On a side note, the time it takes to shift affects gearing choices. I recall that a world record run at Blackwood in the LX6 only used 3 gears (other than the initial start). With a much faster shift time, a Formula 1 race car will use all 6 or 7 gears or at least all but 1st gear, on a track like Spa.

Link to image of rear wheel force (in g's) versus speed for a old ZX11 motorcycle. If shift time was instant, you'd shift where the lines cross, since shifts take some amount of time, you'd shift at slightly higher speed. 1st to 2nd gear shift is at red line, then the rpm for ideal shift point decreases with the gear.

http://rcgldr.net/misc/shiftpoint.jpg

Last edited by JeffR, Fri 1 Jan 2010, 11:37.

I assume this is an indirect comment on the slow pace of development with a 3 person team. I recall back in late 2005 or early 2006 making what seemed like a pessimistic prediction that S3 would be released in 2009, in a post at RSC, and later in 2007 here at LFS forum:

http://www.lfsforum.net/showthread.php?p=586307#post586307

One of the issues I mentioned back then is if a flaw was discovered in some core part of the physics model, it would take a very long time to rewrite that part of the code with just 3 people, and this is essentially what has happened now with the tire physics. My guesstimate is this will end up being about a 10 month hit on LFS's progress.

In the meantime, S2 has ended up with more content than what was expected back in 2005, at least in terms of cars, so what we have now is more like S2.7 than S2.0. I'm not sure what the original planned differences between S2 and S3 were supposed to be.

Regarding the "10 years", a more appropriate time period would be 10 years after the July 2003 release date of LFS S1, and I'm sure that the final version of LFS will be released before July 2013. The main issue is that it will seem outdated compared to other racing games that will be released by 2013, in terms of physics as well as graphics. I still think there will be a fan base that likes the "feel" of LFS.

LFS has always been unique. There haven't been many cross-overs from the Papy/ISI community (most of these moved on to rFactor and the GTR series, a small number are iRenters now). Instead LFS has developed its own unique base of players, which has evolved somewhat over the years, but still draws a relatively large number of online players compared to other games.

To put this into perspective, how many 10 year old games are still being played? For the NFS series that would be High Stakes, with less than 50 'active' players in the community (iplounge, hsscoring). Grand Prix Legends is 11 years old, with a surprising amount of effort put into mods and visual updates for a game that was never that popular, although most of that work is completed now (JR Pearson is still doing a solo project working on the Isle of Mann track for GPL). I'm not sure how big that community is. EA F1 series players and modding community moved onto rFactor, so it's essentially gone. NR2003 activity is way down since the server was shut down years ago, and it was released about the same time as LFS started.

On a related note, how will the current players interests change over the next 3 to 10 years, and how will this affect the fan base of the current or future racing games?

Last edited by JeffR, Tue 29 Dec 2009, 12:25.

Quote from amp88 :

I've seen a few sites that extract newsgroup posts into HTML in a forum-like manner. I don't know if you could post on the extracted page and have that reflected in the newsgroup though.

This is how microsoft newsgroups work. You can access, including reading and posting of messages, to the newgroups via usenet or via html web sites (several initial links, but they eventually end up accessing the same newsgroups):

http://www.microsoft.com/commu ... groups/en-us/default.aspx

As mentioned in their web site, it's hard to believe two failures in separate systems and hard drives occurred at the same time, and there wasn't some form of archival backup that could be used to restore RSC to a state perhaps a week or a month old.

Quote :

control freaks

I still have bad memories of this from the earlier days. RSC was the only forum ever that felt they needed to punish their members by using a public warning system. I quit RSC after receiving one of these for posting a old (2002) video of an onboard lap at Spa spaf1.wmv claiming they could get sued for copyright violation, demanding I get a letter from Bernie Ecclestone and mail it to them before they would remove the public yellow card on every post I ever made at RSC, even though it's the team, track, or video maker owners that own copyrights, (which is why LFS can have a BMW F1 car even though it doesn't have the rights to the F1 series), and in spite of the fact that video had been posted at other web sites for years without issue before I posted a link at RSC. During this period, anyone posting questions about the public warning system was permanently banned.

Quote :

repository

Towards the end, RSC mostly served as a good repository of information, and it will be shame if they can't recover the data. For a while, there was some activity in the iRacing section, but that had virtually ceased by the time of failure, and there are game specific forums now, such as LFS.

Quote :

central hub

The newsgroup, rec.autos.simulators, served this purpose, but there's been little activity in the last few years, since there are now so many game specific and other generic forums.

Last edited by JeffR, Mon 28 Dec 2009, 01:31.

Quote from Torben :

You bought an S1/S2 licence for not even an beta Version of a game

Although the dev's call these "alpha" releases, these are "beta" releases, according to the the terminology used by the rest of the computing world:

alpha build of the software is the build to the internal software testers

http://en.wikipedia.org/wiki/S ... _release_life_cycle#Alpha

Beta is a nickname for software ... released to users for software testing before its official release.

http://en.wikipedia.org/wiki/S ... e_release_life_cycle#Beta

If you have a graph of rear wheel torque versus rpm for all gears, you'll see the lines intersect on the right side of the lines (except perhaps between 1st and 2nd). If shift were instant, you'd shift where these lines intersect. Since shift times are not instant, you shift at a bit higher rpm. The idea is to maximize the rear wheel torque for the speed near a shift point. This assumes a normal torque curve where the torque peak is significantly less than redline.

One of the early "are we bored?" threads.

http://www.lfsforum.net/showthread.php?t=10706

Quote from birder :

Racing games seem to hold their own as far as computer games.

I meant PC based racing games. Other than some NFS games (note the NFS series recently broke the 100 million total sold count), no racing game has made into the top 10 list of top selling games, and I don't think any sim-oriented PC based racing game ever made it to the top 50.

Using the NFS series as a benchmark, Underground 2 and Most Wanted sold 9 million copies, each, while Carbon and ProStreet about 5 million copies. I don't know numbers for Undercover, but assume it was less than 5 million. NFS Shift had sold 2.5 million copies by the end of September (From EA quaterly report), so it may end up reversing the trend. The online play in NFS series peaked with Underground 2, with a few hundred players online at peak times. Now there are probably less than 50 players online at peak times with NFS Shift, some of this is due to console sales splitting up the players though. LFS has somehow managed to retain a relatively high number of online players though.

I know the console games were and are very popular in the USA (where I live), are they also starting to catch on in other parts ofthe world?

Quote from ChiliFan :

Excluding drift servers we have 9 combos:
GTR at AS National
FOX at KY National REV
XFG & XRG at BL GP
TBO & GTR at BL GP
FBM at BL GP
XFG & XRG at BL Rallycross
GTR at KY Oval
FOX at AS North REV
TBO at WE International REV

I haven't played LFS in years, and based on this list the online play has really changed. Perhaps my memory is faulty, but when I played online, the server would change the track after every race, or perhaps every other race (only repeating the same track once). You'd have the equivalent of multi-class racing with some difficult to drive (before S2 version V) LX6's in the mix that were faster if you could manage to keep them pointed forwards, but mostly these followed the true sense of mutli-class racing where different class cars were running on the same track at the same time, so it was like a group of races within a race.

Quote :

old timers

I think there's a burn out factor for the older guys, some who have been playing games for over 10 years. I spend more of my computer time on forums than gaming, and I spend more of my free time with other activities than on the computer. The fact that LFS is stuck in the past with it's slow development pace isn't helping either, but that's another topic.

I wonder about the future of computer games. My gut tells me that the heydays of racing games and perphaps PC based games in general was 2003 -> 2007, and it's diminishing now. I'm wondering if iRacing will survive more than a few years, if rFactor 2 will reach even 1/2 the sales of rRafctor, if the NFS series end with the planned 2010 releases, if there ever be an LFS S3 (what was it supppose to have that S2 didn't?), if there will be any more TombRaider games, ... A lot of game production simple stopped or slowed to a crawl in the last 2 or 3 years. Already I've seen the demise of the Star Wars Knights of the Old Republic series (there will never be a III). Half-Life 2 Episode 3 will be nothing like the previous 2 games.

Quote from Ball Bearing Turbo :

LFS started as the most hardcore, realistic, no holds barred sim that was made at the time.

Actually the physics of S1 was quite flawed, but the slowness of most of the cars back then masked the issues, except for the LX6. GPL and NR2003 had better physics than S1. I had the impression that early versions of LFS had little input from the real world, based on things like a vicous rear end differential, something I don't think was ever done with any real car (although some 4WD vehicles use viscous coupling between the front and rear driveshafts). The GTR like all or nothing traction of the tire physics was another issues, but all of these issues were mostly fixed back with S2 version V.

As far as the old timers not playing LFS as much, it's because they are old timers. Some of us have been playing racing games for over a decade now, and the newness factor is gone, we're aging, and our interests change.

Last edited by JeffR, Sun 20 Dec 2009, 12:51.

So how many km on the Scirocco now?

Quote from ultrataco :

Anyone have any driving tips for SHIFT?

Techniques:

Drive like you were running a drift event. There's some hidden traction control assist that can't be turned off, making it difficult to accelerate while turning. High speed turns are generally power limited as opposed to grip limited, which is why I use 0/1 downforce settings for most tracks. Willow Springs with the Zonda R is a good track to learn this at, mostly turn 2 and 9 since they are long sweeping turns. For turn 2, I ease off the throttle, turn in sharply to get the rear end out, then go full throttle and use steering inputs to guide the car. Steering hard will increase oversteer a bit (as opposed to inducing understeer), scrubbing off speed. If the car oversteers too much, it ends up losing more speed than easing off the throttle, so there's a balance here. Turn 8 is flat out in 5th gear (with my savegame setup), just be sure take an inside line and allow the car to drift to the outside at corner exit. For turn 9, I ease off the throttle and downshift into 4th gear, aim a bit inwards, then let the car scrub off speed and drift back to the outside to setup the apex. I'm back at full throttle well before the apex. Video of Willow Springs in Zonda R demonstrating drift style of driving the game seems to favor:
http://www.youtube.com/watch?v=lBSg1eG_cys&fmt=22

Exploits:

Autoclutch off for fast launches - you get dragster like launches if you turn off autoclutch and traction control, if you use manual shift. Shift into 1st before the start, and hold rpms between pefect launch and redline, then go full throttle once car starts to move. Shifting doesn't need to use clutch, even with autoclutch off, but it's better to lift throttle on some cars to prevent clutch slippage. In my savegame, the MX5 is the only car that likes lift throtle.

Downshifting while at speed before braking - The game has the equivalent of an slipper clutch that limits engine braking. Some cars like the EVO IX setup for 10.0 events, downshift slowy under lift throttle, usually upsetting the car if the downshift occurs while turning. To avoid this, you can downshift quickly while at full throttle before braking during corner approach. For the Donnington chicane with the EVO IX setup for 10.0 level, I'm flat out in 6th and downshift into 2nd gear while still at full throttle just before braking.

Braking is faster with ABS off and downshifting while braking - For most cars in Shift, braking distances are shorter if ABS is off and you downshift prematurely while braking, unless the car has the slow downshift issue like the EVO IX.

Link to a savegame of the cars with the setups I use for online. I just updated these for patch 1.02. I don't use any of the new cars, but there doesn't seem to be anyway to delete them, so just ignore them. I use low downforce in all cars, except Murc and Z06 which are setup for short tracks. The Z06 isn't as good as it was before the patch, because maximum downforce was decreased from 60 to 24.

http://rcgldr.net/nfss/nfss.zip

My patch 1.02 "review":

Bounce issue - problem cars like the works Camaro and the McLaren F1 were "fixed". The tracks weren't changed so the bouncy spots are still there, and although the bouncing is a bit smoother, the cars can still end up bouncing off track in to a wall. Dakota Tri Oval and Rustle Creek still have the same "seam bug" where car body parts snag on some invisble seam in the track, causing the cars to bounce badly, usually slamming into walls.

Handling - Steering sensitivity seems decreased. The hidden traction control is decreased, affecting the works Z06 the most, which now requires more precise throttle inputs if you run all assists off. After adjusting to these changes, my lap times are about the same as before.

Grip - This wasn't changed directly. However the grip level is affected by the players system (fast systems with nvidia cars seem to get the most grip), so if your frame rate was increased by the patch, your grip level may also have increased due to the game's sensitivity to system performance.

Setups - The setup values were tweaked by patch 1.02. Max values are now working, but they also lowered the limits of some settings. The Z06's max downforce was reduced from 60 to 25, so it's not the super high downforce car it was before, good for the short tracks like Alpental and Tokyo Circuit. Tire pressures settings may need to be moved one click left to get them the same as before. The bars in the gear menu are working, but when setting gearing just use the top of the bars (max speed at redline), and space them somewhat evenly. The displayed ratio values are way off in some cases, but the top of the bars seem to be OK. The only car I had that really needed to be regeared was the MX5 I use for 4.0 racing, I had to move 4th, 5th, all the way left, and 6th 1 click from left.

Lambo Murc - it's fixed now so you can tune it. Zonda R is still faster at most tracks, so I'm trying max downforce setting with the Murc as an alternative to the works Z06 for short tracks.

Z06 - I noticed differential settings got changed to 0 0 80, resulting in a very oversteery car. Not knowing what else was glitched by the patch, I reset to default, then retuned the Z06.

Mods - I manually do this myself, using the unpacker, editing files, and then repacking. I remove headshake, and turn off collision and motion blur effects. So I just redid these.

Headshake - bootpersistent\cameras\cameraconfig.xml - parameter HeadPhysicsScale set to (0;0;0) (there are multiple instances of these in cameraconfig.xml).

Collsion Blur - igphaseactivate\cameras\collisionblur.xml - parameter maxblur set to 0.0

MotionBlur (cockpit) - igphaseactivate\cameras\speeddepthoffield.xml - parameter MaxBlur near_low="0.00" near_high="0.0" far="0.0".

Quote from Bose321 :

Are those cars PS3 only, or all?

All versions. It's a bit of annoyance, because the put those cars in your garage as opposed to the car lot, and you can remove them from your garage. When selecting cars for online play, you have to scroll past these 'collector' cars.

Quote :

Releasing games/beta's on a console before the PC

Somewhat separated teams from what I undestand.

Quote from G!NhO :

The reason they do that is probably because of piracy.

Since the consoles are fixed platform, the code is probably simpler. Regarding piracy, I recall some recent console game was cracked before it was officially released.

Quote from thisnameistaken :

That line is cheating and shouldn't be allowed.

The racing line is mostly giving a player the equivalent of more visible braking markers that are not as easily seen on a typical computer monitor playing a racing game versus the real world. I would compare it to the exaggerated tire sounds that are not realistic, but give a player a sense of when the tires are at the limits to compensate for the lack of forces that would be felt in a real world situation. From what I read, iRacing exaggerates the visual yaw (oversteer) from the in car view during a race, but yaw isn't really happening and it doesn't show up in the replays. NFS Shift exaggerates the yaw even more, and it does show up in the replays, so my guess is that they've exaggerated the tire's slip angle physics. There's a trade off between total realism and playablity in a racing game.

Last edited by JeffR, Fri 4 Dec 2009, 02:26.

Quote from JohnPenn :

Mosport ... nostalgia for GPL.

Dual view video of the GPL version of Mosport (Greger Huttu replay), with graphic updates from GPLPS installation pack:

http://rcgldr.net/gpl/gplmsltgh.wmv

Quote :

The video is without borders in fullscreen and thats what matters but your upload seems to be a low bitrate anyway.

If you're referring to the .MP4 video you can download from youtube, that's after youtube has compressed the video and lowered the bit rate. The orignal video was 1440x720p, 30 fps at 10 mbps, and a .WMV file. The youtube recommended H.264 format was coming out blocky looking when they first started doing 1280x720, so I switched to .WMV format. I don't know if the H.264 issue at youtube has been fixed yet.

Last edited by JeffR, Wed 25 Nov 2009, 07:29.

Quote from JeffR :

I thought the latest update reduced the excessive engine braking. Was that fix only done for the Spec Racer Ford? I'm not an iRenter, but follow what's going on at a few forums.

Quote from jtw62074 :

I just tried both again and the Skippy seems a lot better than it was before (default setup). Perhaps that's my imagination since it's been so many months since the last drive in it. Pretty sure I remember it being a lot worse than the Spec in this regard.

I did the one one free Radical trial a while ago. Since I didn't buy the Skippy I don't know about it, but read many posts that stated the SRF was much worse than the Skippy. As you mentioned ealier, lift throttle resulted in spin, not oversteer in the SRF (and perhaps the Skippy).

Before the recent update, rather than mess with setups, some players adjusted the throttle throw so that zero throttle pedal translated into 10% to 20% throttle input, and then adjusted the brake bias in the setup rearward to compensate when driving the SRF (or the Skippy?). I don't know if any players are still doing this with the current update.

The player size is 854x480, now with some black bars on the side. For about a week or so, the player size was 960x480, no black bars, and I uploaded one test video at 2:1 aspect ratio to test it, and during that one week or so, it filled the player window. However now, it's back to black bars on the sides and tops. The original size of the video is 1440x720 (which is what you get if you use the javascript to download the .mp4 file). Link to that test video:

http://www.youtube.com/watch?v=vC72qIiP57U

Is there anyway to force a specific window size? I don't see an option for resizing the optional window mode to a specific size. Seem the only way to force this is to set the desktop size to some fixed value (1280 or 1920), then go to full screen.

Note that 1920x1080i or 1280x720p HDTV from a cable company uses 38.8 mbps, but that 1080i is 60 fields per second, equivalent to 1080p at 30 frames per second. Are there any consumer camcorders that record at 38.8mpbs?

Quote from jtw62074 :

hate the Skippy. It's just massive off-throttle oversteer.

I thought the latest update reduced the excessive engine braking. Was that fix only done for the Spec Racer Ford? I'm not an iRenter, but follow what's going on at a few forums.

[QUOTE=hazaky;1303293]Bouncing ...I wonder why this havnt happened to me yet./QUOTE]Depends on the car. It happens in the last invitational event, time eliminator at Nordschleife with the McLaren F1. The McLaren is very bounce prone.

Quote from Kamrock :

As you have said the Ackerman is there to decrease the toe in corners... and as the perfect racing line includes all four wheels having as much grip as they can

All this Ackerman stuff is ignoring the fact that the outside tires have more downforce than the inside tires because the lateral forces translate into an outwards torque on the car. Because of the difference in downforce, the inside tires achieve maximum lateral force at lower slip angles than the outside tires, so you'd want some amount of toe-in at all times.

In a non-downforce car, the forces will be about the same regardless of the cornering radius. In a high-downforce car, the forces are much higher in larger radius turns due to the higher speeds involved. The setups for those cars will take this into account.

Another issue is that few real race tracks have small radius turns, and for those that do, the driving method used for hairpin like turns relies more on braking and acceleration than cornering speed at the apex.

Last edited by JeffR, Mon 2 Nov 2009, 00:38.

Quote from tinvek :

explain to me why rolling starts are so prevelant in road racing on your side of the pond?

Perhaps to reduce incidents, and reduce the strain on the car parts. I assume the days of drivers doing a standing start sprint to their cars, strapping in and taking off for the start of a race are gone also. I think USA motorcycle races are still done using standing starts.

I'll toss in my two bits here.

Note that profile = tire sidewall height / tire width. A wider tire is going to have a lower profile number even if the sidewall height is the same as a narrower tire. So it would seem that a better question would be the ideal rim size for a given width and outer diameter tire.

Tire construction can compensate for sidewall height. Indy Racing League cars have tall but very stiff sidewalls, and the slip angles involved at the limits for IRL cars are less than that of Formula 1 cars which in turn are less than the lower classes of racing cars.

One issue with "low profile" tires, is the larger wheels and the greater amount of weight, increasing unsprung weight. Since the key factor is ratio of unsprung weight to total weight the weight of the car is a factor.

I can only think of two examples: Light cars like a Formula Ford, Formula Mazda (Rousch), Caterham, seem to do best with 13 inch wheels, and fairly tall sidewalls. Most Porsche 911 track cars are running 18 inch wheels, with relatively shorter sidewalls, but still smaller than the 19 inch wheels you find on some street versions.

Formula 1 cars also have smallish wheels and tall sidewalls, but I don't know the actual numbers.