Laguna Seca Laserscan Complete, Release for Assetto Corsa Expected Late 2017

[Paul Jeffrey]

Marco Massarutto has confirmed laserscanning of the epic Laguna Seca Raceway has been completed, suggesting the track may appear in game before the end of 2017.

Already widely acknowledged to be making its way to Assetto Corsa sometime in the reasonably near future, the noise around Laguna Seca and the simulation has died down somewhat in recent months, leaving fans of the game to wonder exactly what is happening with the famous American venue.

With something of a well timed update post on his personal Facebook page Marco Massarutto, Product Brand Manager of AC developers Kunos Simulazioni. has gone on record to state that the laserscanning element of the track creation has now been completed (albeit later than anticipated), and the track is in the works to be completed and released to the game by the end of the current calendar year.

Long anticipated by fans of both Assetto Corsa and American themed race tracks, Laguna represents one of the most impressive driving experiences in the United States and has been host to numerous high profile national and international level car and bike racing events throughout its illustrious history.

Well known to sim racing fans thanks to its almost ever present representation in games such as Gran Turismo and Forza Motorsport, finally PC players will get the opportunity to sample a fully lasterscanned version in Assetto Corsa.

Despite the delays Massarutto confirmed scanning completed just this week after unexpected delays, with the following statement made via Facebook released by the Italian Kunos founder:

"After a pile of burocratic and logistics issues that resulted in a sensible delay of our operations in US, [finally] Yesterday we have completed our laserscan survey at Laguna Seca, so the track should be completed and available for Assetto Corsa within the end of the year."

Exciting times ahead for fans of the track and simulation as it should prove to be incredibly interesting to drive a full scanned version of a track that has been so popular in racing games over the years. At this stage it is not clear if the track will be a free bonus content addition or form part of a future DLC release for the game, however it is expected that more details will be released in future as development back at Kunos HQ continues to progress.

Assetto Corsa is available to purchase now for PC, Xbox One and PlayStation 4.

More fun happens at the RaceDepartment Assetto Corsa Racing Club, so sign up, head over get racing today! Its a blast and you won't look back once you've been bitten by the bug! If you just want to catch the news or hang out with fellow AC fans then fear not, we accommodate that too! We have a thriving Assetto Corsa Sub Forum where you can do exactly that, or even check out and download some of the great mods on offer and upload and share custom car setups! Its all here at RD, what are you waiting for?!?

Laguna! Anyone else excited? What are your memories from this track in other racing sims / games? Let us know in the comments section below!

 

[Rupe Wilson]

If laser scanning is costing them all this time to release a track maybe it's time to drop it. It's not like tracks that were scanned a month ago are the same thing now anyway, let alone stuff from 6~12+ months ago in places with hard summers or winters.
But that's just me, I'm not picky about laser scan, my enjoyment of a track never increased/decreased because the place was or wasn't laser scanned, as long as it is well done I'm ok with that.

Could not agree more with you,
and i do realise Laser Scanning does speed up production of a track.

 

[Rupe Wilson]

Now leave the personal stuff out of this thread please. and stay on topic.. i will be watching.

If you are unhappy with a post you can always use the report button and a member of staff will look at it,, you don't need to get into a verbal ping pong battle with any other member..

 

[Michael Hornbuckle]

I don't know how a racing game that features licensed real world tracks in 2017 can even start to call itself a 'sim' without said tracks being laser scanned.

It's not like tracks that were scanned a month ago are the same thing now anyway, let alone stuff from 6~12+ months ago in places with hard summers or winters.

Do you guys really think that tracks go through major elevation changes or that the turns somehow gain or lose width/radii because of a cold winter or hot summer?

 

[CC]

I am forever grateful that Kunos can provide unreal accuracy of some of the worlds greatest racetracks. But i can't help but think AC needs more circuits and this is a very time consuming process. AC has the cars, They have the graphics and they have most importantly the community. Whilst community tracks are great and there are some real gems out there, I do wish Kunos made some more tracks for the game. That aside, Can't wait!

Simply do that maths man ,
Compare the prices of iracing content , And then ooooh lets look at AC content + DLC ,
The difference is enormous ,
We all agree more are needed , Just simply takes time to do so , 4 Tracks a year would be a good start

 

[F_B]

Great addition and nice track. I know it from Iracing so the differences shouldn't be huge.

 

[Rupe Wilson]

I don't know how a racing game that features licensed real world tracks in 2017 can even start to call itself a 'sim' without said tracks being laser scanned.

Do you guys really think that tracks go through major elevation changes or that the turns somehow gain or lose width/radii because of a cold winter or hot summer?

Major elevation changes.. you do make me laugh
maybe you can point me to that part so i can read it for myself..

 

[Jockey]

I'm quoting something just for fun since some started again the to be laser scanned or not to be laser scanned. Just a big food for thought.

A - "Right. I usually start explaining track accuracy from a different point of view, namely the physics engine itself.
Let's assume we have a physics engine that runs at 400 Hz (which is a pretty common frequency, although I know not all run at exactly that frequency).
Now let's also look at a car driving at 200 km/h, which is about 56 m/s.
For each step of the physics engine, that car moves about 14 cm.
I think this is an interesting starting point for further discussion, as it means that for example a pointcloud with a resolution of 1 cm^2 will have a lot of points that will be skipped because the physics engine is running at a frequency that is way too low. You could even drive over a smaller speed bump, or crack in the road, if it's less than 14cm wide.

B - "true, but still the poitn remains you want to pick up the exact position you are on when updating right?"

A - "Sure, so the question is how high the resolution of your road mesh should be so the points where the road should be are (very) close to the real point cloud.
Given that small bumps, like cracks and small uneven things might not be picked up when you only sample every 14cm."

B - "wild guess. 1-2 cm no clue honestly"

A - "Well, 1-2cm would be way too high, at that speed you would waste almost 95% of the points because they would be skipped. Small bumps would not be noticed that way. Btw, I honestly don't know the resolution of the point clouds that AC and/or iRacing use (or even if they use those to drive on). We don't drive on point clouds. Our real road mesh is fairly high resolution and that is our basis for driving on. We do add a few things, such as some high resolution noise to simulate bumpiness of the asphalt and those two things combined get us fairly close to what a point cloud would offer.
A point cloud is a great source of information, but you still need to create polygons out of it."

B - "but does laserscanning not offer other advantages which outweight the scanning costs? back to textures and modeling again. as modeling should be straight forward and textures could be used eventually? i have seen an example where this was done and it looked stunning, but probably to heavy on performance. but at least the important bits yould be used (kerbs, track surface etc)"

A - "No consumer PC can render polygons created from every single point in a point cloud, so you have to simplify. Try loading a track with 1cm^2 polygons into any sim. It's way too much."

B - "well at the end there are allways more ways to get to rome i guess not having touched the amount of time one has to spend with one way or another however last point remains is the sheer selling point."

A - "Yes, that is nice, but you can also do a lot of that from detailed photographs. It's a selling point for sure, iRacing started marketing it, and it's working. And it is a good way to get accurate tracks (if your point cloud is accurate and you take the time to properly create a 3D model from it). There are also other ways that are accurate. For example, I think for ******** we are building now, we took about 300GB worth of photographs. And we are building from 3D CAD data.
And then of course we all run into the issue that tracks get repaired, repaved and (slightly) changed over time.
3D CAD data is not captured, it is the original design that the builders used."

C - " if its not captured then it can be off by quite a bit i assume building a track is not like building a car or other small object"

A - " Well.
I'm not going to comment on the accuracy but I do know that tracks need safety approvals of all kinds and those are based on the drawings so I am sure there is someone monitoring if the build is accurate.
Btw, even laser scanned data is not 100% accurate when captured, there are always errors.
In theory you could probably extract the Silverstone track from rF2 and AC and compare the meshes"

C - "maybe but for me what makes the track is also the bigger stuff like cambers in the track surface if you drive circuit zandvoort in real life and then in iracing the way the car behaves over the surface is 1:1
because they capture all the undulations and cambers in the surface with the laserscanning

A - "Yes, you work with a good reference as a starting point. I don't deny that. But your average track no longer contains all those points when you download it. So there is a lot of data being aggregated. I don't know exactly how much, but you can probably do the math. Does anybody here know how big the average track is (download size)? from memory, we're talking about a couple of 100's of MB's, right?"

C - "iracing's imola is 500MB "

A - "Ok. So that is 3D model and textures. Typically most of that will be textures (they take the most space). Let's assume about 80% is textures, 20% is points. so we have about 100MB worth of point data. And let's say Imola is 5km long (4.9 according to wikipedia). "

C - "but in iracing its stored in a .dat file which is compressed i believe"

A - "Sure, want to do a guess on the average compression ratio? 50%?
So if you had a 1cm^2 point cloud, that would be about 5000 * 100 points in length, and let's say the track is 5 meters wide with another 5 meters left and right of it, so 15 * 100 points in width.
Each point has an X,Y,Z coordinate, and each coordinate takes 8 bytes (a "double" for those of you who program), so 24 bytes per point.
That is about 18GB of data.
so they probably at most ship about 1% of that data
and this is all rough math, but I don't think we're too far off "

 

[RasmusP]

Very interesting to read, I like detailed discussions!
One thought on that: for me laserscanning is not about the road mesh, but instead about 100% correct corners, elevation changes and camber/tilt of the track as whole.
Perfect example: Projekt cars vs AC on the Nordschleife. The "Karussell" is just wrong in pcars. It's way too hard, small, I don't know... It looks perfect but the difference when driving through it is huuuuge!
14cm is more than enough for that kind of things!

To laguna seca: oh ma gawd, oh ma gawd oh ma gawd! It was my favorite track when I was 7 years old, playing Sportscar GT with a Joystick and even back then needing a pack of (little) balls going down the mighty Corkscrew

Again, I am not sure that the pcars version is Spot on, some things feel off compared to real life footage so I am really happy to have it scanned and spending my christmas on sunny laguna

 

[Michael Hornbuckle]

for me laserscanning is not about the road mesh, but instead about 100% correct corners, elevation changes and camber/tilt of the track as whole.

QFT

 

[Nevini]

Well I'm still happy, even 1 track a year is better than no tracks.

 

[Jockey]

Very interesting to read, I like detailed discussions!
One thought on that: for me laserscanning is not about the road mesh, but instead about 100% correct corners, elevation changes and camber/tilt of the track as whole.
Perfect example: Projekt cars vs AC on the Nordschleife. The "Karussell" is just wrong in pcars. It's way too hard, small, I don't know... It looks perfect but the difference when driving through it is huuuuge!
14cm is more than enough for that kind of things!

To laguna seca: oh ma gawd, oh ma gawd oh ma gawd! It was my favorite track when I was 7 years old, playing Sportscar GT with a Joystick and even back then needing a pack of (little) balls going down the mighty Corkscrew

Again, I am not sure that the pcars version is Spot on, some things feel off compared to real life footage so I am really happy to have it scanned and spending my christmas on sunny laguna

Well, who knows how they built the not British tracks in pcars. But I got the feeling if you use a 3dcad data you will get those things right.

 

[BhZ]

We decided we didn't want non-laserscanned tracks a while ago when they released Zandvoort. I think we have to reconsider that decision in order to get more tracks... There are mods out there of tracks that we wouldn't get laserscanned anyway like Le Mans and Bathurst that deserve to be in this game imho. If the other option is to have 1 track per year, then this is something we have to accept.

 

[Chris]

Track looks great! I love Laguna Seca
My second favourite North American circuit after Road America.

 

[CobraCat]

20 cars a year 1 track a year, really..priorities are so arse about face with AC now.

If only there was a resource were you could download quality addon tracks for Assetto Corsa. Its a shame such a resource doesn't exist. I suppose if I were to make up such a resource it might look something like this..

http://www.racedepartment.com/downloads/categories/ac-tracks.8/

 

[JPL083]

Did anyone check the tarmac's condition after scanning?...

 

[Will Mazeo]

I don't know how a racing game that features licensed real world tracks in 2017 can even start to call itself a 'sim' without said tracks being laser scanned.

Do you guys really think that tracks go through major elevation changes or that the turns somehow gain or lose width/radii because of a cold winter or hot summer?

Elevation changes can be done correclty without laser scan and no, that's not what I'm talking about on the winter or summer part, I'm pretty sure you are intelligent enough to realise that.
In the end you drive the same laser scanned track in different sims and they still feel different. I'm pretty sure there are non laser scanned tracks being sold as "laser scan" when they aren't and people are happy with it. Or do you really believe everything everyone is telling you.
From my part I don't care, reason why I use mods (as long as they have a decent quality), it's just better to have it from the dev itself.
I feel sorry for the console crowd tho. I'm pretty sure majority of them would prefer 30 well done tracks than 10 laser scanned stuff

 

[hitm4k3r]

I don't know how a racing game that features licensed real world tracks in 2017 can even start to call itself a 'sim' without said tracks being laser scanned.

Do you guys really think that tracks go through major elevation changes or that the turns somehow gain or lose width/radii because of a cold winter or hot summer?

So according to your logic AC isn't a sim because their licensed version of Zaandvort isn't laserscanned ... what a brilliant logic. I consider laserscanning a great tool, but people go a bit too much nuts about it and imply that other ways of doing a track are wrong. It is very nice to get some general accuracy, but it is really just a a snapshot of how a track is at a certain moment and you don't drive on the point cloud. Tracks change all the time: general layout changes, weather, temperatures, insolation or let it even be a repavement. The LS versions of Monza in AC for example is not up to date anymore. Same goes for the iRacing Silverstone, Lime Rock Park, or Suzuka and people are still forced to pay a fortune to get them, when ISI's non laserscanned versions of some of the tracks are more or atleast equaly accurate and more up to date. And lets face it, you won't feel or see a difference when you drive Silverstone in AC, rF2 or even Project Cars.

 

[Jockey]

So according to your logic AC isn't a sim because their licensed version of Zaandvort isn't laserscanned ... what a brilliant logic. I consider laserscanning a great tool, but people go a bit too much nuts about it and imply that other ways of doing a track are wrong. It is very nice to get some general accuracy, but it is really just a a snapshot of how a track is at a certain moment and you don't drive on the point cloud. Tracks change all the time: general layout changes, weather, temperatures, insolation or let it even be a repavement. The LS versions of Monza in AC for example is not up to date anymore. Same goes for the iRacing Silverstone, Lime Rock Park, or Suzuka and people are still forced to pay a fortune to get them, when ISI's non laserscanned versions of some of the tracks are more or atleast equaly accurate and more up to date. And lets face it, you won't feel or see a difference when you drive Silverstone in AC, rF2 or even Project Cars.

I like you comment but Imho you are wrong.
Some people can measure the track with cm accuracy just by looking the monitor.

 

[Peter Hooper]

I absolutely love Laguna Seca. It's been my favourite tracks since my days racing it in Indycar Racing back in the early 90's. Getting it for AC will be immense

 

[hitm4k3r]

I like you comment but Imho you are wrong.
Some people can measure the track with cm accuracy just by looking the monitor.

If you show me a person who is able to meassure track width on a monitor with cm accuracy just by looking at the monitor while actually racing at speed I will be fairly impressed. Just considering that there are so many variables like seating position, camera angles, FOV, pixel sizes and monitor resolutions I highly question that you will see a difference between a laserscanned track and a properly developed track using other sources and good research, as it was nicely explained in your previous detailed post. Ofcourse there are some things that you will notice instantly while driving LS Nords in AC versus the old modded rF1 version for example, as those tracks are off by several meters in places. But at the end of the day there is more placebo involved than anything else.