Quote from SimulatorRental.com :

Hopefully in the future, LFS, will have a color pallet value allocated as a background color (clear), that can change without changing the whole graphics color.

In the meantime I could knock you up a webpage where you upload a PNG skin with transparency, play with a colour selector to get the colour you want, and click a button to download a coloured JPG version.

I've been meaning to do it for a while, but it's been one of too-many things I want to get to

Right, so, I made a simpler layout with 3 checkpoints and a finish, with an insim split over each one. I've added the replay and the layout if it's needed. I did two laps, and the log is below.

The good news is that the gap looks consistent That was my big worry from earlier in the thread.

(The first UCO time is the time it takes from /restart to green lights + reactions + inertia)

Sorry, 'raw' is just a dump of the whole packet by byte. SPX is a split, LAP is the finsih, and UCO is an insim checkpoint. The lines directly below the raw are the timing information extracted from each packet.

I'm a bit confused as to what's supposed to happening where with that track, so I just dumped the lot in case you could figure it out just from the timings

I'll try on a simple test track later on to get my head around it

I did a straight 2 lap race with a log for UCO, SPX and LAP if that's any help

Cheap sums it up perfectly, Niko. I thought the new owners might be better at seeing the bigger picture, but alas they're solving aero-related passing problems by adding... more aero. Genius. F1 is one step away from having drivers deviating from the normal line in order to collect virtual items for powerups. *weeps*.

But at least BTCC starts tomorrow

Quote from Scawen :

No, but what I am saying is that a good tyre physics model should produce a tyre that reacts in a reasonable way, similar to a real world tyre. So it should be able to reasonably simulate a lorry.

I appreciate that this is both "far future" and only "don't see why not", but racing trucks has been on my wishlist since before I even found LFS... it's made my day that your response to his question is anything other than "no chance"

I've always imagined the extra wheels could cause you some programmatic grief, but then, you've had the bike hidden in there for a while and the LXs' tracks are different front-to-back, so maybe it's not as big an issue as I fear

Quote from KeiichiRX7 :

Toss a disclaimer on it. Under construction.

You get to decide the minimum acceptable quality of your releases...

I did a quick test, but route checkers don't work how I thought they did - when you edit the id, it reorders them all so you can't have two with the same id. Still, that part is doable in insim as sinanju says.

PiranMOTO servers live again

I didn't solve the problem, but I did find a way around it. Early on, one of the things I tried was moving the port range. This worked, but only for the first port in the range, none of the others. Today I thought to try putting each one on a port that wasn't part of a range, separated a bit in number, having its own port-forwarding rule.

That's done the trick, although it doesn't help me work out why it broke in the first place

But, it's worth adding to the list of things-to-try when servers don't respond

Sorry, I said 'start' but I meant 'start and finish' in the same direction.

If you use split 1 and 3, you can have them in the same place in opposite directions without people being able to cheat the lap. If you use route check points, you can use any/all splits like that.

The track you've posted above could be done like this, with what we've got already: split1 and finish line in the same place, opposite directions, and two route checkers in each of the loops. (just one needed to enforce the loop being driven, or use two to also enforce the direction the loop is taken).

The cars having to start in the same direction would be the only limitation, I think, but I prefer that way anyway

I don't know anything about the single-pass stereo technology, and I'm sure nvidia know what they're doing... but that description they give has a whiff about it.

Thanks! The new gearshift debounce works beautifully with my crackly switch - I'm able to get the value down from 250 to 100 now, and can drag my finger for as long as I do when I'm not conscious of it without any double-shifts. No more excuses on the track

Will do, but I'm afraid it can't be until I get home from work, around 7/8pm UK. My engines thank you in advance

Quote from Scawen :

I'm pretty much finished now, need to get back to the graphics work. At this point I'll only fix serious issues, although your suggestion is probably good. The only thing I have on my list is to allow down to 0.05 on the view lock filter time.

Sorry to ask late, but I never usually remember until it happens and this reminded me... I'm hoping it might only be a matter of changing a constant.

Is it possible to increase the gearshift debounce upper limit?

(My downshift doesn't click any more, it fires constant static when held down instead. The debounce takes care of it most of the time, but 250ms is just a tad longer than my hand sometimes lingers on the button, apparently, as I often get double downshifts if I don't concentrate on making my downshift a click instead of a squeeze).

Quote from GT4_PL :

Please make increase 30 to 100 selected objects in editor layout?

If they're going to exactly the same place in the new layout then I have an insim that can help. If you want to move them about afterwards, that'll be in the next version sometime in the future

I meant to last night, but meatspace happened again *sigh*.

Seems to be fixed, I couldn't get it to happen. Cheers!

Quote from Draggo :

Ahh Avast, that explain everything. My favorite antivirus that flags everything as a virus.

Not true! There were multiple viruses/malwares running on the last avast-protected computer I came across, and it certainly didn't flag any of those

The mini is Scawen mentioning some distant-future 'maybe' thoughts. Seems reasonable. But wait, what's that noise? Did I see movement over there?

Yep. Here come the 31 misc cars: 10 people with feature requests related to the maybe; 10 people with feature requests a step beyond the maybe; 10 people requesting the same couple of cars that make up 99% of the cars-we-want-to-see thread, and, finally, 1 person who saw someone programming once, thought it looked easy, and now can't understand why LFS doesn't release a new fully-featured version every other week. "Has Scawen thought about using such-and-such entirely unsuited technology?" the last one wonders aloud as the sound of gentle weeping fills the air...

Quote from o000o :

Do you have the pedals showing? maybe you are not getting full throttle.

Many a time I have spent ages tweaking my top gear trying to match others' top speeds, only to find that my pedal was impeded a tiny bit and I was at 99% throttle, not 100%. I leave the pedal display on permanently now for quick checks

Ciccio: When the pedal is at full, there is a white bar across the top of the pedal display so that you know it's definitely pressed as far as it can go.

Sorry, I didn't grok the S part of SPR, I think I've only ever done it in multiplayer as I was playing with the insim detection on my debugging server. I'll have a go in single player for reference if you still need that by the time I get back home.

I can dig out or make one, but not until I get home from work... likely 7/8pm tonight.

My servers have become part of the no-reply count in the multiplayer list. I've run most of them for a couple of years now with this exact setup, but since a week or two ago they've not shown up in the multiplayer list. If I try to connect directly by name they time out. (I have DCON excluded from AV and firewall shield and have tried with those both disabled also.)

I suspected a problem with ports/forwarding so I changed all the ports and added new rules, now one will connect (PiranMOTORaceOfChampions) but not the rest. I've checked all the ports, UDP and TCP, from a website and they're reported open, and each check shows up in the relevant DCON with 'Accept: <IP>'. They all show up in the LFSW host list, LFSRemote can connect so packets can get through.

But, when I try to connect directly I see 'connecting to host' on the client and 'Guest wants to join' on the DCON but it doesn't then show an 'Accept: <IP>', and I get 'connect timed out' or 'clientconnect: connect failed' on the client. That would suggest that DCON isn't receiving the request that the client sends directly to the host (as directed by the master server), but why not if the ports are open and working?

I can't think what to do next to solve this, any suggestions?

Edit: Also upgraded to T3, both client and DCON.

Last edited by Racon, Sat 9 Mar 2019, 18:14.

I had all my driving lessons, passed my test, bought a banger and stalled it three times on the forecourt before someone thought to mention to me that petrol cars require a bit of throttle as you move off from standstill. I'd learned in a diesel, been taught never to touch the throttle until the clutch was fully engaged, and nobody thought it worthwhile to mention that the two types of engine require different approaches. Literally 5 minutes after that revelation and I was on the motorway for the first time ever, following my boss in his very expensive porsche as he was driving like a dick. "I thought you'd have been raring for a burn", he said. *rolls eyes*

Come join the club matey, no shortage of people beating me these days

Yes, sounds like a good idea