[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/vt/ - Virtual Youtubers

Search:


View post   

>> No.10466918 [View]
File: 803 KB, 1600x1600, trialsandtribulations.png [View same] [iqdb] [saucenao] [google]
10466918

>>10466259
so, ninjaripper 2.0.4 is an entirely different beast. Unlike 1.7.1, it is much, much faster, and 2, it has different import options.

Unlike 1.7.1, it actually seemed to work real well with 2k20. we did another test export and found it loaded faster in blender too. Didnt have to wait approximately 20 minutes for it to import and instead only took 1 minute or so. hooray. Oh, and the addon packaged with it actually functioned in blender. cant forget that.

We initally just exported the match on default export settings. For some reason, ninja ripper 2.0.4 has a need to manually input the y value of the thing you are ripping. Basically, if you get it wrong, all your shit that you imported get's slanted to one side. As an example, a sphere will turn egg shaped.

So our test match imported great! ... well, kind of. There was some slant to it, you see, and we thought "eh, we'll just guess and check the y-value ourselves." Again, a sign of pure naivete.

We switched over from our test match in the ring to ripping the model from the character creation screen since ninja ripper rips EVERYTHING and makes like, 600 different files to import and the character creation screen has less geometry forit to retrieve, cutting down the number of imported objects by half and saving me harddrive space.

And thus begins the tale of slant alton.

You see, we did some tests on alton because he has a super simple appearance and the character creation screen actually takes place in a sphere, and that gets imported too. We could tell if wee were off with teh Y-value if that sphere got all egg shaped.

after hours of trying, we were not able to get a Y value that worked.

SO, we looked for other solutions. Turns out, there's something called "matrix import" that uses a value from the log file on export to do the proper import. So we searched the log file a bit and found a matrix value listed under "float4x4 g_fm44ViewProj" and NOW we had a sphere and a proper model to work with.

And so the journey continues.

Navigation
View posts[+24][+48][+96]