Tundra-Urho3D 3.0.0-prealpha

["Lasse Öörni"]

Hi,
the development of the "new" Tundra aka Tundra-Urho3D has reached a
significant milestone: it can run as a server for both native & Web
clients, and it can run JavaScript code (including the Tundra "avatar
application") via the lightweight duktape JS engine. At this point it also
features basic scene & entity-component editor windows.

I have tagged the latest development version in the github repository as
3.0.0-prealpha.

The repository is available at https://github.com/realxtend/tundra-urho3d.

Build scripts exist for Windows (VS2010-2015 supported) and Ubuntu Linux.
Compiling for Android is also possible. However, at this early stage there
are no prebuilt binaries available.

For those unfamiliar with the Tundra-Urho3D project, it's a lightweight
reimplementation of the realXtend Tundra 2.x.x core without the heavy
libraries Qt and Ogre. Instead the MIT licensed, multiplatform Urho3D
(urho3d.github.io) game engine is used as the base framework and as the
renderer.

--
Lasse 嘱rni
Game Programmer
LudoCraft Ltd.

[Lord]

10 months and nothing new ?  Is Tundra dead ?

[Lord]

Where has Tundra and RealXtend Association gone ?

[Toni Alatalo]

Hi, sorry for missing your question earlier.

Tundra and realXtend have not gone anywhere, in the sense that the
open source project is there, as are the realXtend association and
realXtend foundation.

However, there has not been active open souce initiatives now, for
reasons that are good to discuss more - I am happy if this turns out
into a fruitful planning discussion, thank you.

To put it short, Tundra (as WebTundra) is still there as the core of
Mapgets which is a commercial service. It is also in FIWARE, where the
same folks that developed it for mapgets were (/are?) active. I have
not been personally involved on that side of things recently. Several
companies previously involved in Tundra development are using
commercial suites for business (Unity, Unreal and friends), but that
is nothing new per se -- also back in 2012 LudoCraft and Playsign used
Flash, while also using Tundra (Playsign mostly three.js / WebTundra).
In Wizard Quest Playsign used Cocos2D-iphone in Obj-C.

My main responsibility is Playsign business and product development
and we are now not using Tundra there, but Unity instead. Main reason
for Playsign to use Unity starting 3 years ago was good native mobile
app performance. Nowadays we are happy with the WebAssembly and WebGL
implementations there. For certain bureaucracy reasons we have not
been involved in the recent phase of FIWARE so my knowledge from that
side (basically from open souce) are outdated. I help also the
University Ubicomp city models etc group, where they decided 1,5 years
ago to start using Unreal for VR app / research prototypes (they/we do
not do engine dev).

Regarding Open Source, standards and interoperability, I actually find
that there is still crucial work to do to achieve more of the goals of
the realXtend effort. One lesson has been, an old one already, that
due to the generic and common Entity-Component model in Tundra and
other game engines, and nowadays IoT backends (also in FIWARE) and
what not, one possibility for realXtend development would be to
prototype and standardise common virtual world applications
components. The spec being implementable cross systems, including
Tundra, Unity, Unreal, FIWARE Context Broker, Open Simulator, name it.
In Playsign's current business in city planning, we have discussed
this sort or things related to planning information about areas, in
land use planning / city zoning. Very essential in all this are the
existing standards, from BIM to CityGML and GeoJSON to 3D-Tiles. Much
of my work both for Playsign and the University has been staying more
or less on track in geospatial & 3D city app related standards
development.

That stuff in previous paragraph was worked on by the Oulu University
comp science folks in the national 6Aika city model spearhead project
in 20-16-2018 (ended this summer). Work is still on going to take
lessons from that work forward.

But that's not the only option - in the meeting that we had already a
long time ago, it was also discussed whether Urho3D development
support would be worthwhile.

I'm sure anyone here is also open for suggestions for development
ideas. The association and foundation are there, but parties here have
not had so much need for them now, and we have been all very busy also
with applications and usages that realXtend was created for.

Thanks for asking, hopefully this explained the situation somewhat
helpfully - am happy to discuss more. Will also talk about this later
with prof. Ojala who is the current chair of the foundation (also very
busy with many things but does have time for this too).

Humbly yours,

> --
> --
> http://groups.google.com/group/realxtend-dev
> http://wiki.realxtend.org
> http://dev.realxtend.org
> ---
> You received this message because you are subscribed to the Google Groups "realXtend-dev" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to realxtend-de...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.