New Orbiter

[psychotix]

Sorry if this is in the wrong section but I was just wondering about the progress on the new version of orbiter. I've seen the new planet textures but I was wondering when it will be coming out and what it will feature.

 

[Woo482]

I can tell you one of the fetures: it will seperate the phisics core from the grafics

 

[psychotix]

How can they be the same thing in the first place?

 

[Linguofreak]

Sorry if this is in the wrong section but I was just wondering about the progress on the new version of orbiter. I've seen the new planet textures but I was wondering when it will be coming out and what it will feature.

As with everything here, it will come out on a Tuesday. That Tuesday could be sometime later this year, or sometime in mid 2010.

 

[John_L]

How can they be the same thing in the first place?

What's meant by that is that right now, and, since the beginning of Orbiter, the graphics code and physics code have resided in the same source code, in other words, it's all in one program. The Orbiter Visualization Project will have the physics and graphics code separated. Dr. Schweiger will continue to modify and improve the physics portion of the program, (and maintain the privacy of the source code), and other programmers will have access to the graphics portion of the code. Those other programmers will move forward to improve the graphics, and, also different graphics engines will be able to be developed, so that Orbiter can be ran on different systems. As it stands right now, Orbiter requires DirectX 7 (or higher), which means it's 'held in place' by the limitations of DirectX 7. Dr. Schweiger did this so that people with older, less powerful systems can use Orbiter, giving Orbiter the widest possible user base. By separating the graphics from the physics, multiple coders/coding teams can develop newer, better graphics engines taking advantage of DirectX 10, OpenGL, etc, so that those of us with newer machines and video cards can use the 'bells and whistles' that our hardware has, and much more graphically rich engines can be developed, taking Orbiter to the 'next level'. This will also allow DirectX 7.0 specifications to be maintained so that no one is 'left behind'. It's the best of both worlds. Low end support for those of lesser means who have older hardware, and high end support, (not to mention non-DirectX support, so that those using say, Linux can enjoy Orbiter).

In the (hopefully near) future, There will be diffrent versions of Orbiter to choose from, and you'll select the one that's appropriate for your particular system / setup.

 

[psychotix]

So since the two components are being separated does this mean that there will be collision models?

 

[Woo482]

no I dont think there will be but i cant remember that much at the moment its 6:15 AM over hear in the uk and I have not had much sleep

 

[hypersonic]

Simple yes or no answer......

Without pressing for a 'when', thus getting another 'next Tuesday' answer...

To all the beta testers (& martin), a simple Yes or No ....

Is it looking likely that there 'will' be a new Orbiter release before 2008 is over?
If 'No', will it be early 2009?, or is it transpiring to be over half a year away at least?

If a simple status update were to be given, then all the 'when when when' crowd would keep quiet.

Cheers

 

[Artlav]

Without pressing for a 'when', thus getting another 'next Tuesday' answer...

To all the beta testers (& martin), a simple Yes or No ....

Is it looking likely that there 'will' be a new Orbiter release before 2008 is over?
If 'No', will it be early 2009?, or is it transpiring to be over half a year away at least?

Both yes, no, undefined and possible at once.

To resolve a fundamental misunderstanding - software development, like any other artistic activity, cannot be planned ahead for any kind of reasonable schedule, especially not when an artist is doing the thing alone and unbounded.

 

[Biscuit]

Both yes, no, undefined and possible at once.

To resolve a fundamental misunderstanding - software development, like any other artistic activity, cannot be planned ahead for any kind of reasonable schedule, especially not when an artist is doing the thing alone and unbounded.

I really wish my bosses understood that when they ask for an update to something and figure that it should only take 30 minutes to plug in a whole new routine, add the output to the report, test it, and get it out ...

I mean, it's only one little routine, right?



Some things with projects can be forcasted and planned, but most people never think about the point that you've made. Writing code can be (as far as I see it) an art. Art takes time and imagination. It's not just plugging a bunch of lego blocks together to make a tree ... well, for good code, anyway.

... still though, some days it does get a little hard waiting for Tuesday to arrive. :rofl:

 

[hypersonic]

Yes, thankyou for the 'nebulous' replies.
As it happens i am involved in software development & have programmers working for me. - & YES i USed to Code myself, so appreciate the 'art form' that it is.

Whilst i don't press for precise deadlines, unless there is a commercial reason for doing so. - Which of course Orbiter is NOT that.
My developers ARE able to give a rough PERCENTAGE complete.
So when i ask them, 'how's it going guys', how far from completion do you feel you are?

I Don't get flippant... "Next Tuesday" or yes, no, maybe..
They reply, based upon their own gut feelings... "80% or 95% complete, or nowhere near..."
So that's all the question really needed from people who ARE in the know...

75% plus, without any deadline would warrant a 'Yes'
or less that that , based upon your 'collective' GUT feeling .. would be a 'No'
Simple as that.

So whats your gut feeling on whether a release might happen this year, based upon your assessment of Percentage 'complete' ... a yes or a no?

Many thanks

 

[simonpro]

Martins is the only one who can answer that, it'd be doing him a disservice for anyone else to do so.
It's also abundantly clear what Martins policy on talking about future releases is.

 

[Pilot7893]

And I know there is a tool that will include an Atlantis launch autopilot

 

[jb42682]

If you search the forums there are many post on this topic already. one of them has a link to the beta.

 

[TSPenguin]

My developers ARE able to give a rough PERCENTAGE complete.
So when i ask them, 'how's it going guys', how far from completion do you feel you are?

I think the real problems are that this is not a project from scratch, nor one with a defined final result.
There is updating, reinventing, inveting, implementing new features, thinking of more features, taking 'features' out again, seperating two tasks from one code (inventing a lot of new cool things along the way and implementing an ever growing todo list). Ordering new paper for the todo and "cool things to think about" list.

At least this is how I see the development happening. I might be completely wrong and Martin has a fixed todo list, a very well structured plan how to handle everything and a blueprint of the final code. But somehow, I doubt that... Maybe Martin will let us know one time.

My point is, that this is not like "write a program that can do this" and not even "hey, let's make a game, where you can grab your oponent and throw him around. Let's give him some ubercool body suit with neat features and let him use vehicles...."

There was a defined goal for the next release, that was seperating the physics core from the graphics engine. Even if you only by accident sometime read in the beta forum (here and in the m6 dungeon) you will know that so many, sometimes little things, have changed, been added or revised to work better. So much for keeping to the task

I like developers to say "Ooohhhh, shiny" and go for it. And that is why I prefer tuesday over monday.

P.S.: This post was supposed to be a small 4 part list. Now it took me 25 minutes to write it and that is why, oh, look at that, shiny...

P.P.S.: [fun] And hey, Artlav was so bored waiting for martin to finish the dx7 client so he could continue on the OpenGL one, that he singlehandedly wrote orulex and pimped it to the point where he says that he reached the absolute limits of orbiter 2006. [/fun]

 

[Quick_Nick]

And I know there is a tool that will include an Atlantis launch autopilot

That's the Lua scripting language. A bit different from being 'just an autopilot'.

 

[Biscuit]

... - & YES i USed to Code myself, so appreciate the 'art form' that it is.

Whilst i don't press for precise deadlines, unless there is a commercial reason for doing so. - Which of course Orbiter is NOT that.
My developers ARE able to give a rough PERCENTAGE complete.
So when i ask them, 'how's it going guys', how far from completion do you feel you are?

...

I apologize if I came off as totally flippant ... personally I'm used to so many people, usually ones in my command chain, trying to play the role of the 'pointy haired boss' and thinking that it's exactly like an assembly line task ... this program will take 35 widgets to build, you can put together 5 widgets per day, therefore you will have this program done at 4:15 PM Friday afternoon ...

You can get a rough idea when something will get done, but sadly so many people take an 'estimate' as a concrete guarantee. *sigh*

You actually seem to understand there is some art and skill involved, give your guys the leeway they need, and stay involved and keep them on track (if I ever find the rogue spreading that scurrilous lie that techies are easily distracted :blink without the micromanagement ...

... hmmmm ... you guys hiring?

As far as the new features of the new Orbiter, I tried the Beta ... even with all the graphics options turned wide open and higher level textures, I got a better framerate on my old POS PC. Sadly, OrbiterSound doesn't work with the Beta ... amazing how much difference sound makes.

Anyway, Tuesday will be here eventually ... sooner rather than later I hope (and from what I've seen of the Beta so far, the release should be just downright glorious).

 

[hypersonic]

"You actually seem to understand there is some art and skill involved, give your guys the leeway they need, and stay involved and keep them on track (if I ever find the rogue spreading that scurrilous lie that techies are easily distracted :blink without the micromanagement ...

... hmmmm ... you guys hiring?"

We actually have a product launch in October.... So ya never know ;-)

 

[Buzz Lovell]

My developers ARE able to give a rough PERCENTAGE complete.
So when i ask them, 'how's it going guys', how far from completion do you feel you are?

Software development by hobbyists is a little different to software development in a commercial/development house: more freedom to change scope and requirements, work, waxing and waning enthusiasm, other hobbies and interests, simply choosing to spend more time with family, etc. etc. throws the gantt chart out the window and with it even general indications like "before end 08." There's a lot of wisdom behind the answer "Tuesday" IMHO.

When I am paying Martin and Artlav and the testers & co. their salary, THEN I'll ask for dates. In the meantime - thanks for all the fun so far, guys!

 

[dbeachy1]

Sadly, OrbiterSound doesn't work with the Beta ... amazing how much difference sound makes.

Some good news on that front: ObiterSound does work in the newer beta releases (including the new public 080910), just not in the previous public version (080516).