referring to katsbits, we would gain in both due to that, power AND quality

if you like i can do further investigations at gamedev.net

but i guess that's not needed no?

i guess the above linked article explains the reasons well, why rescaling shouldn't be left neither to the program nor to the GFX card.

not because it's not possible to do so, it's the "weaker" solution rather a simple "keep the POT* rule". 😉

well industrial standards, they have their reasons, equal to why still often 8bit depth is used, even when anything else is possible to.

*power of two

that excerpt of NVIDIA's CG tutorial is also very interesting to read, when you leave away the permanent tapping on their own shoulders.

http://developer.nvidia.com/node/76

Well, I replaced the alpha 10 facegen folder with potsmoke's 'POT' converted folder. Unfortunately, it made no difference. If the compression is different in alpha 10 than 5d61822 (which works fine for me) then that's probably the reason. I guess in a way it's good to have a crappy old computer after all! 😆

Replacing the images will not make the slightest amount of difference. Its nothing to do with the size of the images.

which surprises me not little 😮

because else i can actually see no reason for that...

compression really makes no difference, as long as only the huffman table is used and no restrictions to the original data have been made, which is not given by .png compression levels, it's lossless, you don't loose one bit of the picture and chksum is the same after un-compressing them like the original (unlike a .jpg which looses many information, but i don't have to tell you).

anyway to avoid a complications the pics in the folder i send you was compression level 0, even if there is no difference, except loading speed.

and packet size of a zip is the same, you can compress lossless only once, there is no further or only very little you get out of that.

for a data stream that's a bit different, because there you have to set frames for the huffman coding, you like to listen or watch in real time, for a static data not.

can you describe the appearance a little further (attach a screenshot)? and i forgot actually what system (your computer) you use, will be also of importance

perhaps i can find a reason for that, because it seems no one else has this problem.

[/hr]

for the devs, what would be if you really would pass all texture data un-scaled to the process? i mean all textures even icons would have to follow the POT, imo that's only professional.

problems with some gfx cards is not the only reason, read the dammit katsbits stuff, textures get bad of scaling from irregular to regular and has nothing to do with blender or not, only because that this is a article of blender community, it's important for ALL computergames. further if the "artist" scales his pictures he has appearance in his hands and didn't have to leave that up to a softs. i know i am only gernot and sometimes i feel like "ahh not that odd gernot again..."

usually i go and investigate, check a lot of articles where i find information to a certain problem i have. one could never know everything, but one can ask and one can know where to ask (old rule of my chemistry teacher).

if i have some facts, i can ask some "wise guys" and I have EARS to listen... i didn't think i am super, in nothing i do (even when i am perhaps, i can't see that), sometimes this hinders me, sometimes this is only good. 😀

i am not to proud to ask a stupid question..., might be because one has first to LOOSE EVERYTHING that he has loved.

there is not much left that i am proud of, even not my paintings, that's somehow stupid or bad, but no way to get back.

Go it. Facegen has to be modified to output a power of two texture so even the oldest ATI cards can work with it. Thanks for everybody's interest in this.

And Gernot, I would have loved to hang out with you and your dad back then. He was an :ugeek: but I'm not up to that standard. Maybe a bit of a :geek: , but if anybody around here is an :ugeek: , it's you. 😎

I'm thoroughly enjoying the assumptions that the devs are not listening, not reading, etc. As I've already stated, the issue is here solely because I screwed up and threw a non-power-two texture into GL. The problem is understood. There is an open issue to track the problem, with a suggested fix. All its waiting for now is for someone to find time to implement it.

wait some fun stuff... i posted before i know but i love it

there are many kinds of :geek: and :ugeek:

btw, i still have my old comic books.... 😉

Luv it potsmoke66 😆

dont be to hard on em guys they are doing a good job with pioneer im playing the nightly builds and loving it. Just figured id throw some encouragement in there 😀

A minor quibble with alpha 10. On the Planetary info screen Discovery Base is depicted orbiting Jupiter, or possibly Callisto. It's actually orbiting Ganymede.

Actually it is orbiting Jupiter, but its distance is "wrong" so it appears on top of Ganymede. Issue #116.

OK, thanks for the reply. Since I was 'evicted' from Mars High, I've toured the Sol system and decided on Discovery Base for my home ..um.. base of operations to work on detailing the nice_spacestation. There's some nice photo-ops with Ganymede and Jupiter both in view... for now anyway. 😀

Is there going to be a search for planet feature implimented? its hell looking for a certain star in amongst all the mayhem on screen. or even if the mission would target the planet you were supposed to be delivering package/killing someone at?.

Regards

I've been wanted the targeting button in the mission screen for a while now. You've motivated me to finally log it: issue #128.

I would like to do a system search but I haven't figured out how to do it well. The problem is that except for a handful around Sol all the systems are generated on demand. To search all of them would require a prohibitive amount of time. One possibility is to limit it to "nearby" systems, but for what value of nearby? Once we start thinking about that we end up with a solution that may not work for all the cases, or comes with a complicated user interface.

Does that mean you're thinking of implementing a "target system" button next to accepted missions? That's a fantastic idea but what happens if your rubbish jump drive requires you to travel through a few systems to get there? Perhaps instead it could plot a course for you, linking your current system to the required destination with coloured lines, a bit like freelancer?

I knew you were thinking about Freelancer even before I finished reading your post Subzeroplainzero. 🙂 That feature was indeed handy in that game, but for Pioneer, maybe even a different coloured blob on the target system when you access the galactic map would do? In that, once you accept a mission the game will highlight the system for you. Then you can plot your own route.

Another thing I always wanted was a log function where, say the co-ordinates of last 10 systems you visited, could be stored and used to speed up navigation. That way if your lost (happened to me all the time in Frontier 🙄 ) you can back track.

some sort of log- mission log would be a good idea if there was anyone to do it. Put it in the to do list

Actually all I was thinking is that it would switch you to the sector view and show you the system. If you were out of range it would be up to you to figure out how to get there. Of course if we ever get a better sector view with a path-finding mechanism then we'd be able to use it.

That's a great idea. Please open a feature request.

That's really cool. Do you know how it might work if you have multiple missions open?

Perhaps each mission accepted on the list could have its own color blob that corresponds to the system on the galactic map (as long as the colors weren't the ones used for the present system and hyperspace navigation blobs).

To answer robn first, keep it simple, use different coloured blobs. At present Pioneer is using blue to show you where you are and green for a selected system, so basically any other colours that take your fancy.

And for SeanN, yes that is along the lines I am thinking, but it could be refined by either by having different colours to represent different kinds of mission and further refine that by increasing the brightness of the colour depending on how soon the mission time ends. Or you could use coloured shapes around a star. Anything really to designate to the game player that these are important. Eg, a bright yellow triangle might mean a soon to be completed bombing mission or a dull purple square might mean a delivery mission that has plenty of time left. The game player would be able to tell at a glance looking at the galactic map what they need to attend to first. 🙂

And I will enter a feature request for the flight log, but need to think about how it might work first. I dont just want to write "gimmie a flight log!" 😆 I will try and think about how it could work. 🙂

Incidentally, if people are thinking about features they would like to see in Pioneer, do check out the super massive "Ideas For Elite IV" thread at the Frontier forum. Never have I seen a game's features analysed and argued about by a gameplaying community in such detail and over such a long period of time. Someone might as well use that honey pot of ideas 🙄 Be warned though, it's big as in the Douglas Adams sense of the word, here is a link http://forums.frontier.co.uk/showthread ... 86&page=21 😉

no the elite 4 ideas if they can be slowly added to pioneer by the contributing programmers is a good idea, elite 4 doesnt exist so all youre taking are ideas from this and other space based games and putting em into an open source mix

I mentioned that thread because it is a cornucopia of ideas for the ultimate space sim. Many of them are perhaps beyond the resources of the Pioneer project, but there are many more that might be possible to implement. It's not like say, FRONTIER DEVELOPMENTS are making use of them! 🙄 Besides a good idea is still a good idea no matter it's source 😉

as u say resources and time this is all personal developer free time contributions and its all appreciated. Having a look at the ideas about 1/3 my guess are simple things that can probably be done the others would require sourceing more willing contributors