*****GP4 Track editing*****

Oleg only tested it on cars and not on rims, wheels, tracks etc. If he did he would have known that gp4 files can contain multiple scenes.



[1991] [2003] [2004] [gp2 world series]

well i dont know what he did i only posted whats in the readme :P

can we get

DaveEllis Wrote:
-------------------------------------------------------
> too bad that ZM only exports one object at the
> same time
>
>
> Is that ZM restricted or filter restricted? If its
> the filter cant it be modified?
>
>
hmm, maybe i'm wrong because cars exist out of more objects. I think i had to say gp4 editor only imports one object at the same time. (like one object export)
Sorry i didn't edit for a quite some time now, so i'm easily mistaken.

ZM also creates carlightmap textures when exporting, these have to be removed for tracks too.





Edited 1 time(s). Last edit at 01/09/2005 02:24PM by erwin_78.

I have Oleg's gp4 source code that he released 6 months ago in his forum

That's cool, now we need someone to modify it...



[1991] [2003] [2004] [gp2 world series]

I also found a link to download it so here it is:

[zmodeler.nfscheats.com]

I don't know if it has all the codes used , maybe there is something missing because someone asked Oleg that he could also give dff source plugin.



Edited 1 time(s). Last edit at 01/09/2005 09:18PM by vesuvius.

or ask him if he could edit it too :P

OK, I was out for the week end and just came home...

@Xero: I read your essay and will reply more precisely tomorrow (I'm not 100% reliable this evening )

@All about ZM: The one object at once feature is not a ZM limitation (for this time ) but a 3DEditor one I deliberately did as a first step for editing to see if no unknown bugs could appear in the import/export features.

I Got ZM GP4 source filter and as Stan found, it wasn't designed to save anything else than cars.

deleting objects would be nice too, files are only growing specially with these large grandstand.

@Erwin: Yep, an old time missing feature...

Beware : long text below
OK, I made another test, getting orientation of Monaco's hairpin track sectors in excel to be able to quickly average them.
Well, in original Monaco hairpin, neither track sectors are pointing to next one, nor track sector orientation equal average value of surrounding sectors...
So all I can guess is that the orientation average method could be fine for very first track sector of each cc-line section to smooth line if you manually edited orientation of sectors in the area around this first sector.
It can't be used as a rule to fix a whole track for example.

Displaying more files content is possible, but giving hand to edit them at the same time is another thing.
So solutions I actually thought about are :
- displaying many files simultaneously with editing posibilities (not the easiest way)
- auto-generating other files from one (but this has issues as behaviours (for .col files) are not described in .dat files and track sector refinements (for .bin files) are not defined in .dat ones)
- simultaneoualy editing the same vertex files .gp4/.bin/.col (I'm thinking more and more at this one as I think I could do it now, one of the matters would be it might not be easy (for me) to create new vertices/faces)
I don't know yet which one I will be able to realize.

About sectors definition in auto-generated files, that's the matter I pointed out above, because the most usefull is to create .bin file from .dat one, but .dat do not seem to store sectors' sub-divisions count and thus auto-generated .bin files from .dat are actually a bit raw as sectors are not subdivided.
At the opposite, .bin files created from .gp4 files are well subdivided but obviously need an updated .gp4 files to fit new .dat one.

.col file are not ordered by behaviour originaly. I did it to simplify the search amongst all vertices as they are all in a big list with no other flag to find the one you need.
I could re-order them by index but them you would have a single loooong list...

Multiple select will become possible in the list but this can't be usefull for this file (maybe not much in others either). Badly multiple mouse selecting should have to wait because I really don't know how to solve it in code (that's the reason why single mouse selecting do not exists either in 3DEditor).

Well, you and others can't help me much on programming that's true, but reports as the one you did the latest days about cc-line first sector particularity were really helpfull and that's the kind of work that can help me much.
So the best you (and other testers) can do is continuing to test and try features for any particular usage and keep inform me on any result. There's still much to learn about GP4 track files data and you can really help there.

There's no problem with your essay, I think I did the same with this post

I really hope to be able to fulfil your wishes, I'm just sad track editing is starting so late; I could have done so much one or two years ago...

I will do my best anyhow to bring a new version as soon as possible at least to simplify .bin and .col editing to match .dat/.gp4 changes.

Lo2k Wrote:
-------------------------------------------------------
> @Erwin: Yep, an old time missing feature...
>
> Beware : long text below
.....
>
> I will do my best anyhow to bring a new version as
> soon as possible at least to simplify .bin and
> .col editing to match .dat/.gp4 changes.
>
>


sounds promising lo2k. keep it on! dont you got a job already? best wishes for 05 from here!

Well, in original Monaco hairpin, neither track sectors are pointing to next one, nor track sector orientation equal average value of surrounding sectors...
So all I can guess is that the orientation average method could be fine for very first track sector of each cc-line section to smooth line if you manually edited orientation of sectors in the area around this first sector.


Yes, thats what I mentioned in my post. Only the track sectors containing the beginning of a ccline section need to be averaged. I also think the phrase "pointing to the next one" is slightly off, but I think you understand what I meant. A few posts back when I discussed working out the angles is much simpler than I first realised. The equation (as you probably have already worked out) for track sectors where a ccline section begins is:

Where TS = Track Sector -> Sector's Orientation -> Direction

TS = (TS-1) + (((TS+1) - (TS-1)) / 2))

I'll have a look at the unknown values next to see if I can make any sense of them.

About sectors definition in auto-generated files, that's the matter I pointed out above, because the most usefull is to create .bin file from .dat one, but .dat do not seem to store sectors' sub-divisions count and thus auto-generated .bin files from .dat are actually a bit raw as sectors are not subdivided.
At the opposite, .bin files created from .gp4 files are well subdivided but obviously need an updated .gp4 files to fit new .dat one.

I think it would be best to generate .bin and .col files from the .gp4. The .dat is arguably is easiest to edit in terms of layout, so being able to overlay the .dat over the .gp4 or the newly generated files would be the best option IMO.



Edited 1 time(s). Last edit at 01/10/2005 08:59PM by Xero.

@Stifflers Mum: Thanks

@Xero: Well, just for info, you forgot to divide by 2 the ((TS+1)-(TS-1)) part in your formula
Many people do not think like you about track generation and they prefer working on .dat, leaving the .gp4 file for one of the latest operation...
for me, as either .dat and .gp4 have to be update manualy, they are both important.

Lol... so I did! This cold I have is playing with my head!

Just thought i'd show you what i've been up to recently.



The circuit still needs some reshaping to give a better representation, but the harder part is at least done. Those grass verges were a nightmare!!

I've stumbled across a problem involving the ccline however, and think one of the unknown values affects where the ccline section crosses a track sector. I'll play about with it and explain as best as I can later when I find something more concrete.



Edited 1 time(s). Last edit at 01/11/2005 05:22PM by Xero.

great stuff

the verges are indeed a nightmare, they're what have actually stopped my experiments for the moment (I also need to be able to remove sectors, lol)

what dat number is the nurburgring so I can have a look at the original layout as Im not sure what it looked like :S

Keiran


vMax.Keiran
Velocity Maximum Simulation Racing
www.vMaxSR.com
www.liveforspeed.net LFS Status:

how can you create the track sectors? It is not possible with 3Ded