why 250GB

[laserman]

I was wondering why is the terraGIT planet 250GB ?

The scenery 2.0 terrain is 87GB
Let's assume , that the .git folder is also 87GB
-> 174 GB

are we automatically downloading all old commits(versions) of a tile ?

[IAHM-COL]

I was wondering why is the terraGIT planet 250GB ?

are we automatically downloading all old commits(versions) of a tile ?

Short answer. No
TerraGIT only copied/ported the terrain content from terrasync over. No history import.

ex.
https://github.com/FGMEMBERS-TERRAGIT/w ... ts/Terrain

In fact, the terrain subdir has not received much modifications since 2.0
(maybe 1 or 2 recently)

So, it is not like you'll be seeing lots of history content out of terrain

And in the objects, most editions are plain text stg files. The rest are just addition of new binaries (png or xml) or new text files (ac).

Rarely (actually never for what Ive seen) something in the content of this scenery represents modification of binaries.

The reality is that I can engineer terraGIT a bit better, so it is around 180 G as you present, but I am waiting for scenery version 3.0 with all LOD and stuff (that psadro is working on) before I go that way. Anyhow, it sounds like that new system is going to be much lightweight anyways. We'll see what he comes along with.

(Keep in mind that terrain is what carries most weight)

[laserman]

TerraGIT only copied/ported the terrain content from terrasync over. No history import.

In fact, the terrain subdir has not received much modifications since 2.0
(maybe 1 or 2 recently)

Japan, D-ECHO's islands, EDLM, ...

There were some Terrain commits since 2.0/terrasync version.

what about the repaired and updated airports from
http://media.lug-marl.de/flightgear/Terrain/
did you include any of those ?
if not, anybody here who likes to do that ?

We'll have an issue with the airport elevations:
when you repair/update an airport, the elevation might become different from it was before.
then the objects will sink or hover.

[IAHM-COL]

TerraGIT only copied/ported the terrain content from terrasync over. No history import.

In fact, the terrain subdir has not received much modifications since 2.0
(maybe 1 or 2 recently)

Japan, D-ECHO's islands, EDLM, ...

There were some Terrain commits since 2.0/terrasync version.

I wasn't aware that Japan terrain was patched in terrasync. That history hasn't be ported over terraGIT. Who constructed Japan and based on what sources, by the way?

I am asking, because if they simply using my Japan content from FGMEMBERS, without asking me, you'll notice that

• 1. They are not asking the authors and respecting their wishes as they use to propagandize around.
• 2. Objectively they could. It is GPL. But again 1. They certainly (no-one has) did not asked me if I was interested in proceeding with this merge
• 3. I would find extremely hypocritical that they are speaking around saying that my work does not benefit their "upstream" but they sneakily still take my work when they find fit, and never recognize it -- nor to me, nor to much anyone else
• 4. I mean #3, because, this was not announced in either the devel.list nor any of the forums (Curtis or this one)

what about the repaired and updated airports from
http://media.lug-marl.de/flightgear/Terrain/
did you include any of those ?
if not, anybody here who likes to do that ?

I had not included them. Basically, I've got a work overload.
Most people had not wanted to learn how to do that, basically Thanks to a propaganda of terror that spread over on the other forum. So collaborators are not very keen to jump to help.

On this sense of ideas, the task is not terribly difficult. Are you willing to port these changes over to GIT and send these as PR to the respective terrain tiles?
(I know, it is not very simple task, but as you and I know how to script a bit, it can be computationally optimized)

If you can't I definitely have this in my queue.

The major issue I have is that you provide btg.stg files for those airports layouts but frequently you don't offer the reconstructed terrain. The final outcome is not 100% optimal. Sometimes terrain overrides over parts of the new airports.
When I patch an airport I tgconstruct at least 2x2 degree tile surrounding the airport, and make sure to transfer enough btgs to completely reconstruct surrounding terrain (somewhere between 1 and 4 additional terrain meshes, depending on where the tile boundaries sit along the airport)

We'll have an issue with the airport elevations:
when you repair/update an airport, the elevation might become different from it was before.
then the objects will sink or hover.

I scripted a fix for that. If you mean, the guys in terrasync has the issue, then yes. If you mean we in terraGIT, then NO.
Look on fixed/reconstructed airports with objects in terraGIT for evidence: example: KSFO/KHAF, you'll see I ported this correctly.

[IAHM-COL]

about the Maldives and EDLM. yes. TerraGIT got those fixes before terrasync actually.


A quick question about Japan and EDLM:

How did they solve the problem of tile stiching in continental tiles?
I understand that is a "deal" breaker?