Isle Of Man

[doubledragoncc]

FUCK BIG BEN!!!!!

Its just a big stone cock anyways lol

DD

[Storris]

Can one or preferably some of you developer types wade in with some advice?

The mesh makers, RFPro.com, have asked, "what parts of the scan, exactly, do you need meshed?"

My instinctive answer is that the entire scan should be meshed as a single unit, but I've no real idea.  From the point cloud, would anything need to be meshed separately, or is 'everything' the right response?

[matty0l215]

Can one or preferably some of you developer types wade in with some advice?

The mesh makers, RFPro.com, have asked, "what parts of the scan, exactly, do you need meshed?"

My instinctive answer is that the entire scan should be meshed as a single unit, but I've no real idea.  From the point cloud, would anything need to be meshed separately, or is 'everything' the right response?

Basically everything would need to be scanned but mainly the track. Things like building and terrain will need to be seperated but i imagine this can be done at a later stage. Hawk can give a definet answer

[Storris]

Cheers Matty.

Hawk?

The scan will encompass a lot of non-track items.  Within a single IoM mesh file, would there be a single continuous mesh covering everything, or would each thing (house, wall, lampost) need to be meshed independently?

I see the first option as being just as useful as the second, but with a much lower cost and much lower system requirements.

[Hawk]

Cheers Matty.

Hawk?

The scan will encompass a lot of non-track items.  Within a single IoM mesh file, would there be a single continuous mesh covering everything, or would each thing (house, wall, lampost) need to be meshed independently?

I see the first option as being just as useful as the second, but with a much lower cost and much lower system requirements.

Sorry guys, I've been out all day......

Okay. To put the IOM track together in GPB we would need the track, terrain and all the extra peripheral objects like walls, trees, pavements, kerbs, buildings, signs, spectators, barriers etc, etc, modelled as separate objects in the scene.


Edit: We're not even sure GPB could handle a track the size of a full 37 miles IOM track..... If you imagine all the objects that would have to be involved? Tests need to be done to find out how far we can push GPB before it starts coughing it's guts up with the strain. 

Hawk.

[janaucarre]

It will need big improvisation to have as less as possible polys

[RiccoChicco]

The TT Mini was running great while there was a lot (I'm talking about thousands) of objects in the scene. The game just loads the parts you need, so the track could be 1000km long, it will work great too.

[Hawk]

The TT Mini was running great while there was a lot (I'm talking about thousands) of objects in the scene. The game just loads the parts you need, so the track could be 1000km long, it will work great too.

I really hope we have no issues, but I'm more on the side of caution and awaiting to see what the tests will show before any major work is started. 

Hawk.

[Storris]

Official quote is in. £9,850  Waiting for conversion & data storage/webhosting quotes.  I'll be back with links when we launch.

[Storris]

Ok, so, I'm ging to have to creat 200-300 survey grade GPS control points for the scan.

Will they be useful to any of you guys in sorting the .GPX out?

[Hawk]

Ok, so, I'm ging to have to creat 200-300 survey grade GPS control points for the scan.

Will they be useful to any of you guys in sorting the .GPX out?

I haven't got a clue what your talking about there. Lol 

Hawk.

[Storris]

  lol, me neither really.  Just been told that part of the scan preparation requires "client to provide Ground Control Points", essentially ~1cm accurate GPS co-ordinates, every 250-300 metres of the track!

I thought this might allow the .GPX map to be a massaged into better shape, but thinking further, it would probably just make things more difficult.

[Hawk]

lol, me neither really.  Just been told that part of the scan preparation requires "client to provide Ground Control Points", essentially ~1cm accurate GPS co-ordinates, every 250-300 metres of the track!

I thought this might allow the .GPX map to be a massaged into better shape, but thinking further, it would probably just make things more difficult.

1cm GPS co-ord for every 250-300 meters of the track for ground control points!? 
I presume ground control points are to do with fleshing out the basic height map of the IOM track surface?

If that is correct, then taking a 1cm accurate track GPS point into consideration at only every 250-300 meters is going to lead to major inaccuracies at some points on the track surface height and shape surely? 

Hawk.

[HornetMaX]

I don't see any other usage of the control point outside "massaging the point cloud into better shape" (I like the image, very hands-on:) ).

What I can't figure out is why they don't take the Ground Control Point themselves, at the same time they do the laser scan. Would be much easier.
Unless ... Is the company doing the scan different from the company "converting" the cloud into something more manageable ? It's the 2nd company who's asking for the GPS coords os is it the 1st company ?

[Storris]

If I've understood correctly, the GCP's are to allow an accurate overlay of the geo-reference image, and to correlate the data properly with the rest of the world.   I'm hoping to keep them to a minimum since the track only has to relate to itself.  They will be logging GPS, and setting up a temporary GPS base station, but they need pre-determined reference points to alleviate any potential drift during scanning & processing <<= Accuracy.