I have followed this LFS-scene for a while now, and one thing makes me wonder. Why developers are still publishing a small and quite meaningless patches to LFS S2. It´s quite a ready product now.

So, is it about time now to move on and concentrate resources to making LFS S3. Or is already under construction? Or maybe cancelled. Hope not.

Define "meaningless."
If you mean the lack of content, well that is Eric's job, and we are expected to see many updates and hopefully some surprises with the comming patch Y. I think there is much more going on in the background than what appears on the outer surface.

S2 is not done. There are many features still in the works for S2 before it gets to Beta stage. Beta stage will be short, though, since we have been bug-testing since the beginning

Then, S2 final will come out. After that the devs will probably take some time to rest, and then S3 will be worked on. I wouldn't expect anything regarding S3 to surface for at least 6 months after S2 final, though. Most likely a year or so I'd think.

1. It's only one guy (Scawen) actively developing LFS, as he's the only coder. That's the main reason LFS development moves quite a bit slower than that of other sims.
   
2. Even those "small and meaningless" things are things that need to be done sooner or later, so why not do them now?
   
3. First they have to even finish S2, so no, it's not time to concentrate on S3. S2 is still a long way till it can be considered feature complete, and at least if the devs follow the original plan, S3 will be mainly about graphics updates and weather implementation. Physics and damage model improvements are very likely planned to be included in S2.

But actually, why don't you have a look at the 581762349 other threads discussing this topic?

Quote from XCNuse :

Define "meaningless."

Quote from AndroidXP :

First they have to even finish S2, so no, it's not time to concentrate on S3. S2 is still a long way till it can be considered feature complete, and at least if the devs follow the original plan, S3 will be mainly about graphics updates and weather implementation. Physics and damage model improvements are very likely planned to be included in S2.But actually, why don't you have a look at the 581762349 other threads discussing this topic?

Okay. if S3 is only graphics updates and weather implementation, and not as big leap forward as S1 was from S2, that changes the situation a lot. So S3 will be finished version of S2. I expected some bigger changes car- and trackwise mainly, when writing this thread.

And what I ment meaningless, was something like those countless language-versions, and features like send SMS from in-game.

And I simply don´t have time to read those 581762349 threads. :-)

S3 will be seperate of S2, S2 is still in its alpha stages as said before, S2 has a long ways to go before it can be official done and over with.

SMS from ingame is just something Victor did, Scawen didn't put much work into that I do not believe, but language is still something that matters, that isn't that pointless.

All we can do is hope for the most for S2 final.

Quote from Hartsa_550 :

Okay. if S3 is only graphics updates and weather implementation

Mostly, not only.
At least that is what people are expecting. In reality nobody knows anything about what S3 will be like.

Quote :

So S3 will be finished version of S2.

No. S3 will be LFS's next installment. S3 will be the final version of LFS, and therefore finished. S2 will have a final version and that will be the finished version of S2.

Quote :

And what I ment meaningless, was something like those countless language-versions, and features like send SMS from in-game.

Just because they are meaningless to you doesn't mean they are meaningless to others.
Besides, small changes often take a small amount of time to implement so I don't see the problem. And most of the work on the language versions is done by the translators. All Scawen has to do for that now is update his version.

Quote from AndroidXP :

It's only one guy (Scawen) actively developing LFS, as he's the only coder. That's the main reason LFS development moves quite a bit slower than that of other sims.

Is that really true? If that is true, then clearly LFS must be very far behind all the other sims by now. But I don't feel it is.

I think that when you work with a large team, you can sometimes progress at a good rate, for some time, as long as there is a lot of good morale in the company (and that is not a consistent thing). But later, as different people's code gets interwoven with other people's, and some others leave the company, and the new guy can't see what's what, and redevelops his section from scratch, and some things can't be updated because they affect too many areas and the company needs to call meetings all the time to try to figure out what's going on... finally they aren't progressing so fast any more.

Sometimes game development houses can actually go backwards. They can work on something for two years and what they have at the end is worse than what they started with. It's very hard to see how LFS can go backwards, because there aren't any other programmers to leave. And that's why LFS continues to go forwards while the development of other games and sims can sometimes just slow down or stop.

Quote from Scawen :

Is that really true? If that is true, then clearly LFS must be very far behind all the other sims by now. But I don't feel it is.

I think that when you work with a large team, you can sometimes progress at a good rate, for some time, as long as there is a lot of good morale in the company (and that is not a consistent thing). But later, as different people's code gets interwoven with other people's, and some others leave the company, and the new guy can't see what's what, and redevelops his section from scratch, and some things can't be updated because they affect too many areas and the company needs to call meetings all the time to try to figure out what's going on... finally they aren't progressing so fast any more.

Sometimes game development houses can actually go backwards. They can work on something for two years and what they have at the end is worse than what they started with. It's very hard to see how LFS can go backwards, because there aren't any other programmers to leave. And that's why LFS continues to go forwards while the development of other games and sims can sometimes just slow down or stop.

There's more...

http://www.simbin.se/main/technology.htm

http://media.iracing.com/downl ... ds_Irwindale_16001200.jpg

Quote from Scawen :

Is that really true? If that is true, then clearly LFS must be very far behind all the other sims by now. But I don't feel it is.

I think that when you work with a large team, you can sometimes progress at a good rate, for some time, as long as there is a lot of good morale in the company (and that is not a consistent thing). But later, as different people's code gets interwoven with other people's, and some others leave the company, and the new guy can't see what's what, and redevelops his section from scratch, and some things can't be updated because they affect too many areas and the company needs to call meetings all the time to try to figure out what's going on... finally they aren't progressing so fast any more.

Sometimes game development houses can actually go backwards. They can work on something for two years and what they have at the end is worse than what they started with. It's very hard to see how LFS can go backwards, because there aren't any other programmers to leave. And that's why LFS continues to go forwards while the development of other games and sims can sometimes just slow down or stop.

This matter is highly subjective. Yes, working with a small team is good in some way, as you just don't have other people to merge code with, or whatever, but it also brings numerous problems, the biggest being obviously that it cannot be as fast. It applies to everything, not only program developing.

What's interesting about that is that we've never ever seen something that we could call the perfect middle. It's either too slow, or either crappy because it was done by too many people. The day that a very dedicated team of, say, 40 or 50 people will do work as if they'd be 5, something might top LFS. Look at all the other things, they've all been produced by a huge team, and I personally get the feeling that they've tried to finish them as fast as possible.

Quote from Scawen :

Is that really true? ...

Oh, I hope I didn't leave the impression that I'd like more people working on LFS

Anyhow, yes, I think LFS development would be faster if more people would code on it, though overall it would depend very much on these coder's morale and discipline, if I can say it like that. If they were as talented and dedicated as you, then it could work, though I doubt it's easy to find someone like that . Coding alone and without stress of managing other people surely increases the quality and "coherency" of code, that's for sure.

In my opinion, a deciding factor is how the project is structured - mainly how many separate and independent modules it has. I, for example, am a part of a three man coding team, developing a web based enterprise resource planning system. Due to its nature, it is split in many rather independent modules, making the code easy to merge - thus in our case more coders equals faster development (to a certain extent). But I also know from experience that debugging/enhancing "foreign" code can be quite a PITA if a coder got ill or has something more important to do. That's where the set coding standards and coder discipline come in (which can be quite horrendous in stressful situations).

In LFS' case it depends, for example, on how much the graphics engine is coupled with the physics core and whether completely separate development/improvement is possible without interfering too much. In the right circumstances the time gained by parallel development of big complex chunks could far outweigh the time needed to merge the code. However, more people would certainly also mean a higher risk factor regarding bugs. I think LFS is a very special project and the quality of its code surely contributes to its success, and I'm not really sure if the decrease in quality would be offset by improving the most criticised point of LFS (slow development). That said, at this point in time where a very good and stable code base already exists, an extra coder would probably be less problematic than he would've been in LFS' early stages.

I think a lot of people seem to think that all games are developed in a matter of a few months to a year. That is simply not true for 'real' games. By 'real games' I mean games which have their own engines developed. For instance, EA pumped out a new F1 game every year for several years. Every single one of them used an old, out-dated engine (Physics and Graphics engines). (the very same ones the SCGT engines)

Now, when you look at a game which is being developed with its own engine, you see development takes significantly longer. For instance, UT2007 has been in development for several years (before UT2004 was release, IIRC) and it has an all-knew physics and graphic engine. (Yes, an FPS with a physics engine... sorta. :P )

Also, there is Will Wright's upcoming title "Spore." Spore has been in development since 2000. Nothing have been said about the engine itself, but seeing as how the game is completely unlike anything else out there, it probably has an all-new engine as well.

Both games have very large development teams and Very large producers backing them. To make a full game takes years of work. Fact is, most games out there run on someone else's game engines. The Unreal Engine has been used for dozens of FPS titles, as has the Half-Life engine. The ISI engines for several racing games, etc etc. Many of the 'games' that are released are no more than mods, really. (Most Unreal Engine based games are just mods in a standalone package, the EA F1 titles were basically poorly done mods, etc etc.)

LFS is it's own complete thing. It doesn't use another game's physics sound or graphics engine. Everything was made for LFS. That adds years to a game's development no matter how many coders you have.

Quote from Hartsa_550 :

So, is it about time now to move on and concentrate resources to making LFS S3.

That is EXACTLY what the small "meaningless" patches are. You might think that releasing the whole alfabet (not to mention all the test patches) is a waste of time, but you are wrong. Solitary coding for years on features that doesnt work well in the real world is wasting time (and you cant know if they work or not unless you try them on a large scale)
The developement model (release phases etc) used for lfs is optimal for getting things done. Its also optimal for me becuase I dont have to wait for years for S2, Im driving it already there are just a few features missing and its done (unless Scawen invents a new extended alfabet

Quote from MAGGOT :

as has the Half-Life engine.

I'm being pedantic, but half life 1 was actually built on a modified quake2 engine, the other games based on halflife 1 are mods of halflife 1, not standalone games.

OT: I haven't seen any real breakthroughs in all of the simbin/blimey/isi related games there all so similar to that old sports car mod based on F1 Challenge or whatever it was from memory, i would prefer the LFS way rather than putting new clothes on a tired old dog every year or so..

The only sim they have released that has some character is GTL but it still feels like the cars pivot around the center rather than on four axis ( you know what i mean)

So back OT yer keep doing your thing Scavier your way as your product imho is better than any current sim atm...

Quote from ev0 :

I'm being pedantic, but half life 1 was actually built on a modified quake2 engine

Quake1 engine actually. On a side note most of the engines are built on older engines. In software development it is rarely a good idea to start anything from scratch.

Quote from Mike85 :

I just wonder when will the damage model be completed. When are there gonna be animated pit crew? Etc....

Scawen should tell us what are his plans and a timeframe.

But at the same time I dont want it to be rushed to S2 final just because we want it fast. But I get a feeling that Scawen writes alot of code which is not used in the game right now.

What I, and probably most of the community wants, is an LFS with features to match rFactors for example. Then that would be a full-blown engine. It would be THE best racing game. Right now its just the best in racing physics.

Right now LFS is full of ideas, but they need to be put on a list and prioritised. We need to say what will be in the game and what wont.

Scawen you should read the wishlist at the top of this thread to see whats missing

And the best guidance for that is currently realism.

I would love to see LFS be a full-blown car simulator in the range of a flight sim. With all the switches and all that. That way it could even be used as an industry simulator for testing new cars.

But I guess thats overly optimistic and dreamy. And Scawen would need a team of 100 and a budget of millions to do that.

And in patch X we get false starts and some InSim improvements.

OK, false starts is ok. Its a feature. But all these InSim and OutSim and downloading skins... Thats not for racing really. We just want features.

I would recommend making a test patch with everything you have done so far. With all the code so far. And then we'll make it stable with testing. And see what comes out

Then you make some new code

I am SO GLAD you aren't in control, because LFS would be rubbish if you were. You have no idea, have you!

sorry for being offtopic...

...but @richo I think you should change your avatar, shouldn't you?

Quote from Mike85 :

OK, false starts is ok. Its a feature. But all these InSim and OutSim and downloading skins... Thats not for racing really. We just want features.

Erm, the InSim and OutSim doesn't directly benefit you, but without it stuff like the CTRA manager wouldn't work - and imo by developing this interface Scawen has outsourced some of the work to the community, who are more than happy to write stuff like managers etc.

Quote from Mike85 :

What I, and probably most of the community wants, is an LFS with features to match rFactors for example. Then that would be a full-blown engine. It would be THE best racing game. Right now its just the best in racing physics.

...what else should it be best in then? How can 3 men make a game with the best graphics? They're in a niche market of simulators, and going for the best physics seems to be the only way 3 guys can make something that can be compared to things made by big companies such as SimBin. LFS (imo) has the best physics, and netcode - yes, i wish LFS had better graphics and sound - but if those features were that important to me i'd be playing NFS or CMR...

The Skin Downloading is something i REALLY like, as while LFS doesn't have the best graphics, it can look really good with the right textures, AA etc, and yet the low-res skins let it down - now thats fixed, and LFS looks (imo) fairly good in it's own special way

Quote from Mike85 :

I would love to see LFS be a full-blown car simulator in the range of a flight sim. With all the switches and all that. That way it could even be used as an industry simulator for testing new cars.

But I guess thats overly optimistic and dreamy. And Scawen would need a team of 100 and a budget of millions to do that.

I'm sure that games like NFS, GT4 etc were made with a team that size and that budget, and yet they somehow have been beaten realism wise by in effect one man (Scawen).

Quote from Mike85 :

I would love to see LFS be a full-blown car simulator in the range of a flight sim. With all the switches and all that. That way it could even be used as an industry simulator for testing new cars.

nK Pro was a rather dramatic illustration that it doesn't work in a race sim and people don't want it. Most race cars are relatively simple to operate, if nothing else due to the fact the driver doesn't have time to start playing with stuff. Given that fact most things can be assigned to buttons. I think that trying to make a clickable cockpit in 3D is a mistake as well (in nK Pro you could operate the switches from the TV cam if you clicked in the correct place). Flight simulator gets away with using 2D panels still because given the nature of flying if you're having issues with seeing the 2D fixed panel against the 3D world it usually means you're doing aerobatics or are in trouble.

In FS I find the better option though I feel is to have a virtual (3D) cockpit but set to give the best view and having only the essential gauges in view. This solves the issue of the awkwardness of a 2D cockpit against a 3D world. Then I have the conventional panel on my second monitor, this gives additional instrumentation at a scale where I can read it and clickable switches. I'd love it if someone made some kind of gauges/panels for LFS so I could have all the info and overheads on my second monitor, the only thing you actually need when racing is the tachometer anyway

No real time commercially available simulator is going to offer a sufficient level of detail with current computing power.

Quote from Mike85 :

I would love to see LFS be a full-blown car simulator in the range of a flight sim. With all the switches and all that. That way it could even be used as an industry simulator for testing new cars.

That bit is only fun in flight sim when you do procedure simming, on a complex jet doing a full IFR , trying to fly precise approach and manage your fuel usage, where you have direct control on the plane for merely 10 minutes in a 10 hour haul.

But sim racing is all about direct sensory and input instead of procedures, it's more like VFR, and when you do VFR you just want to turn on the magneto and fly.

Quote from Scawen :

Is that really true? If that is true, then clearly LFS must be very far behind all the other sims by now. But I don't feel it is.

I think that when you work with a large team, you can sometimes progress at a good rate, for some time, as long as there is a lot of good morale in the company (and that is not a consistent thing). But later, as different people's code gets interwoven with other people's, and some others leave the company, and the new guy can't see what's what, and redevelops his section from scratch, and some things can't be updated because they affect too many areas and the company needs to call meetings all the time to try to figure out what's going on... finally they aren't progressing so fast any more.

Sometimes game development houses can actually go backwards. They can work on something for two years and what they have at the end is worse than what they started with. It's very hard to see how LFS can go backwards, because there aren't any other programmers to leave. And that's why LFS continues to go forwards while the development of other games and sims can sometimes just slow down or stop.

I do think, the only part where LFS could improve speed of development is content in terms of cars and tracks. Cause designing with given tools, especially texturing, is a work of Art which can be done by a crafted person and overlooked by some group-boss (would mean Eric changing way of working ).

Code to be done by a single person - with some help here and there - is imho a very good solution for clean, almost bugfree and working code.

Quote from ksa_land :

sorry for being offtopic...

...but @richo I think you should change your avatar, shouldn't you?

Its rubbishing EA not supporting Nazi's i think anyone with a drop of common sense could appreciate the humour...

I guess each to his own personally i think its bloody funny

i thought you were saying EA are Nazi's. although you might find it funny, i don't think some of our German friends will.
back to topic. LFS is moving forward every day imo. the devs are taking small confident steps and making sure everything works before they take their next small step. bigger companies like to release a product and then bombard their customers with huge patches. although it is a sign of progress, it makes you ask "how unstable was the original version that i payed for?" i think if bigger developers took a leaf out of the lfs devs book, most/all games would be better.

the development of lfs is the only part the should be "slow and steady"

Quote from Mike85 :

Scawen should tell us what are his plans and a timeframe.

No, have you EVER seen a game released on the date it were supposed to get released? And no, he shouldn't tell us all his plans, he might come to the conclusion that some of them just isn't doable or shouldn't be in the game at all.

Quote from Mike85 :

Right now LFS is full of ideas, but they need to be put on a list and prioritised. We need to say what will be in the game and what wont.

I'm sure Scawen allready have a list like that. And no, its not "we" that should say what should be in the game and what won't, thats the developers decition.

Quote from Mike85 :

all these InSim and OutSim and downloading skins... Thats not for racing really. We just want features.

uh, that doesn't make sence. The new InSim features makes it possible to control the servers better, and thus making better racing. Just look at the CTRA(STCC) servers.