[captain_bazza]

In a weird, but wonderful, quirk of Fate, no sooner had I typed the topic heading when I received a phonecall inviting me out to lunch with family...so forgive me if I continue on my return. http://www.elvastower.com/forums/public/style_emoticons/default/sign_thank_you.gif

CB.

[atsf37l]

No! Get back here! :D :wallbash:

[atsf37l]

I will say one thing, I've been running Build 2578 for the last week and it just doesn't get much better than this. Some of the most intense graphics areas of several routes are running better than ever before, either on ORTS or its predecessor. Excellent work, folks.

Now that being said, I don't normally build activities with AI operation and I see lots of posts about issues there, nor have I dealt with signaling and its ramifications. I do know that there are things like semaphores that are not working, at least in ABS mode, so I have to assume there are continuing problems but as far as general operation of trains and the way the simulator operates in conjunction with current hardware I couldn't be happier. :wallbash:

[Genma Saotome]

Well... while Barry wastes his time with totally unimportant things, like a family lunch, we're stuck here wondering what he had in mind.

Taking a guess... other than the unexpected new bug is there a defined list of missing features that are necessary to get to 1.0? What are they? Or is it down to the bug list... and if so, how many are there required for 1.0?

I would hazard a guess that burn-out / life changes demanding time elsewhere is a real risk to further progress too... perfectly understandable.

[wacampbell]

Well I guess we can hijack his thread since he wandered off.

To attract new content and bring existing content up to date we really need a route editor. The old MSTS editor is unbearable and lacking so many features of the commercial alternatives.

Solving the terrain resolution limitations and providing better terrain texturing methods would go a long way towards delivering some unique advantages.

And we could use a couple of enhancements for modelers like environmental shaders, AO and multilayer shaders, normally maps, light maps etc.

... and the list goes on ... I'll stop now.

[captain_bazza]

"I'm baaack!"

Throws a pebble into the pond.....

First, let me thank all those great developers who are contributing to the project, and those who find the bugs. Any remarks I make are not personal, I assure you, I'm too grateful for what you do.

I feel we are at the stage in Stage One, where some serious consideration needs to be given to what remains be done to reach that ever receding (so it seems at times) target of an OR Stable Version #1.0. Some prior discussion has been done on the subject, but no definitive answer came forth.

Maybe we need a hiatus on the development of NEW features so that work can be done on current features to bring them to a stable version. The decision of what constitutes 'stable' is up to discussion by everyone involved, whether active developer, or other contributor.

The cat just parked itself on my typing arm, but enough said by me....what do you all think?

Cheers Bazza

PS Cat's too not happy I'm continuing type like a one armed paperhanger.

B.

[Genma Saotome]

 wacampbell, on 15 October 2014 - 04:15 PM, said:

Well I guess we can hijack his thread since he wandered off.

To attract new content and bring existing content up to date we really need a route editor. The old MSTS editor is unbearable and lacking so many features of the commercial alternatives.

Solving the terrain resolution limitations and providing better terrain texturing methods would go a long way towards delivering some unique advantages.

And we could use a couple of enhancements for modelers like environmental shaders, AO and multilayer shaders, normally maps, light maps etc.

... and the list goes on ... I'll stop now.

You having second thoughts as to what belongs in Phase 1.0-1.99, 2.0-2.99, etc?

[wacampbell]

 Genma Saotome, on 15 October 2014 - 07:10 PM, said:

You having second thoughts as to what belongs in Phase 1.0-1.99, 2.0-2.99, etc?

Not suggesting any change. Just looking forward to these key elements.

[disc]

 wacampbell, on 15 October 2014 - 04:15 PM, said:

To attract new content and bring existing content up to date we really need a route editor.

Leaving .s format to die and replace with a new 3d format would attract new content too, while it seems to be easier to implement than creating a new route editor.

[James Ross]

 captain_bazza, on 15 October 2014 - 05:08 PM, said:

I feel we are at the stage in Stage One, where some serious consideration needs to be given to what remains be done to reach that ever receding (so it seems at times) target of an OR Stable Version #1.0. Some prior discussion has been done on the subject, but no definitive answer came forth.

Maybe we need a hiatus on the development of NEW features so that work can be done on current features to bring them to a stable version. The decision of what constitutes 'stable' is up to discussion by everyone involved, whether active developer, or other contributor.

We have a plan for the versions but it proved problematic stopping people adding new features. We tried that earlier in the year - it was meant as a stability push for 1.0 with bug fixes and only approved new features allowed - but people couldn't keep it up for long.

We have made progress since that previous stability attempt - but there's still some things on the 1.0 list not done (e.g. "Mouse operation for cab controls").

An idea I've mooted before is that only blueprints that are targeted at the next release and bug fixes are allowed on trunk. Anything else (whether it be new features not previously planned or ones planned for a later release) need basic approval before they can be committed to trunk. People could do whatever they linked in private or on personal branches.

I think two things failed last time:

• No target that could be measured or seen - no date or number of unfixed bugs or anything like that.
  
• A lack of a sandbox for people to commit things not approved for 1.0.

For the first, if people are okay with how blueprints have worked so far, it might be worth creating a blueprint for every remaining feature for 1.0 (based on the original plan) and then discussing if we actually need it for 1.0 or it can wait for 1.1, etc., and then people can pick a feature to work on from a clear list (they're shown on the milestone details page).

For the second, there are a number of options; first of all, people can simply work on other things privately, or they can use a distributed version control system and even publish their personal branch - on Launchpad, GitHub or wherever suits. As long as they don't attempt to alter the license (that's not allowed without everyone's consent no matter where the code is), I believe this is totally fine and perhaps should even be encouraged. With some care, we could even move the code to GitHub and let people fork it directly.

 disc, on 16 October 2014 - 12:19 AM, said:

Leaving .s format to die and replace with a new 3d format would attract new content too, while it seems to be easier to implement than creating a new route editor.

Yeah, we have tried having that discussion before. I am in complete agreement that we need a new model format, to complement DDS for textures, but we've not found an obvious choice. Implementing the choice seems like the easy part now. :rolleyes: