search

TrilogisIT / WorldWind_Android

WorldWind Library and Example code for the Android framework.
19 stars 10 forks source link

Improve ETC1 support #5

Open nicastel opened 10 years ago

nicastel commented 10 years ago

This is a big one Nicola! Take your time for the review.

This pull request :

I have also made this changes :

ndorigatti commented 10 years ago

Yes, it's a very big one! I'll be out of office for the whole week and hence have no time to check/merge. Status bar hiding is something we wanted to be but was in low priority. Thank you very much, i'll review it as soon as possible. I think now we need to fix elevation and after that the sdk can be considered quite good!

nicastel commented 10 years ago

I have found some texture corruption problem when using the client side renderscript compressor. They are mostly visible when you tilt the view (wuth a very strange effect) wich means this is mot likely due to some problems during mipmap compression. For the time being it is better to deactivate client side texture compression by putting :

.jpg in BMNGWMSLayer.xml and EarthElevationModelAsBil16.xml There is no such problem with server side on the fly compression with Geoserver (based on my JavaETC1) and Mapserver (based on gdal/crunch)
ndorigatti commented 10 years ago

I cant understand what you mean for, can you attach a screenshot?

Sent from my Nexus 4 Il 23/mag/2014 00:44 "nicastel" notifications@github.com ha scritto:

I have found some texture corruption problem when using the client side renderscript compressor. They are mostly visible when you tilt the view (wuth a very strange effect) wich means this is mot likely due to some problems during mipmap compression. For the time being it is better to deactivate client side texture compression by putting : .jpg in BMNGWMSLayer.xml and EarthElevationModelAsBil16.xml

There is no such problem with server side on the fly compression with Geoserver (based on my JavaETC1) and Mapserver (based on gdal/crunch)

— Reply to this email directly or view it on GitHubhttps://github.com/TrilogisIT/WorldWind_Android/pull/5#issuecomment-43947995 .

ndorigatti commented 10 years ago

Hi @nicastel , i'm trying to merge the commits but i have problems with RS classes. Eclipse blames about supportv8 package, i dont know if committing RS part since you told creates artifact...

ndorigatti commented 10 years ago

For now the merge has been done to a new branch, we need some testing and checking on DDS conversion (seems to be not completely flexible). Thank you for the PR!

nicastel commented 10 years ago

To be able to build Renderscript you need to install with the SDK Manager the Android SDK Build Tool 19.0.3

ndorigatti commented 10 years ago

Yes, i then manage it later, I did not merged the project.properties file. Up to now, there is a branch dedicated to the merged @nicastel branch. I'm going to do the same with @keidze branch and then look after to merge the whole (I'm sorry but some things will be not merged due to both us three worked on common things).

2014-05-24 13:35 GMT+02:00 nicastel notifications@github.com:

To be able to build Renderscript you need to install with the SDK Manager the Android SDK Build Tool 19.0.3

— Reply to this email directly or view it on GitHubhttps://github.com/TrilogisIT/WorldWind_Android/pull/5#issuecomment-44084704 .

nicastel commented 10 years ago

Hello, I fixed the image corruption problem by deactivating MipMap generation in the Renderscript ETC1 compressor (see my last commit).

About mipmap I am not so sure that they are useful in worldwind since they increase file size and compression time by ~40% and since WorldWind already have a pyramidal tiling system. I suppose that they can avoid some drop of performance during fast unzoom. Anyway the JPEG/PNG textures do not contains them so it will not create a regression of performance.

I will be happy to review keidze stuff also but as you noticed there is some build problems for now.

ndorigatti commented 10 years ago

Hello, glad to read the fixing of image corruption. About mipmap, i dont know too, my colleague told me that there is no gain with mipmap but probably some loss (no technical values, just impressions).

I managed to make the library to compile from keidze pullrequest, the client is not updated i suppose, so we'd need to create a new one that uses his changes. My colleague is working on keidze transparency and elevation commits, we have a code freeze before friday, so it's kinda a rush. After that we will dig better on the code and try to make a full merge of all the commits.