• Welcome to PiBoSo Official Forum. Please login or sign up.
September 23, 2020, 03:41:10 PM

Oliver's Mount

Started by Grooveski, February 06, 2015, 06:06:25 PM

Previous topic - Next topic


No worries m8, brain needs to rest  ;)

BOBR6 84

Gold cup this weekend!! Anybody going?


Estoril's turning into another saga so I'm back at OM for a break.   ::)

Looked into the rounding off error.  Think it was when DE changed the model from lwo to 3ds.  Have tweaked some settings(and the model a touch) and fingers crossed for next time.

Finished the landscape poly-reduction.  Had done the poly merging and deleted the free points but not the ones that were attached to vertices...
...so all those thousands of extra points had been turned back into polys before going ingame.  Reckon the background.edf was maybe 140% the size it should have been.  :-\
Oops.   :-[
Anyhow - all gone now.

Ran Mountside through the mill again...

...then thought "what the hell..." and done Mere as well.   :)

Moved back the fence up the rise a little.

Back in SpeedTree again tonight.  Still just feeling my way around but already the Mere tree is down 90-odd-percent in poly count and still kinda almost looking the part - closer to the Dirt3 arrangement and count.
As I say - still just fishing around.  ST is one of those programs that's designed to simplify a single task but in itself is complex as hell. 
(...like Bryce or Poser - with leaves :) )


Nice. Trees in GPB have almost always sucked, if you pull something decent out you already deserve a medal :)


Very nice Groovski!  ;) 8)



Quote from: HornetMaX on October 12, 2016, 08:46:20 PM
Nice. Trees in GPB have almost always sucked, if you pull something decent out you already deserve a medal :)

You'd understand how SpeedTree works better than I do.  At the top end it's a vegetation physics plugin for games and modeling programs.
I've chosen Dirt3 models as a benchmark because of the games I've looked at those seem a good balance of what works and what I may be able to do.  D3 was an improvement over 2 and Rally is beyond my means but 3 is just about right as a goal...  :)

I'm using it to knock up individual(static) trees but it's designed as a total solution.  Has been used in every game I've played that's had moving trees since Skyrim or thereabouts.


I had a look indeed, quite a piece of kit.


Indeed, SpeedTree does look a very nice piece of kit....... Have you seen the price of SpeedTree Cinema?! WOW!  :o



Like a kid with a new toy here.   ;D

Reasonably pleased with the overall match.  It's definately the same track I was trying to model.  Some bits are quite close and some that aren't I had my doubts about anyway.
...like the second of the esses - which I always kind of knew should have had a deeper dip.

Slightly frustrating are the bits I had close once but seen fit to change during the build.  Turns out he pits are as deep as I originally thought.  Farm Bends are too shallow now as well as is the dip under the bridge on the drop down to Mountside.   ::)
The lidar's going to be real handy for finishing the TSOs.  Real heights for all the buildings and a rough guide for some of the vegetation.  :)

Memorial has a rise through it that I hadn't noticed.

I'll be using the geometry of the field at the startline.  Probably other areas too but this field is first on the hit list.  Bit of poly-red and it'll be much nicer than what's there at present.

Mountside to Farm - The jumps straight...
I always knew there was a lot of guesswork there and reckoned I'd have to fine tune it as the versions went on(by comparing game and real onboards).  It was too flat to get much of a clue from contour maps and half of it's under trees on the aerial photos, is a wild ride on the onboard vids and being at the tail end of the lap never done it any favours - I was usually sick of whatever job I was doing by then and in a hurry to get on with the next step.

Right - that's enough excuses...
...maybe - just!  :P  There's no two ways about it - it's a mile out.
White line is my attempt.  :-[

So, the plan?
hmmmm....   :-\

- Tweak the model roughly into place so the landscape and junctions are matching up.  Get the back of the model verges matching with the lidar.  A lot of sections are less than a degree out so I should get away with just tilting some of the fences.  Other bits will need realigned but shouldn't be too big a deal.
- Back to loft for the road.  Take the lidar into max, redo the spline heights and export a new road model.
- Redo the camber and subdivides.
- Blend the two together.  An unexpected and somewhat tedious job that I've already done once but I did reckon that seam needed some poly-red - that's an ideal time to do it and will cut down the joining work a fair bit.
- Back to business as usual.   :)


Nice indeed Grooveski, lot of work. But looks encredible detailed.


Great Work Grooveski!

Really looking forward to this track ;D


Wow what a million hours of work :)


I'm in no hurry.  Check out the build's start date!   :P

Third time's a charm.   :)   First round was for GP500 and lay untouched for 6 years or more - where it was when I arrived over here.
Second time is what I've been doing for the last year or so.  Would have been ok, have got better over time and probably most folk(myself included) would have been pretty happy with it.
This time though it's going to be right - and that's a nice thought to have.  :)   Worth a little extra work.   ;D


New road loft is done and fresh into LW.

The lidar data was a little lumpy(around 30mm variance between one point and another that should both have been at the same level).  Still soooo much nicer to work with than contours and the likes though.   :)

The rise through Memorial - which I had just over a metre before is now closer to three.

Now we're talking!   ;D


I can see this is going to really be very good! Excited!!  ;D 8)