with some fixes and shore-up is ready to go in post-4.0 when conveinent
  • Work to E/C-ify game classes into GameObjects started and a few classes are already complete or mostly complete
  • Initial testing of modulification of the Tools directory shows it should work fine, and can be done in chunks as we update the tools suit, piecemeal instead of needing to do it all at once
  • The overall behavior of the Asset pipeline and the Browser are ready, just needs a solid testing beatdown to find flaws
  • Assimp integration is almost complete, it looks to only need the bone offseting for animations corrected and it's ready to roll
The Roadmap, Q1 2019
  • PBR finalized and integrated
    • Fix binding issue preventing probes-in-array from rendering properly
  • E/C work finalized and integrated
    • Finish component-ification of major game classes
    • Testing of system to ensure to major issues with use
  • Asset Pipeline finalized and integrated
    • Assimp Integration finalize to ensure consistent art flow(Only animation logic needs to be fixed as far as can tell)
    • Cleanup of code
    • Test to ensure no major issues with the asset workflow arise
  • BaseGame Cleanup finalized and integrated
    • Finish update of options menu
    • Fix outstanding bugs relating to toggling editors
    • Fix remaining crashes with popupmenus
    • Make the default template, remove Full
Immediately following 4.0's launch
  • Lock in initial nu-Project Manager's 1.0 implementation, including Legacy Project Importer
  • Begin using the Asset and E/C stuff to start building out Rapid Prototype Kits, or at minimum, start converting old asset packs to work with the new flow
  • Look into a sort of centralized repository or whitelist for the PM to reference for easy downloading of new assets and RPKs
  • Update Shape and Material editors and convert them to modules
  • Implement Camera-driven render pipeline with dynamic resolution scaling
  • Begin research into overhaul of terrain system(implementation for later build)
Beyond 4.0.1, which would be the *astonishingly* uniquely named 4.0.2 and so on, will emphasize updates to the tool suite to maximize usage of the Asset Pipeline and E/C stuff.

As you can see, definitely walking back the planned targets for the different milestones to be able to push harder for a rapid update schedule. There's a lot everyone wants rolled into the engine, of course and as more contributors are able to toss their hat into the ring, we can duely expand on what's notated in the roadmap, but I'll definitely apply pressure to keep it more segmented and practical, rapid chunks.
(I'm a little sick at the moment, so I feel like some bits were missed, so this may be updated tomorrow after I've slept, but consider this the "core" plan currently)
Posts: 1208
Joined: Sun Feb 08, 2015 1:51 am
by Duion » Tue Jan 29, 2019 8:39 am
How about making it first priority to make everything more transparent and democratic.

All I see is "We decided or I decided..." but I don't know who that is or where it happens, it just happens and I cannot see anything nor influence anything.

Before you do anything you need some kind of planning phase and take into consideration what everyone else is doing or wants to do.

Why not let everyone make suggestions, see what we have and who is developing what and base our Roadmap on that, instead of one of a few people just single handedly deciding it.

Before we had a Trello board with cards of each tasks, I did not like the platform, but the concept was very great, because everyone could see all the tasks and pick one and everyone else could see at all times who was working on what, at the moment nobody can see anything.

If you always want to do everything alone you won't get much done, the secret to leading a successfull company is to not do the work yourself, but organize others to do it for you.

You want to be like the boss and the worker at the same time. Why not divide the tasks up, for example I wanted to improve the art assets since forever, so I would contribute there, but so far my contributions got ignored so I stopped doing anything in that regard.
Posts: 212
Joined: Tue Feb 03, 2015 8:58 pm
by Bloodknight » Tue Jan 29, 2019 1:07 pm
Democratic, lol, this is pretty much a codeword for jack shit getting done, the ultimate democracy is everyone takes their own forks and does their own work, democratic is not people who cant telling the people who do what to do.

People are making suggestions all the time 'We' is often the result of conversations in discord or on forums. Where there is lack of interest in a particular upgrade or choice in how the upgrade should be done then somebody has to be the decision maker, here again, democracy fails if only one solution from many can be applied

YOU abdicate your 'influence' by refusing to engage in the most populated and interactive areas of the community, its just that simple. The rest of the community does its contribute, it engages, it informs and it suggests, it is neither the fault of the community or its 'leadership' if you don't want to engage, nor does it mean nothing is done, just because you don't see it, it is only not transparent because of that, to everyone else its clear as air and plain as day.

As for the Trello suggestion, everyone should sign up for Zenhub, then we have a Github repo with built-in Trello like features that directly links to issues, PRs, commits and documentation, seems like a near perfect solution. I'm pretty sure you cant self assign, but then you couldn't do that in Trello either, least not without admin privileges but you can use the various forums for declaring you are working on feature xyz.

As for art, I don't recall ever seeing any art improvement PRs, or replacement art suggested on the forums, given that the templates/demos are crying out for quality art I don't see how such a thing could have been refused, perhaps you can post some links for the community to review.
Posts: 431
Joined: Tue Feb 03, 2015 9:50 pm
by Azaezel » Tue Jan 29, 2019 4:08 pm
"We" would be the folks that show up and contribute towards completing things, official titles, or no. Period.

As an example, folks may note that further SDL is no longer on the list. It was placed there as an in for folks that wanted to get their feet wet contributing by working with a more familiar codebase. We didn't have many along those lines, so what was submitted will likely be it for a bit, though we'll likely always accept contributions along those lines, just not push for them in the face of lack of interest.
4 posts Page 1 of 1

Who is online

Users browsing this forum: No registered users and 1 guest