[Virtua_LM] Official tracks' topic (last update: 06-14-2014)

MarcG said:
whats the changelog for Mid Ohio please? always enjoy reading change logs!
Click to expand...
In a nutshell:
- all new shadow configuration: you can choose the shadow level to match your capacities of your mashine
- new fast path for the aiw
- different shader and higher resolution for the trees
- modified track naming for a better support of rf2 Rank
- the flags are animated
- some minor fixes
 
If you update the tracks could you please update the .GDB files with more relevant Event Names.

Currently we have

SEBRING_12H.GDB
Code: 
  TrackName = Sebring 12h Course
  EventName = 12h Course

This makes it show up in the server list as 12h Course. Club layout shows up just as Club Course. You look at it and think ok Club Course but what track in the world is it I need to download.. The server list really should show TrackName under the Circuit Tab but instead it shows EventName.

ISI themselves fall foul of this with all the tracks. Plenty of server called Outer Loop, Inner Loop. If the user didn't already know it was Mills it could be any track in the game!
 
Sam Moss said:
This makes it show up in the server list as 12h Course. Club layout shows up just as Club Course. You look at it and think ok Club Course but what track in the world is it I need to download.. The server list really should show TrackName under the Circuit Tab but instead it shows EventName.

ISI themselves fall foul of this with all the tracks. Plenty of server called Outer Loop, Inner Loop. If the user didn't already know it was Mills it could be any track in the game!
Click to expand...

Yes. you are right.
But maybe all modders wait for changes in displaying data format. To be honest it the place where things should be changed. Track name concatenated with Event Name should be shown in dedic as well as on server list in game
 
ethone said:
In a nutshell:
- all new shadow configuration: you can choose the shadow level to match your capacities of your mashine
- new fast path for the aiw
- different shader and higher resolution for the trees
- modified track naming for a better support of rf2 Rank
- the flags are animated
- some minor fixes
Click to expand...

Awesome thanks mate, especially liking the AIW update. Thanks VLM for Sebring too (sorry forgot in my original post!)
 
Very nice track ! No low fps as some other tracks... The new fast aiw path works fine, it's a pleasure to race against bots :)
Thank you !
I keep the 2 releases because of vmods. I'll unsintall all, mods and V1.90 when those vmods will be updated. it's not an easy way to manage track update...
 
Last edited by a moderator:
jimcarrel said:
It will not let you uninstall old version because you have vMods installed that refer to the old versions. (as do most users). You have to uninstall the vMod that refers to the old version of track.

Most tracks so far, have done okay with two versions installed. However, who wants to eat up disk space with multiple versions of the same track. This has been a big discussion.

On a better note! Thank you for continuing to support VLM tracks! What a marvelous opportunity that VLM continues to offer us!
Click to expand...

How i manage to discover witch vmod is llinked to the old version?

ps: thanks ISI to make things easier
 
Mine show to be connected to 5 vmods.. count 'em, FIVE. So, easiest thing and worst thing is to give up on vMods. and pick thru "All Cars and Tracks" series, and stay off of servers. Just imagine, another 5 track releases with updates here and there, and you too may give up on vMods and servers.

just saying
 
And if this team who relase the full version and just launch the update to old version, same thing occur? It will be a necessary a new vmod?
 
SLuisHamilton said:
And if this team who relase the full version and just launch the update to old version, same thing occur? It will be a necessary a new vmod?
Click to expand...
If you want that vmod to utilize the new track, yes, it would still require a new vmod.
At the moment it's a moot point though since the track update option seems to be bugged.
 
tried it for a few hours, very nice for first beta release

just a few points ( im sure they are already covered)

a slight to much satuation
track surface needs a tad more life ( cant feel anything on the filled tarmac areas (hole infill detail)
no slight cambers anywhere

maybe its a completly flat track in real life but im guessing there would be a bit more charactor in the surface

again thanks for this nice track, cant wait for future updates
 
Thanks a lot VLM! Your tracks are simply the best! Mid-Ohio is so addictive! ;)
 
@Adrianstealth: If you are talking about the Sebring track, it really is flat in real life, it is in Florida, USA, built on an airport's grounds. It used to use one of the active runways as the back straight, but they moved the track off of that area.
 
gearj.

yes -sebring i see wow that flat, ok i understand, that makes sense -thanks for the info

if poss a bit of charactor added to the track repairs would be nice then (small bumps under tires etc)

p,s the track is still a top piece of work im just trying to add constructive points etc
 
Thanks for this track guys :)

There's one suggestion I have, though.
When you look at tracks by ISI you can see they consist of 600-900 GMT files. The only exception is Spa, which has just below 2000 GMT files, but this track is large, so you may say that "concentration" of GMT files is similar.

In case of Mid Ohio we have over 2000 GMT files and Sebring has over 4000 GMT files.

This is the main reason why these two tracks perform much worse on my system. Cutting track geometry into bunch of small objects will not optimize the track too much for GPU, but will greatly increase CPU demands.

I've made a test once with rF1. I've placed 1000 identical cubes in one GMT file. Rendering time was 0.2ms. Now I've put each cube in separate GMT file - rendering time was 4.2ms.

The reason for it is that GPU is typically capable of rendering a few thousand triangles by the time CPU can prepare next mesh for rendering.

As far as I remember, you had all energy posts in separate GMT files, even though they were concentrated in given areas. Another exampe would be tents in pitlane - despite being identical and standing nearby, they were placed in separate GMT files.


Rendering a mesh requires selecting shader, textures, setting material parameters, binding vertex arrays - all this takes time. While CPU is trying to do all that, GPU is starved.

Collision models are a different story. Game engine should optimize them while loading, but if it doesn't it's better to have collision models chopped into smaller pieces.
But for rendering it's always better to merge objects and limit unnecessary CPU-GPU communication.


Just for a comparison - CryEngine 3 can render entire scene with like 400 "draw calls" - including all lighting, shadows and reflections.
Now imagine that rFactor must render 1000 meshes separately, then again for shadows, then again for reflections.... There's no chance it's performance will be on par with other games in market this way.


I hope this makes sense to you guys :) And good timing - I was just looking for a track for some funrace in Nissans :)
 
You must log in or register to reply here.
Back
Top