F1RS: Consistency.

> >I've decided what the best thing about F1RS is. I's not the fancy > >graphics, the great sound, or the superb handling. It's the fact that > >you can see the front wheels. Read on... > ne SNIP! SNIP! Here is a thought. I had GP1(World Circut) before ICR1. when my friend told me about ICR1 he was going banannas over the spinning wheels..Just loved seeing goodyear spin into a blur. So I had NO wheels and then, with ICR, WHEELS. I didn't miss them in GP1. But I think I would in ICR1/2. So why not an option. Everybody is screaming about making the arms/wheel of GPL a switchable item, why not the wheels on all open wheel sims. ABC, ICR,CPR,F1RS,GP2 etc etc. That way those comfortable with the wheels get their cake and those who don't want it aren't forced by a programming choice to be forever boxed in.dave henrie hi to all TPTCCers!

> Any comments? Well, I wonder about the effect on consistency of GP2's strange design decision (IMO) to allow time to compress/expand like an accordian in favor of maintaining frame rate. - Matt

Quoted text (29 lines)

>I've decided what the best thing about F1RS is. I's not the fancy >graphics, the great sound, or the superb handling. It's the fact that >you can see the front wheels. Read on... >A friend came round who's played GP2 a lot. >He saw F1RS, with me driving, and was amazed by my consistency. I was >clipping exactly the same piece of kerb, and just locking up slightly at >the last moment, in exactly the same way - lap after lap. He was amazed, >and gave me many compliments on my driving. He then pointed out how much >difficulty he had doing that in GP2. >What I had to admit to was that I was the same as him in GP2. I used to >get a bit fed up with it, because there were times when I went off, and >didn't know why or what I did wrong. This NEVER happens is F1RS. If I go >off I ALWAYS know what happened. >Eventually I worked out why. It's because I can see the front wheels, >and they are so accurately modelled. >You can SEE when you've locked up, you can place the car very accurately >in a corner, you can see when you're using too much lock and that's why >the car is understeering. Once you get used to the sim it all becomes >very natural, and because it's a sim you need this sort of feedback. >I wonder if you'll be able to see the front wheels in GP3? >I have to say "Well done UbiSoft". >--

I really don't agree with this. The reasons being... Modelling the front wheels is dumb because if you really were sitting in a formula one car, you would not be able to see them given that the monitor is only displaying a limited view ahead. Only if you had a monitor a metre wide would you be able to see them. However, showing them anyway wrecks the perspective you see, and it is difficult to appreciate how the road you are seeing is supposed to be ?? metres across when the front wheels you see are only 40 cm across and yet are supposed to be about a 1/10th of the width of the road. Its curious how so many people seem to agree that the perspective you get with GP2 (or GP1) is the best of the racing games (ie, the view out of the***pit looks so real and believable) but then also criticise it because you can't see the wheels like in other racing games. Perhaps these two factors are connected? Also, the fact that Geoff Crammond who undoubtedly is a genius at creating racing simulations, and is evidenced by the sense of being there in his games, did not include wheels with both GP1 and GP2, despite I believe having included front wheels with some of his earlier driving games, as well as undoubtedly trying to put front wheels on GP1 and rejecting it, should tell you something. Modelling the front wheels adds little feedback when driving. While the wheel may lift when you hit a curb, you know you have hit the kerb anyway because of the sound, and you are generally looking further up the track. Similarly, wheel locking is also indicated by smoke in the rear mirrors and also sound effects and these are both arguably more noticeable. Also, the wheels don't necessarily help you to "avoid clipping kerbs" because you can only see the inside of the tyre, not the outside which will be the first to hit the kerb, nor the kerb itself when you are close enough to hit it. The amount of lock shown by the front wheels is also very ambiguous and therefore unlikely to be of assistance, especially if you are using a wheel where you can feel how much you have turned it. Therefore, the bottom line is that the front wheels don't really give you any feedback you wouldn't be getting from other sources more easily, but seeing the front wheels also wrecks the perspective so that the monitor isn't representing what you would actually be seeing if you were sitting in an F1***pit. However, I do agree with you that it is much easier to drive consistently in F1RS than GP2. The main factors I attribute this to are: You do not need to be as precise with the steering in F1RS compared with GP2. What I mean by this is that in GP2, to turn into a corner properly without missing the apex, you would have to turn the wheel to a very precise angle to give you the right amount of steering. For instance, an angle of 45 degrees. If you only turned it 42 degrees, you would go wide. If you turned it 48 degrees you would turn in too sharply and cut the kerb (this is a hypothetical example). In contrast, F1RS is much more forgiving in this regard as the steering is less sensitive (ie, small differences in steering lock don't have such a great effect on the attitude of the car). Therefore, if you accidentally turned the wheel only 42 degrees, it wouldn't reduce the steering as much as in GP2, and you wouldn't go as wide. Only if you turned the wheel 39 degrees which represents a much bigger error would you not get enough steering and you would then go wide. Therefore, in F1RS, you tend to get close to maximum steering effect irrespective of the exact positioning of your controller and therefore, it is easier to position the car consistently. Secondly, once you have turned into the corner, it is easier to change the attitude of the car in F1RS than GP2. What I mean by this is that if you turn in too soon, for example, you can open out the steering and the car will go wider and the tighten up the steering again and the car will be brought back on line. In contrast, GP2 seems to give the car more momentum so that once it is set to go in a particular direction, it is more difficult to change that direction, PARTICULARLY WITHOUT UPSETTING THE CAR. This again makes it much easier to drive the car consistently through corners in F1RS. Finally, the better graphics on F1RS also definitely make it easier to drive consistently. For instance, I found that I could drive more consistently in GP2 when I would switch the preferred line on. This was not because I needed the line to show me the correct path around the course and rather, on some corners I would actually take a quite different line. However, the preferred line gave you a much stronger indication of your cars position on the track. For instance, when approaching a corner, without the preferred line you would just be presented with grey tarmac and you would have to judge the turn in point from markers such as trackside objects which were far away from your car and therefore, difficult to judge precisely. However, with the preferred line, you could see it turning into the corner and it therefore gave you a much better indication of just how close you were to the corner, when the corner was going to open up again, etc. With F1RS, it is worth noting that really everyone uses the preferred line by default since, if anything, it makes the graphics look more realistic. However, other graphical features in F1RS would also aid you in the determining the position of your car much better than in GP2. While the graphics in GP2 are undoubtedly good, few dispute that the graphics are better in F1RS and in particular, detail on objects which are close to the track are very good. This then allows you to do things such as recognise where in the corner you are much better, judge the speed of the car much better, and judge the position of the car relative to its surroundings. All of these things add up, making it much easier to drive the car consistently because you are much more certain when is the appropriate time to turn the car in, etc. Also, if the car does start to go off line, you are aware of it much sooner and can therefore correct the problem before it is too late. Any comments? Having said all of that, I think that F1RS is undoubtedly the most enjoyable racing game available at present, although, as many others have already said, it does have some minor faults with it which prevents it from being the ultimate sim. Lets hope that Ubisoft listens to the criticisms and patches accordingly. If so, F1RS could truly be a rival for GP3 when it finally comes out.

I agree that the having the front wheels is terrific. I missed them in GP2 and am glad that Ubi thought enough to have them in the best F1 sim around..... PAPA DOC Pierre PAPA DOC Legrand Infamous Pink Flamingo Pilot...

Great comments from both Richard and Dennis. Two great sims with striking similarities and striking differences. Both with flaws, both with brilliant aspects. Right now, for me, F1RS' better visibility, as Dennis points out, and many aspects of its physics model make it my current favorite. If I didn't have a Voodoo card, I'd be playing GP2 a lot, now that CPUs fast enough to run it well have come into the price range which I can afford. Alison On Sat, 24 Jan 1998 11:37:16 +1100, "Dennis" >>I've decided what the best thing about F1RS is. I's not the fancy >>graphics, the great sound, or the superb handling. It's the fact that >>you can see the front wheels. Read on... [snip] Alison

> > Any comments? > Well, I wonder about the effect on consistency of GP2's strange > design decision (IMO) to allow time to compress/expand like > an accordian in favor of maintaining frame rate. To my mind this is a much better solution than maintaining a 1:1 time ratio and suddenly finding yourself 20yds down the track in a single frame at a busy corner (ala papy's ICR). WRT 'canned spins', I can't imagine a program designer modeling the physics of a car in the real world, then reverting to a 'canned spin' when the end's swap round. Having now given F1RS a fair go, I can only say that it is a fine siumlator, badly let down by bugs and poor design in the menu and configuration screens. Keep rolling out the patches UBI, this product needs them. -- Suck The Goat

Quoted text (16 lines)

>On Fri, 23 Jan 1998 21:07:59 -0800, "Matthew V. Jessick" >>> Any comments? >>Well, I wonder about the effect on consistency of GP2's strange >>design decision (IMO) to allow time to compress/expand like >>an accordian in favor of maintaining frame rate. >>- Matt >Im very glad you posted this, I was afraid to be the only one to >realize this. That is a very crucial difference, IMO disqualifying GP2

Nope. This is a very well known "feature" of GP2 (and World Circuit). When processor occupancy goes above 100% you will sllooooooow waaaaaaaaaaaaaaaay dddoooooooooooooown. Just do a lap with a stop watch and you will see Einsteins Theory of Relativity demonstrated :-) You were certainly not alone there! I did as well. -- Header address intentionally scrambled to ward off the spamming hordes.

Quoted text (7 lines)

>>Well, I wonder about the effect on consistency of GP2's strange >>design decision (IMO) to allow time to compress/expand like >>an accordian in favor of maintaining frame rate. >Im very glad you posted this, I was afraid to be the only one to >realize this. That is a very crucial difference, IMO disqualifying GP2 >(though, when there was nothing better, I stuck with it for quite some >time).

I too have posted on this sporadically since it came out. I admit to not following the group consistently so don't know if a ng-consensus has been reached (when does it ever?). I hate this design choice. For this reason alone, I can't believe the creator of this game is so often called a "genius" here. If you set the fps for the majority of the track, you'll have the occasional time when it goes above 120% and things slow radically. More than annoying, it upsets timing that is crucial to driving. Set it low for this eventuality and you'll be playing with a framerate a few fps lower than otherwise possible. And I find framerate to be even more important to me in racing sims than flight sims. Further, you'll find that time is slightly accelerated! Test it, you'll find that a 1:20 lap actually takes only, say 1:18. Maybe not enough to upset timing. But annoying nonetheless. Bottom line is: for me, I play at lower detail, and lower framerate than I would have to if it used a variable framerate design like most sims. BTW, F/A-18 Hornet similarly plays with game time.

>Further, you'll find that time is slightly accelerated! Test it, you'll find >that a 1:20 lap actually takes only, say 1:18. Maybe not enough to upset Huh, I thought it was the other way around. I thought that GP2 slowed down to make sure every frame was displayed. Making the actual time on a stopwatch longer than the game lap time. It didn't? -- Header address intentionally scrambled to ward off the spamming hordes.

>>Further, you'll find that time is slightly accelerated! Test it, you'll >find that a 1:20 lap actually takes only, say 1:18. Maybe not enough to upset >Huh, I thought it was the other way around. I thought that GP2 slowed down >to make sure every frame was displayed. Making the actual time on a >stopwatch longer than the game lap time. It didn't? Usually what you said is the case, ie you set an optimistic framerate, get high PO, then the game slows and your watch reads longer than lap time. But what I referred to above was when you anticipate this and instead set a low fps to keep PO always below 100%, you'll have times where the PO drops around 50% and a lap actually takes less time than elapsed stopwatch time. Me, I think games like this should be real-time, period. Is it that hard to do? Sim programmers response??

OK good I'm glad you were specific about this. I agree completely. By contrast F1RS displays in real time as far as I know. So obviously the bad part is whenever there are pauses the framerate will pause, but your progress around the track does not. Kind of like warping in Warbirds or Airwarrior. I have had pauses on turns and ended up in the gravel with not hope of preventing it. I will point out though that for me with P5-200MMX, 32MB ram, 3dfx and a partridge in a pear tree, pauses of this type are *VERY* rare. Usually if I run the game after connecting to my ISP without re-booting (reset button after a proper W95 SHUTDOWN. -- Header address intentionally scrambled to ward off the spamming hordes.

Quoted text (14 lines)

> >>Further, you'll find that time is slightly accelerated! Test it, you'll > >find that a 1:20 lap actually takes only, say 1:18. Maybe not enough to upset > >Huh, I thought it was the other way around. I thought that GP2 slowed down > >to make sure every frame was displayed. Making the actual time on a > >stopwatch longer than the game lap time. It didn't? > Usually what you said is the case, ie you set an optimistic framerate, get > high PO, then the game slows and your watch reads longer than lap time. > But what I referred to above was when you anticipate this and instead set a > low fps to keep PO always below 100%, you'll have times where the PO drops > around 50% and a lap actually takes less time than elapsed stopwatch time. Me, > I think games like this should be real-time, period. Is it that hard to do? > Sim programmers response??

In my experience, this is wrong. When the PO is below 100% (anywhere below 100%) then GP2 is in real time. I'm sure I would have noticed otherwise!

<...> > In my experience, this is wrong. When the PO is below 100% (anywhere >below 100%) then GP2 is in real time. I'm sure I would have noticed >otherwise! On my machine when PO is below 100% (like 50-70%), GP2 runs faster than real time. I've used a stop watch to verify this, and also a friend of mine reported the same behaviour. The only "reason" I can think of is that after GP2 installation I've overclocked my machine from 200 to 225 (and the bus speed from 66 to 75), could this confuse GP2 ? Then again, my friend hasn't changed his configuration... -Teemu. --

Quoted text (11 lines)

> <...> > > In my experience, this is wrong. When the PO is below 100% (anywhere > >below 100%) then GP2 is in real time. I'm sure I would have noticed > >otherwise! > On my machine when PO is below 100% (like 50-70%), GP2 runs faster than > real time. I've used a stop watch to verify this, and also a friend of > mine reported the same behaviour. The only "reason" I can think of is > that after GP2 installation I've overclocked my machine from 200 to 225 > (and the bus speed from 66 to 75), could this confuse GP2 ? Then again, > my friend hasn't changed his configuration...

I think it is a result of small inaccuracies in the model adding up. At a fundemental level you have 'blocks' of code which must be completed, the timescales for these may not happily fit into a 1:1 representation (like, try making the number 10 when you only have addition of the integer 3). GP2 delivers the frame rate you set for it. On a 120s lap (real time) at 20fps there are 2400 frames delivered, if latency is present (ie lower than 100% occupancy) then tollerance on the idle function is within 0.004% per frame (no doubt a lot less per itteration), but of course that all adds up over a lap. The alternative, is to keep a close eye on time as measured by the system clock and adjust the required duration inputs into the physics model. This way you achieve 1:1 time ratio (within some miniscule error over (say) 60s) at the expense of fidelity in space, i.e. at high occupancy you suddenly find yourself traveling twice as far in a single frame as you normally would (i have to say that usually for me this results in my car wanting to occupy the same piece of space as the car ahead). -- Suck The Goat

Lets do the math :o). 1:18 is 78 sec, 1:20 is... 80, yeah that's right :o). It looks as the error is around 2 or 3 percent. Not much really and hardly noticable. What it means is that when driving on the straight and your speedo meter is showing 300 kph, you're actually at around 307 :o). No matter what PO you have at average below 100% the diff should not be more than around 3 percent. This is easy to check for anyone. There are machines around now that can average 50% PO, and if the real clock then shows half the time of the GP2 clock, I'll eat my hat :o). I bet the difference is still only about 2 to 3 percent. /Christer, have 800 hours in GP2 below 100% :o)