Quote from three_jump :

Would it be possible to get the interior updates for the cars (FX0, RB4, ...) at some point as well? Those were nearly finished in 2009, if I remember correctly.
Or would updating those require an incompatible version (and a reset of the lap-times?) and are therefor kept until the physics are updated?

The cockpits of some cars are just so outdated and "ugly" by todays standards, especially when compared to the level of detail in Westhill and Rockingham.

Yes some of them are really back in time.

I know Eric wants to update the car interiors and other things. But we are agreed that another track is the priority at the moment. The S3 license looks so limited in what it offers, that updating S1/S2 stuff is just not the priority at the moment. He likes to focus on one thing at a time.

I don't know what will be the priority when the track is done, or if the tyres will be before or after the track. We're just trying to crack on.

Quote from vitaly_m :

Thanks for sharing the plans.

If Eric finishes a track sooner than tyre physics are ready, will the track be released sooner than tyres?

Yes, now that we made the decision to sell S3 licenses, there is no reason to hold back tracks because of tyre physics. The quicker we can get a track out there, the better, as the S3 license upgrade will look like twice as good value. But it will take a while as Eric is working to a high detail level as you can see in Westhill and Rockingham.

Originally the tyre physics delays held up some things, and still holds up the Scirocco. As most people know - and this has never changed - tyre physics was and still is the reason for the delay in the Scirocco.

The tyre physics update is the best thing that can happen to LFS on the code side. On the content side, new tracks are the best thing.

It will also allow us to simulate more different types of vehicles because the tyre model should scale correctly from small tyres to large tyres. There are so many possibilities from having a better tyre model, not only that it feels better and more fun to drive as you put the power on coming out of a bend and so on. A physics model based on real physical principles that produce the mathematical functions that produce the slip curves and the way of combining lateral and longitudinal forces (they aren't 'combined' as such, they just come out of the model naturally).

And no, I'm not going into details now. Right now I'm trying to get this quick update out of the way. And as I said then I'm working on functions for Eric and at last I'll be back on the tyre physics. If there's something interesting to say about it in a few months, then perhaps I will.

I noticed that my previous post could be misread, so I just want to make it clear there will be no tyre physics updates in the coming patch.

1) Test patch in a few days with crash fix, a few small lesson editor fixes, possibly an MPR smoothing.
2) Not going to get into a massive patching session. I've just come off one. Just release the full version with a few minor updates.
3) Do some editor functions that Eric has asked for.
4) Work on tyre physics.

Disclaimer: That's the plan, but plans change...

I didn't want to get into a new full version at this point, but a crash bug is enough to change my plans. We can't leave a crash bug out there in the official version. So I'd like to do a few small fixes and improvements to make it more worthwhile. I won't be getting into any monster graphical updates now, as that has far lower priority than the tyre physics which I must get back onto without delay, after doing some editor functions to help Eric with the track he is working on.

I did manage to get the join request working for Garage exit too (for already joined players who may have changed their car or settings). Not finished yet, partially working but it should be OK.

As all this stuff looks good, and with the programmers and other people here showing a lot of interest and coming up with good suggestions, I've decided not to rush this, and just work to the full release next weekend. We can do some tests between now and then, but I don't have to finish all these things by tomorrow afternoon!

Eric's fine with that as he is working on other things anyway. Victor is around next weekend too so it looks fine.

Tomorrow morning I'll round up where we are with these things and what I plan to do, and start a new thread for us to discuss that in the programmer section.

Quote from Ziomek21 :

+1, any more content is welcome, I don't understand why you would make something work perfectly fine and then throw it away.

Guys, did you see my post that there is a sunset config that has been deleted, temporarily, from the public version to allow the version to be online compatible with the official version for this weekend?

There is a maximum of four lightings per track. When the incompatible version comes out., there will be four. Eric has obviously chosen the four that he likes the most.

Last edited by Scawen, Fri 22 Jan 2016, 21:25. Reason : didn't need the last sentence ;)

I don't know all the details, because Eric doesn't send me a list.

But holes are fixed, all reported bugs should be fixed and there are various improvements. I guess you could make two comparison screenshots using a saved camera position and flick between them.

One thing you can't see is he has added a sunset lighting. I left that out just for now so the versions are compatible online.

Quote from Degats :

This isn't strictly speaking test patch related, but since 0.6K (or maybe even before), I haven't been able to see the Anti-aliasing option in graphics settings. The option for mirror AA is there, but not the global one.

Sounds like the post-processing shader is enabled in Misc options?

Quote from cargame.nl :

K3 handbrake warning bug.

I don't know what that message is doing there, maybe Wilk used the handbrake a couple of times .. I am sure it wasn't me because my screen was minimized

I didn't think of that appearing for remote cars online! They are probably not using K3, and they are using a handbrake on a single seater, which is no longer allowed (except for parking).

Quote from PeterN :

I just got my car stuck in a hole on Rockingham, seems to be some of edges slightly away from the track are not solid, especially the little green tarmac patches leading to paths.

Please can you post a screenshot or replay of this, so Eric can see where the hole is. Sometimes it is correctable by adjusting some of the track segments a bit. We just need to know as accurately as possible where it is.

It's only the Rockingham ones that would go OOS, because it's only Rockingham that will be updated in this patch - we aren't planning a car physics change at the moment. I've edited the original post to make that clear.

Many different types of change that Eric could make might make the replays go OOS. Most obviously if he changes any road surface.

I don't think we will archive the 0.6K replays in that case as one month isn't a long enough era to archive the results. But maybe a timing archive would be a nice to have when the new tyre physics update is out.

Quote from Mr_Lonely :

Since there has to be more finished s3 content lying around that arent affected by the new tire physics.

No, there isn't finished content lying around. There are tracks which are driveable but without the scenery completed, that kind of thing. If you imagine from Eric's point of view, the great thing now is that he can decide to work on a track, big or small, and when it is complete, we can release it. There's no more need to try and complete X tracks and Y cars, then when you are ready to release, realise that the tracks already need updating...

I think one at a time is a better way to do it, in such a small team, with the required level of detail being so high these days.

Quote from nacim :

...
However, I'm disapointed that there were no test patch before official release.
...

We did consider this!

But just on this one occasion, we couldn't really do a test patch. Because we would have had to sell licenses only for a test patch without any official version to support them.

So we did some good testing with a small group of beta testers and decided that would be the best way forward, although we do prefer full public tests.

There is the possibility of an update, e.g. in January, if there are enough issues to make it worth while, though incompatible updates are a bit annoying. If it's not needed, I guess Eric would like to get stuck into finishing some other content after the holiday.

Last edited by Scawen, Sun 20 Dec 2015, 20:14. Reason : "another track" -> "some other content"

Quote from nacim :

I thought you were trying to future proofing a bit more LFS on the graphical side while you catch up with modern graphical engines ?
...

Quote from Keiichi_Tsuchiya :

...
I don't know the way you going to push LFS graphics, I'm just trying to point on things that already used in game industry. Anyway, keep it up, Scawen!
...

Thank you for your comments and I just want to say that I don't have anything against cube maps (and other, more than 1 hemisphere methods). I am interested in doing things like that in future, possibly starting with doing that for the nearest car, and using hemispheres for the other cars. It seems to me that for the case of a car (or any shiny painted object standing outside on dark ground) the hemisphere method is just a good optimisation, because the ground is dark and the sky is bright. The sky reflection is important for the general look of a car in many circumstances.

I would say that omnidirectional environment mapping has a lower priority than a better shadow system, reflections on roads and some methods of blending textures that Eric has been waiting for. Along with the shadows I am interested in going for physically based lighting if possible and perhaps a generated sky which paves the way for changing weather. Aside from graphics, I am keen to get the tyre physics done so I can escape the curse of working with two separate versions and merging the changes. That will allow the possibility of separate graphics and physics threads and so on which would be inconceivably complicated to implement into two separate versions.

I sent an update to Eric today and this evening I was able to test the Vive.

It's not really too hard to set up. The instructions are pretty clear and it's just a bunch of wires to plug in. I used a tripod to mount a base station in a location where it can see the floor and my seated position. In my office room there isn't really space for room scale VR so for me it is fine to use only one base station. Lucky that because the first one I tried seems to be faulty. The base stations are very "pre-production" and there is quite a bit of noise / vibration from the base station (mounted on a tripod on the left desk - maybe the desk is amplifying it). A continual hum that is much noisier than my PC, for example. I guess these will be miniaturised for the production version.

For some reason I had to stop and restart the hardware and software several times to make it work, and eventually it started to work. I can run a VR demo where you can just walk around a bit, or in my case take a step in either direction.

The screen resolution seems good in the demo. I can still see the pixels as in the Rift but my first impression is this has higher resolution than the Rift DK2. However it is not really comparable until I can see the same thing in each headset (e.g. LFS). One other thing is better in the Vive - you can turn right around and the tracking stays for much longer. Presumably this is even better when you have two base stations. I have no idea how to access any other Vive software. I'm not too bothered about that, as LFS is my main concern. I plan to have a better look at the OpenVR SDK tomorrow.

Yes, I have received the socket adapter. I am finishing off some functions for Eric and then I will test the Vive. If it works then I'll have a deeper look into Valve's OpenVR SDK. I am hoping it will all work in a similar way to the Rift without too many nasty surprises! As we have some other priorities at the moment, I don't want to get involved if it's going to take more than a few days. One good thing about OpenVR is that it is supposed to support a variety of headsets, not just the Vive.

Got a reply from Valve. Looks positive so far, will let you know if I receive a Vive. I hope it'll be easy to code for.

It's good to support these things. VR and simulators go together very well.

For all the LFS users who aren't too interested in VR, don't worry. Eric and I have been continuing to work on some good updates.

No, this current update does not give shadows to layout objects. I agree the concrete objects look strange and unrealistic without shadows. It's tempting to dive right into a totally new shadow system that gives shadows to all objects in the world. That is something I want to do.

But I can do that at any time. Right now, Eric needs some new multiple texture blending features and it's important to give them to him with the highest priority, so he can complete a track we plan to include as S3 content. I was just surprised to find such a frame rate increase, right now with the current shadow system converted to HLSL and brushed up a bit.

I've been updating the vertex shaders for the track drawing system, so now they are in HLSL. At last I've been able to delete the old DX8 assembly language shaders. I'll be able to add some new texturing features for Eric to use in the editor. So far I have not added new features, but improvements and simplifications in the car shadow code and shaders have resulted in improved frame rate, specially when many cars are close together. Important on the grid or at Turn One.

That'll be fun to release and see a few people getting annoyed that we improved the frame rate again!

It's very important to improve the efficiency of the drawing system. For a start, many people did have problems at Westhill and can now enjoy it a lot more. I want to add dynamic reflections on the cars because it looks nice when the car is near high objects, trees, buildings, under bridges and so on. This depends on drawing an reflected image of the world as seen from each car. Clearly this will give the CPU and GPU more work to do, so it needs to be as efficient as possible.

Quote from Amynue :

https://developer.oculus.com/blog/upcoming-oculus-pc-sdk-0-7-compatibility-changes/

Thanks, they sent an information email about this. Our current official version will no longer work with the new Oculus runtime when it is released on 20th August. It seems to me the best plan for LFS is to finish the few remaining bugs and make the current version official as soon as possible, e.g. next week. Our Rift support in the test patch is based on Oculus 0.6, so it will work with the new runtime. Then I'll get on with the track shaders that can help Eric with the initial S3 tracks.

Quote from valiugera :

@Scawen, I found a mistake on Fern Bay. The wall is in the air, but Z is 0. Can you add ground level check the for new objects or add option for -Z?

I had to make that decision, to disallow Z values below zero, in order to reach the highest altitudes at Westhill. The autocross object data is highly compressed, to allow them to be passed quickly to joining racers. There is only one byte in the object data for altitude, so it can't go below zero. There are very few places in LFS which are below zero so this is the best decision. Maybe in future the sutocross object size will be increased, allowing more flexibility.

I don't think we can do that kind of thing. Eric's busy on the S3 tracks. I'd like to just get this improved performance + Rift support patch finished and released so I can get on with some other things that need to be done for the S3 update.

I am actually interested in releasing a track editor and providing a system to automatically download track updates when the creator updates their track and textures. I think the download would be a low priced paid service (like premium skins) even if the tracks are free (paid tracks are another thing entirely). But this is absolutely impossible to supply to S2 licensed users, unless we want to discard any possibility of making money. So it would have to be an S3 thing. This type of possibility is one of the reasons why Eric and I are working hard to get to the S3 license, because it will allow us to produce and release cool things for S3 licensed users.

Quote from nacim :

After using G3 for a while now, I still don't have problems to report, waiting for next test patch.

There's not much left for the compatible version apart from a couple of minor fixes. I've been focussing on the incompatible things for Westhill. In the last couple of days I implemented the custom start positions in the autocross editor, for the start grid and pit start positions (mid race join, etc). These allow you to make race tracks anywhere. Eric and Victor and I have done some online testing in the new Westhill and it's going well. I think another full version would just slow us down and annoy people. We're on the case finishing things and I have a couple more autocross things to do. We had fun testing online. There will be a public testing stage before the final release. We still have things to finish before that.

Hi, thanks for the comments! That compatible test patch (without Westhill) is getting quite close now!

On Friday I did editor updates to help Eric. Also I was pleased to finally get DX9 debug output out of the development version of LFS. Previously I only got it from the public version. I actually had to list the specific LFS.exe in the DirectX control panel, in a place that looks like it only applies to DX10/11. Very strange, but good to see some debug output!

Edge3147, it's true I don't often put 110% into this. I'm not really into that expression, but I suppose 110% means somewhere higher than a sustainable effort. Well that can happen right before a patch and when I'm really on something, do some late nights and so on, while ending up tired. But I prefer not to do too much of that and I do have my cycling now as well (and a short trail run each week and a little swimming in the summer). As for work, I prefer to cruise along at an enjoyable rate. After all, work is a massive part of life and life should be enjoyed as much as possible.

Actually after 3 years of cycling I don't have a bad back any more. After too many years without proper exercise, and after carrying something too heavy on one occasion, I ended up with a problem where something in my lower back kept slipping to the wrong place and often gave a lot of pain! So it turns out that running and cycling increase strength and bring great benefits!

Today I'll be looking into the hidden object files and seeing what else can be sorted for the compatible test patch.

Working on it

On 24th January I gave quite detailed info about what I had to do, to release the compatible open paths test patch.
https://www.lfs.net/forum/post/1880248#post1880248

I just came to have a look here and I'm a bit disappointed to see old style groaning and moaning. It really is a bit 'old'.

Yeah, LFS needs updates and we are working on it. Are the moaners and groaners working on their personal issues? Get a hobby, ride a bike up a hill, just do something. Try another sim if you like, or something in the 'real world'. Just don't come here and moan and groan! It doesn't make us work faster. It wastes your time and ours.

For the people who 'believe' that we are working on something...

Referring to my post of 24th January...

1) I've completed the hidden object removal system for the external TV cameras which are now done a much better way, with individual draw lists per camera, instead of switching the visible objects in a path-related way. So the hidden object removal system is near complete. There is one more thing to do with that, on open configs, something about seeing start lights on the selected 'X' config, when you are on the path from another config.

2) I've completed, yesterday, the lighting on the cars, so that they use the path lighting data when they are on open configs. So, for example, the car goes dark when in a tunnel. This involves creating lighting textures per path. Seems to work well.

The biggest remaining thing for the compatible test patch is to create a way to save hidden object files for the existing public tracks. After I've done that last thing in (1) then I'll look at generating those files. I also do editor updates for Eric, as he's still going round Westhill finishing things, making sure there is good collision detection on everything, etc. There are some other things to look at for the compatible test patch.

P.S. 1) When I said we don't use the help of others, I obviously meant in the sense of my work and Eric's work. We have a great deal of help from forum moderators, translators, community testing, hosting and support in so many ways.

P.S. 2) I never called cargame.nl a talentless fool. He runs a great server so there's no way I would have said that. I was referring to simple trolls who just pop up and make trouble every now and then, revealing their own issues.