Page 21 of 24

Work progress - 0.9.8.1

Posted: 12 Jul 2017 04:23
by HarbingerDawn
Astrophisiclover wrote:
Source of the post the 3d require more render and more power from the hardware

It might look like it, but that's not necessarily true. Certainly the RAM requirements won't change.

Work progress - 0.9.8.1

Posted: 12 Jul 2017 06:33
by Astrophisiclover
oh ok

Work progress - 0.9.8.1

Posted: 12 Jul 2017 14:04
by Yuno
is the out of vram crashes fixed? current version the game crashes quite frequently when travelling large distances, going up close or landing on multiple bodies, things that require loading. vram usage keeps rising & rising till its full then game crashes.

i have a gtx 1070 8gb, windows 10, latest drivers

Work progress - 0.9.8.1

Posted: 12 Jul 2017 15:47
by HarbingerDawn
Yuno, I have made 9 long livestreams of 0.9.8.1, and in all of those streams SE crashed maybe 3 times total, in over 10 hours of video (I have a GTX 970, by the way). So it doesn't seem to be a significant problem for me.

Work progress - 0.9.8.1

Posted: 13 Jul 2017 01:14
by Astrophisiclover
Yuno wrote:
is the out of vram crashes fixed? current version the game crashes quite frequently when travelling large distances, going up close or landing on multiple bodies, things that require loading. vram usage keeps rising & rising till its full then game crashes.

i have a gtx 1070 8gb, windows 10, latest drivers 

Same here when i play and go in certai zone the vram crash and you see black screen and when you zoom at large distance (like u are one telescope) the screen is black.

But is so strange why my PC no have any problem for other game with the same requirments i try elite dangerous and the game run perfectly

Work progress - 0.9.8.1

Posted: 16 Jul 2017 01:29
by Julian
SpaceEngineer, I've posted this before on the old forum, but please DON'T implement expanding supernova remnants in 0.9.8.1. There would be no point in doing this unless you can model planets being destroyed when they're hit by the shockwave, and then you'd also need to depict every stage of stellar and planetary evolution.

Work progress - 0.9.8.1

Posted: 16 Jul 2017 02:13
by HarbingerDawn
Julian wrote:
SpaceEngineer, I've posted this before on the old forum, but please DON'T implement expanding supernova remnants in 0.9.8.1. There would be no point in doing this unless you can model planets being destroyed when they're hit by the shockwave, and then you'd also need to depict every stage of stellar and planetary evolution.

What? SNRs currently contain no other objects, not even a stellar remnant, and if they did they would only contain neutron stars and black holes. There would be no planets to be destroyed. Saying that you need to implement every stage of stellar evolution to show dynamic nebulae is like saying that you need to implement complex weather patterns to have moving clouds, or simulate the internal geology of a planet to have volcanoes. One does not logically follow from the other.

Anyway, this is all still very much work in progress, and a lot of things will be changed/refined before release, so expanding SNRs as they've been shown so far might be very different from the final product, and might work very differently.

Work progress - 0.9.8.1

Posted: 16 Jul 2017 12:35
by Kamil Cader
Julian wrote:
SpaceEngineer, I've posted this before on the old forum, but please DON'T implement expanding supernova remnants in 0.9.8.1. There would be no point in doing this unless you can model planets being destroyed when they're hit by the shockwave, and then you'd also need to depict every stage of stellar and planetary evolution.

By following Your logic, there would not be any point releasing unfinished engine unless it is capable of simulating every aspect of real universe, including Yourself, who seems to be denying its right for gradual evolution.

Work progress - 0.9.8.1

Posted: 16 Jul 2017 15:00
by SpaceEngineer
New nebulae requires virtually 0 MB of VRAM (just a shader code). But requirement for the GPU increases, if you want to explore them in 3D. When user did not moving to fast or too far (saying inside a planetary system), nebulae does no make any load to GPU (they are rendered to a static skybox).

Work progress - 0.9.8.1

Posted: 16 Jul 2017 15:33
by Julian
Kamil Cader wrote:
Julian wrote:
SpaceEngineer, I've posted this before on the old forum, but please DON'T implement expanding supernova remnants in 0.9.8.1. There would be no point in doing this unless you can model planets being destroyed when they're hit by the shockwave, and then you'd also need to depict every stage of stellar and planetary evolution.

By following Your logic, there would not be any point releasing unfinished engine unless it is capable of simulating every aspect of real universe, including Yourself, who seems to be denying its right for gradual evolution.

It's OK to release an unfinished engine if it depicted a steady-state universe where the only things that move are planets and moons in their orbits. But if supernova remnants include the year in which they became supernovas, then if you wind time back before that year, you would expect to see a star in that location, and then you'd start wondering why the only stars that go supernova are the ones that don't have planets.

Work progress - 0.9.8.1

Posted: 16 Jul 2017 15:59
by XBrain130
Julian wrote:
Source of the post But if supernova remnants include the year in which they became supernovas, then if you wind time back before that year, you would expect to see a star in that location, and then you'd start wondering why the only stars that go supernova are the ones that don't have planets.

Right now SNRs don't even generate from stars, they're simply animated nebulae looping eternally. If you rewind, it would just roll back to the maximum extension before shrinking back again.

Work progress - 0.9.8.1

Posted: 16 Jul 2017 23:51
by Julian
I know, but having stars go supernova is going to require fundamental changes to the way Space Engine works. Objects will need to be able to change their types at specific times: stars becoming neutron stars, planets becoming scorched planets and so forth. I think these kinds of changes should probably be left for 0.9.8.2.

Work progress - 0.9.8.1

Posted: 17 Jul 2017 03:13
by Canleskis
Julian wrote:
I know, but having stars go supernova is going to require fundamental changes to the way Space Engine works. Objects will need to be able to change their types at specific times: stars becoming neutron stars, planets becoming scorched planets and so forth. I think these kinds of changes should probably be left for 0.9.8.2.

It doesn't necessarily require theses changes at all, stars could just go supernova without any consequence on planets.

Work progress - 0.9.8.1

Posted: 17 Jul 2017 10:00
by Zoppadoppa
Canleskis wrote:
Source of the post It doesn't necessarily require theses changes at all, stars could just go supernova without any consequence on planets.

Temperature and irradiation still are factors to consider, along with the mass loss from the explosion itself.  The orbital velocities of the planets would mean that some planets are thrown out while others get much more eccentric orbits than before.
I think that simulating physics on that level is something that shouldn't be seen for a while, if ever personally.  Especially seeing how the timescale system was built w/ watching orbits in mind.  
In regards to what I hope to see by 9.8.2, more diversity in the architecture of star systems would be nice.

Work progress - 0.9.8.1

Posted: 17 Jul 2017 11:19
by Canleskis
Zoppadoppa wrote:
Canleskis wrote:
Source of the post It doesn't necessarily require theses changes at all, stars could just go supernova without any consequence on planets.

Temperature and irradiation still are factors to consider, along with the mass loss from the explosion itself.  The orbital velocities of the planets would mean that some planets are thrown out while others get much more eccentric orbits than before.
I think that simulating physics on that level is something that shouldn't be seen for a while, if ever personally.  Especially seeing how the timescale system was built w/ watching orbits in mind.  
In regards to what I hope to see by 9.8.2, more diversity in the architecture of star systems would be nice.

Of course I was referring to the next version, these features will probably be implemented in a distant future but for the next version stars just going supernova is enough and doesn't need all the other stuff.