Makeshift Shuffle Club - Time Trials & Testing for club car lists - all welcomeOpen

[tarnheld]

Here is the list after @Mires testing:

* 1.06 Toyota MR2 1600 G '86
* 1.22 Alfa Romeo Giulia Sprint GTA 1600 '65
* 1.87 Toyota Celica XX 2800GT '81
* 2.89 Mazda MX-5 (NA) '89
* 3.83 Nissan Skyline 2000GT-R (KPGC110) '73
^ 5.74 Toyota 2000GT '67
v 5.88 Honda CR-X del Sol SiR '92
* 6.16 Nissan Fairlady 240ZG (HS30) '71
^ 8.79 Audi quattro '82
v 9.05 BMW 2002 Turbo '73
* 9.46 Mazda Savanna RX-7 GT-Limited (FC) '85
^ 11.02 Chevrolet Corvette Coupe (C2) '63
v 11.20 Honda Prelude Si VTEC '91
* 13.17 Buick GNX '87
* 13.67 Toyota Supra 3.0GT Turbo A '88

Bold are cars that switched positions in the list. They are pretty close TT-wise, so it is no real surprise.

EDIT: The Corvette seems faster than the Prelude TT-wise, so this could be a fluke in the FT system, they will swap again with more testing.

Maybe we should discuss the list ordering from a fundamental viewpoint, as there are some problematic cars in sensitive positions. I'm not so worried about the MR2, but the 2002 and the Savannah are tricky handling cars that pop up mostly for more funky drivers in the shuffle races. They will have hard time with them, and are more likely to get one of those in the next race. This too can lead to the frustration discussed above (NSX 6th/Evo 3rd).

These problems can not be solved by more testing, as there are faster cars that are more tricky to handle in many lists (Cosmo,Europa,A310, Autozam,Stratos). Part of it makes shuffle fun, but it can be frustrating too. I don't think we should cut these cars because of this, but we should be aware of it, and think about how to make this more unlikely. It's not easy to prevent this, as the car assignment is pretty random and depends on the number of racers -- if there are 8 racers, the funky drivers will get car #8 more likely, but if there are 9 then it will be #9.

So the list is volatile at the car-pairs above, but more testing might just swap the cars more, and come to a rest with arbitrary ordering. We might just as well fix the ordering here by vote.

BTW, @Mires: it doesn't matter if you take some extra turns, as long as you do it consistently, as your times are not compared to the other testers, the times are just used to find out which car is faster for you.

 

[Mires]

What extra turn


Anyone interested in Motorcycles, should have a look at this game.





www.gtplanet.net/forum/threads/ride.317203/

lol, why you all have this MOTO thing in mind, if you hear/read about that layout? never had moto, never will, it is just a possibility which game allows not officially, but allows.

Ok, nvm, cancelling that entry for Tsukuba, holding till i'll have time 💡

 

[watermelon punch]

@Mires - I was just saying I think the extra turn at Tskuba is for the motorcycle races.
I have NO IDEA why the "motorcycles in gt6" thread is controversial. I was making a joke. 👍 Because it seems odd that it would cause controversy. There's no controversy.
Some people like motorcycle racing too.

* 1.06 Toyota MR2 1600 G '86
* 1.22 Alfa Romeo Giulia Sprint GTA 1600 '65
* 1.87 Toyota Celica XX 2800GT '81
* 2.89 Mazda MX-5 (NA) '89
* 3.83 Nissan Skyline 2000GT-R (KPGC110) '73
^ 5.74 Toyota 2000GT '67
v 5.88 Honda CR-X del Sol SiR '92
* 6.16 Nissan Fairlady 240ZG (HS30) '71
^ 8.79 Audi quattro '82
v 9.05 BMW 2002 Turbo '73
* 9.46 Mazda Savanna RX-7 GT-Limited (FC) '85
^ 11.02 Chevrolet Corvette Coupe (C2) '63
v 11.20 Honda Prelude Si VTEC '91
* 13.17 Buick GNX '87
* 13.67 Toyota Supra 3.0GT Turbo A '88

I think the Del sol is going to be faster for most people on most tracks than the Toyota '67...
I think mainly because of the stock gearing.
Someone else could probably comment on that better than I could.

With the quattro 82 & Bmw 73... Quattro definitely should be considered the faster car.
I'm not TOO concerned with the wonky handling of the BMW... but mostly because the Fairlady 71 and the couple of cars up (slower) in the list from it, are not necessarily "easy" cars anyway, and they're definitely a tad slower than the BMW 73.

I wish someone would do a test with these cars at Monza (or Monza 80s or Monza no chicanes)... and come up with some kind of baseline for that... so we can compare the difference.

I would do that... but I STINK at Monza. So I wouldn't be comparing the cars fairly for that purpose.

EDIT: The Corvette seems faster than the Prelude TT-wise, so this could be a fluke in the FT system, they will swap again with more testing.

Maybe we should discuss the list ordering from a fundamental viewpoint, as there are some problematic cars in sensitive positions.

Absolutely we should consider these things.
I wasn't as concerned about things like this in the slower lists, because the cars are slower. If that makes sense.
You have a slip in the Autozam or the Vintage list Europa, or even the A310... and you have a chance to rectify after a mistake.

The historic list is a tad faster...
OTOH, the cars are also overall a bit more "exciting" too.
Even the best handling cars in the Historic list are... have limits.
Either they are plodding or lack torque or overall power, and I think that balances it out.
Example: The Miata is nice and easy to drive... but still, it won't be able to easily stomp someone driving with minimal mistakes in the Skyline 73 or Fairlady 71.

The Corvette vs. Prelude
The Corvette might corner better on some tracks given certain banking/camber (or none). But I don't think it's overall faster than the Prelude, given racing conditions.

Append:
I'm also not worried about cars with peculiar handling way down in the list - faster at #10 or #11... because by the time there are 11 cars being used in the race, you have #1 & #2 cars on the track as well... and they will certainly be faster than those, even with mistakes by drivers using them.

Edit 2:
I don't think the RX-7 is "so bad" in difficulty.
Not comparable to say Stratos or even the Luxury Mercedes.
Not saying it's a picnic to drive. Just that it's relatively not quite as difficult... And I think I'm able to say what's hard, because I make lots of mistakes.

 

[snowgt]

I think the Del sol is going to be faster for most people on most tracks than the Toyota '67...
I think mainly because of the stock gearing.
Someone else could probably comment on that better than I could.

The 2000GT needs to be kept at high revs (which on some tracks might not be possible). It's certainly the better handling car, but the CR-X del Sol is pretty fast down the straights. I guess it depends a little bit, if an average driver finds the 2000GT to be hard to drive fast.

With the quattro 82 & Bmw 73... Quattro definitely should be considered the faster car.
I'm not TOO concerned with the wonky handling of the BMW... but mostly because the Fairlady 71 and the couple of cars up (slower) in the list from it, are not necessarily "easy" cars anyway, and they're definitely a tad slower than the BMW 73.

I wish someone would do a test with these cars at Monza (or Monza 80s or Monza no chicanes)... and come up with some kind of baseline for that... so we can compare the difference.

I would do that... but I STINK at Monza. So I wouldn't be comparing the cars fairly for that purpose.

I might do that tomorrow, if it's needed.

The Corvette vs. Prelude
The Corvette might corner better on some tracks given certain banking/camber (or none). But I don't think it's overall faster than the Prelude, given racing conditions.

My impression was that you have to wrestle the Corvette around the turns, and while the Prelude is understeering quite a lot, it's still easier to drive. Give the Corvette a straight piece of road though and it leaves the Prelude far behind. I was over 1s faster in the Corvette at Motegi East and that's not really what you would call a fast track. Maybe I could run also these two cars at Monza to illustrate the difference. I admit though, that it's not easy to wring that performance out of the Vette.

 

[watermelon punch]

The 2000GT needs to be kept at high revs (which on some tracks might not be possible). It's certainly the better handling car, but the CR-X del Sol is pretty fast down the straights. I guess it depends a little bit, if an average driver finds the 2000GT to be hard to drive fast.

Given the average driver in racing conditions...
I think the 2000GT seems difficult to get out of sticky situations because of that gearing.
So that it's good on straights might not even be a boon on tracks with long straights because of the down-side of trying to get out of turns or maneuver in traffic.
This is my thought on the matter, having experienced similar situations with cars in racing conditions... driving cars which do well on straights.

I might do that tomorrow, if it's needed.

I think it would be acceptable to only run tests at Monza with the "questioned" cars. (ie: no need to do all cars. )

My impression was that you have to wrestle the Corvette around the turns, and while the Prelude is understeering quite a lot, it's still easier to drive. Give the Corvette a straight piece of road though and it leaves the Prelude far behind. I was over 1s faster in the Corvette at Motegi East and that's not really what you would call a fast track. Maybe I could run also these two cars at Monza to illustrate the difference. I admit though, that it's not easy to wring that performance out of the Vette.

At this speed of car, Motegi East can still be considered a miniature "fast track", as there are more straight bits relative to turns.

"not easy to wring that performance out of the Vette"
this is the relevant point ^
Similar issue as with the 2000GT & Del Sol.

The boon one gets on straight bits with the Corvette will wind up negated for the average driver in the turns or in heavy traffic... as the driver attempts to not bump when door-to-door through turns, or tries not to dip wheels into the grass or sand and adjust to the wimbly wombly sway and lean. Compared to the Prelude which you only really have to worry about under-steering off-track... but it's probably more forgiving about sand in tires anyway!

The Corvette is VERY sensitive to sand and grass compared to many cars!! 💡
I haven't specifically tested it against the Prelude on this point... but I suspect it is a lot more sensitive to the sand.

I noticed this in the past enough times that I tested again and scheduled the spec 1-make at Laguna Seca recently...
As is well known, I have BIG pet peeves about racing at Laguna Seca.
The potential for off-roading is huge there and makes me crazy!

Because many cars I believe are NOT punished enough for off-roading in the sand!
I knew this Corvette 63 is well punished for putting wheels in the sand at Laguna Seca... and probably every track with kitty litter sand too.
So I think it will be more difficult to recover after a significant mistake in the Corvette 63 compared to the Prelude. 💡

Right now @bender2011 (miata_1) is testing these cars at Suzuka.
For those who don't know - this is my spouse - who has much less experience in these vintage road cars with peculiarities. He prefers mostly sports cars, nice handling RWD.

Suzuka
2:44.143 Corvette 63
2:44.920 Prelude 91

His thoughts:
He's not sure the Prelude is slower, even though the Corvette might go faster on straights.
He's skeptical of his lap times in any case. Because he doesn't prefer the handling of the Prelude... so he thinks some people who prefer FF would do much better too.

He said that if he were given the choice between the Corvette 63 and Prelude 91 to race at Suzuka with 8-12 other cars he would be near - if he wanted to place better, he would choose the Prelude 91 every time, because of the ability to maneuver and react to events on track and the movements of other cars.
Even if he would be able to pass on the straights in the Corvette... he says it would be preferable to have the stability of the Prelude 91 in racing... easier to avoid accidents or pass cleanly in corners when others make mistakes.

He thinks some people might prefer the Corvette for personal preference of always preferring RWD cars. But he thinks even those people, given a choice, might choose the Prelude if they want to win.

And note: For the average driver - races are almost never like Time Trials.
The average driver gains positions mainly by attrition in any given race... when someone else makes a mistake, to take advantage to pass.
Because the average driver is not going to compete with aliens on lap times, and will be too close in speed to other average drivers to take advantage of any slightly better lap time under racing conditions. The difference will be too little to make good on that tiny edge.

This is normal racing (gaining by attrition) - if everyone were in the same spec car with equal skills.
We try to simulate this with the shuffle system, so EVERYONE is able to "race", not just "hot lap time trial". 💡


APPEND:
bender2011/miata_1
Suzuka
2:47.052 Del Sol
2:49.086 2000GT

 

[tarnheld]

Corvette vs. Prelude: Strange thing is that for those who are faster in the Corvette, it's a big difference of about 1s, whereas those that are faster on Prelude, it's only a minute gap. The big gaps for the Corvette show mainly on high speed or longer tracks: Motegi East, Rome Rev., Suzuka.

 

[John Wells]

Can't race this week going away

 

[watermelon punch]

Corvette vs. Prelude: Strange thing is that for those who are faster in the Corvette, it's a big difference of about 1s, whereas those that are faster on Prelude, it's only a minute gap. The big gaps for the Corvette show mainly on high speed or longer tracks: Motegi East, Rome Rev., Suzuka.

That makes sense.

Also to consider... aliens won't spend many races in 12 & 13 because:
a) that means there are 12 or 13 people in the race (not frequent).. and unlikely for multiple similar aliens to be given these cars together to notice the issue
or
b) in a parity race aliens will be unlikely to have a parity assignment of 13 or 12

So cars that high in the list need to be suited well for the average driver who might actually get these cars on a regular basis with other equally average drivers.

It would be of greater concern if the Prelude & Corvette were car #3 & 4
(3 & 4 being cars that most people spend time in regularly, regardless of their performance level)

 

[WhiterunGuard]

@Mires I love that extra turn at Tsukuba, I always use it when I'm picking the track for my series (if I choose 'Kuba).

 

[watermelon punch]

@Mires @WhiterunGuard
Someone has to see if it dirties the tires like @snowgt mentioned about the extra lane at Silverstone.
If it doesn't dirty the tires, there's no other reason it couldn't be used, in theory, I think... right?

Are there other places on tracks that are paved and dirty the tires?

 

[WhiterunGuard]

I'll go into more thorough testing tomorrow but from my previous races it doesn't seem to affect the grip at all...which is awesome...but like I said I will confirm tomorrow.

 

[Mires]

@Mires @WhiterunGuard
Someone has to see if it dirties the tires like @snowgt mentioned about the extra lane at Silverstone.
If it doesn't dirty the tires, there's no other reason it couldn't be used, in theory, I think... right?

Are there other places on tracks that are paved and dirty the tires?

I did run some laps and doesn't seem to dirt the tires.

 

[watermelon punch]

Cool.
Are there any "track boundary issues" in this offshoot turn?
Like do we need to make a pictorial visual aid for this?

 

[Mires]

Cool.
Are there any "track boundary issues" in this offshoot turn?
Like do we need to make a pictorial visual aid for this?

I don't think it's necessary as it looks like other parts of track. Only, i think, the green asphalt (or, is it a grass? Not sure now) between normal and extra track should be mentioned as off boundaries.

 

[snowgt]

@watermelon punch

Just gave these cars a quick run at Monza NC as per discussion above.

2:12,419 Honda CR-X del Sol SiR '92
2:13,192 Toyota 2000GT '67
2:14,063 Nissan Fairlady 240ZG (HS30) '71

Reverse order of what I got at Motegi East. I guess that goes to show how much on one end of the spectrum Monza NC is. (and that Motegi East is rather at the other end, in my opinion, although by far not as extreme)

And also these two…

2:09,563 Honda Prelude Si VTEC '91
2:07,573 Chevrolet Corvette Coupe (C2) '63

Does that mess up the ranking sheets even more…?

 

[amarynceos]

Finished testing at Laguna Seca finally (offline)

1:53.455 -- Toyota Celica XX 2800GT '81
1:52.556 -- Alfa Romeo Giulia Sprint GTA 1600 '65
1:52.210 -- Mazda MX-5 (NA) '89
1:51.982 -- Nissan Skyline 2000GT-R (KPGC110) '73
1:51.805 -- Toyota MR2 1600 G '86
1:50.737 -- Honda CR-X del Sol SiR '92
1:50.567 -- Toyota 2000GT '67
1:50.119 -- Nissan Fairlady 240ZG (HS30) '71
1:49.274 -- Mazda Savanna RX-7 GT-Limited (FC) '85
1:49.162 -- Audi quattro '82
1:49.106 -- Honda Prelude Si VTEC '91
1:49.007 -- BMW 2002 Turbo '73
1:48.633 -- Chevrolet Corvette Coupe (C2) '63
1:47.262-- Buick GNX '87
1:46.864 -- Toyota Supra 3.0GT Turbo A '88

I was hoping to get a clearer spread at a longer track, but I still got a tight cluster with the bolded cars.

Regarding our run with the Honda del Sol and Toyota 2000GT today:

At Motegi, they were fairly evenly matched. Although I was keeping up with @LongbowX mainly due to the draft, the Toyota's sluggish launch off of the final hairpin, and a few bobbles here and there, I think I could have taken a non-alien driver without too much trouble. If I were out of the draft, I doubt I could have done more than match pace, if that. Very slight edge to the del Sol, I think? Maybe? Obviously LB always has the edge in pace on me, but I'm not that bad at Motegi East, either.

At Deep Forest, I felt the Toyota was definitely quicker; once LB was in clear air, I had no difficulty in maintaining or even pulling a gap over his Honda, even with some sloppy driving on my part. The lack of tight corners played a significant part.

There's also the factor that most of us are already used to the 2000GT from the original Legends list, whereas the del Sol is a new factor -- apart from three laps at Tsukuba and Laguna, I had not driven it before, so it was harder to push in it.

 

[watermelon punch]

@amarynceos

I definitely expected the 2000GT to show its strengths at Deep Forest Reverse because it can keep speed there most of the time.
I didn't expect it to be as better than the Del Sol as it was though, so that was definitely a good test.
Also factor in though, where you & LB started on the grid at Deep Forest, in relation to the other 3 cars. How much did that matter? @LongbowX

At any rate, I think these 2 cars are INCREDIBLY close. And the Fairlady 71 is also VERY close to those 2.

It's a conundrum.

I'd be more concerned if the really problem cars were 1-2-3.

And I know some of you think the GTA 1600 is faster than the Celica 81.
But 9 out of 10 drivers who might participate in any given race are going to wipe out on lap 1 with the GTA on a lot of tracks if they try to push to get all it has to give.
The GTA 1600 has a very steep threshold where it stops forgiving - abruptly.

And 2 out of 3 aliens are going to drive with gusto in the Celica ... and not pay as dearly as any driver (alien or otherwise) who takes risks in the GTA 1600 and comes to a bad end. Or drives cautiously and fails to get the most out of it.

But the GTA is such a nice car.
It's not nearly as crazy handling as the MR2.
It's just that the GTA will bite you by surprise!!

The BMW is rather wonky in handling.
But it's the RX-7 that catches people out sometimes if they push it too far.

 

[LongbowX]

I think overall the Del Sol was better at Motegi, and the 2000GT was better at Deep Forest. The 2000GT is all about maintaining speed, so it'll do well on faster courses and the Del Sol on slower. I'm not sure it's really possible to balance that? It will just be a shuffle luck of the draw.

Personally I'd be happy to never drive the Del Sol again because its exhaust drove me crazy.

 

[watermelon punch]

Well, let me tell you, the Europa on Cape Ring is just as weird sounding as the Skyline 67.
For some reason... well, I'm squeaking the tires is what's going on, but it's more than that, it gets this willowy sound. Kind of sounds like a mourning dove that's upset, or sick pigeon.
It's not a horrible sound - it's not loud, it's just that it's so weird it's very distracting.

 

[tarnheld]

The Fairlady gets up to speed pretty quick, but cannot keep up with the Del Sol on longer straights. I noticed this on Deep Forest, i could keep up with @LongbowX even with sloppy cornering at the last corner, but along the final straight the Del Sol had the better top speed. You can see that too in @snowgt's tests at Monza.

So the ordering is difficult, i fear the math breaks down here. If there is need for the chopping block, and to make matters even more complicated, i suggest adding the Lotus Elan -- might make a decent #1 ...

 

[snowgt]

The Fairlady gets up to speed pretty quick, but cannot keep up with the Del Sol on longer straights. I noticed this on Deep Forest, i could keep up with @LongbowX even with sloppy cornering at the last corner, but along the final straight the Del Sol had the better top speed. You can see that too in @snowgt's tests at Monza.

So the ordering is difficult, i fear the math breaks down here. If there is need for the chopping block, and to make matters even more complicated, i suggest adding the Lotus Elan -- might make a decent #1 ...

Actually the Del Sol is almost always better on faster tracks as it's a lot better in acceleration (and also top speed). The 2000GT has good top speed, but it takes too long to get there, but it's superior to the del Sol in cornering. The only really slow track where the del Sol would have an advantage, is a stop-and-go one with lots of acceleration involved. I'm leaning towards an order...

2000GT - del Sol - Fairlady

...as a compromise, because it seems to be like rock-paper-scissors depending on track with these three.

Good idea about the Elan (and not only just because it was one I had), BUT...

a) the Lotus Elan was almost equal with the Giulia GTA, the MX-5 and the MR2, actually so much so in terms of corner speeds, straight line speeds and overall lap times that you could start a racing series without further balancing with these cars right away, and...

b) unfortunately it runs into the limiter way too early.

Otherwise I would have included it in my testing results. Pity, really.

The math never breaks down, it just shows that the data is inconclusive.

 

[watermelon punch]

Oh how I wish the Elan wasn't so flawed in its transmission.
It hits the limiter on MOST tracks.
That's the ONLY reason it's not made it to any lists.
It is a GREAT PITY.

But that's one of the cars I was considering putting on the "One Make List".

And I agree... we just have to compromise the best we can with some of these.
There's just too many variables it does become luck of the draw depending on the track.

I'm looking at the list now, the order the spreadsheet puts it in:

1. Alfa Romeo Giulia Sprint GTA 1600 '65
2. Toyota MR2 1600 G '86
3. Toyota Celica XX 2800GT '81
4. Mazda MX-5 (NA) '89
5. Nissan Skyline 2000GT-R (KPGC110) '73
6. Toyota 2000GT '67
7. Honda CR-X del Sol SiR '92
8. Nissan Fairlady 240ZG (HS30) '71
9. Audi quattro '82
10. Mazda Savanna RX-7 GT-Limited (FC) '85
11. BMW 2002 Turbo '73
12. Honda Prelude Si VTEC '91
13. Chevrolet Corvette Coupe (C2) '63
14. Buick GNX '87
15. Toyota Supra 3.0GT Turbo A '88

And this is how I think the list probably needs to look:

1. Alfa Romeo Giulia Sprint GTA 1600 '65
2. Toyota Celica XX 2800GT '81
3. Mazda MX-5 (NA) '89
4. Nissan Skyline 2000GT-R (KPGC110) '73
5. Toyota 2000GT '67
6. Honda CR-X del Sol SiR '92
7. Nissan Fairlady 240ZG (HS30) '71
8. Audi quattro '82
9. Mazda Savanna RX-7 GT-Limited (FC) '85
10. BMW 2002 Turbo '73
11. Chevrolet Corvette Coupe (C2) '63
12. Honda Prelude Si VTEC '91
    
13. Buick GNX '87
14. Toyota Supra 3.0GT Turbo A '88

And I know whenever we put this list on a dodgy camber track with 10 cars in the race, someone's going to think that the BMW is mis-ordered.
Because given a track like Autumn or Matterhorn, or even Nurb Nord, the BMW will be a handful compared to the RX-7 & Quattro.
The BMW has a nice speed comparatively, but one does have to nursemaid it quite a bit.

There's just no way around this like LB said... It's just luck of the draw sometimes.
Every car is going to have some peculiarity that makes it better or worse on some track or other.

For example the Vintage Skyline 67. For the most part this car is a stinker compared to the others in the list. But there are a few tracks where it's not so bad at all.
Same with the Retro Accord 88... not so bad in comparison on a tighter track with hinky camber... but on most tracks it's a real bummer comparatively.

In the vintage list we run into this with the 507, Cosmo, & Lancer.
It all depends on the track. Those 3 are really quite close in racing conditions, IMO.

In the Low Power list the Volvo Estate & Skyline 62 are great if you happen to get it when the low power is put on a big track. But Autumn Ring Mini or Eiger, good luck with that.

The Legends Europa was surprisingly fantastic at Cape Ring ... and on the circle. But not sure I would do so well with it at Deep Forest.

Append:
RE: GTA 1600 & Celica.
I'm not TOO concerned about an alien getting the GTA and sweeping or something. YES, the person who came in 2nd and gets the Celica MAY be at a disadvantage on some tracks to the person who won the last race. But it's not always going to be equal drivers (ie: 2 aliens) that get 1 & 2.
Second, there are distinct gaps I see in the Historic List that would make it unlikely for someone, even an alien, to win in the Alfa GTA, given even 6 or 7 cars in the race, and especially if some average driver who likes the Miata gets it, and a fairly competent driver gets the Skyline 73 or Del Sol.

I see the gap points here like this...

1.) Alfa Romeo Giulia Sprint GTA 1600 '65
2.) Toyota Celica XX 2800GT '81
small gap
3.) Mazda MX-5 (NA) '89
4.) Nissan Skyline 2000GT-R (KPGC110) '73
small gap
5.) Toyota 2000GT '67
6.) Honda CR-X del Sol SiR '92
7.) Nissan Fairlady 240ZG (HS30) '71
medium gap
8.) Audi quattro '82
9.) Mazda Savanna RX-7 GT-Limited (FC) '85
10.) BMW 2002 Turbo '73
11.) Chevrolet Corvette Coupe (C2) '63
12.) Honda Prelude Si VTEC '91
larger gap
13.) Buick GNX '87
14.) Toyota Supra 3.0GT Turbo A '88

So basically, anyone getting #1 & #2 cars are going to feel the disadvantage against equal or better drivers getting 5+.

But I think mostly there's never going to be a time where the person in #6 or #7 who drives the race with no mistakes is going to lose to the #1 or #2 car, no matter who's driving them.

If the person in car #9 makes some big mistakes or crashes, and gets beat by the #1 or #2 car - there's no mystery there & no way they can feel like they've been given the shaft & blame the car. Hullo, you have a big mistake or several, you're going to get passed by anyone in a slower car who's not made blunders.

Kind of like my race at Silverstone in the Volvo Estate against @amarynceos in the PAO. Sure he was at a disadvantage on that track in the PAO, and the Estate definitely had more speed.
But I stink at Silverstone in ANY car, and I made huge blunders losing momentum.
It wasn't any surprise to me to come in last.
It was my own fault, I stink at all configurations of Silverstone, and probably wouldn't have placed well even if I'd been assigned the Copen... or the Cube for that matter. Maybe I would've finished before the PAO in the Copen, maybe not. But I didn't have a snowball's chance in a volcano to podium in that race on that track.

 

[tarnheld]

Here is the list as of now, i have reintroduced 4 once-controversial TT times that are not so controversial at all after all the discussion here...

0.79 Alfa Romeo Giulia Sprint GTA 1600 '65
1.05 Toyota Celica XX 2800GT '81

2.31 Mazda MX-5 (NA) '89
2.86 Nissan Skyline 2000GT-R (KPGC110) '73

4.99 Toyota 2000GT '67
5.06 Honda CR-X del Sol SiR '92
5.40 Nissan Fairlady 240ZG (HS30) '71

8.41 Audi quattro '82
8.62 BMW 2002 Turbo '73
8.69 Mazda Savanna RX-7 GT-Limited (FC) '85

8.95 Chevrolet Corvette Coupe (C2) '63
9.01 Honda Prelude Si VTEC '91

12.14 Buick GNX '87
12.71 Toyota Supra 3.0GT Turbo A '88

You see the clustering in the FT values, i think the first, second and third cluster has been resolved. Fourth cluster is also pretty close with average gaps ~0.1s but min./max. TT gaps of ~1-2s -- explained by the peculiarites of these cars. This cluster can also ordered like @watermelon punch's proposal with the Savannah in the middle without skewing the TT gaps.

The only problematic ordering is Corvette vs Prelude. As i said before, the Prelude is slower on average by more than half a second, and the maximum TT gap is 0.267s -- that means of all 10 tests done the best win of the Prelude vs. the Corvette was only by a quarter of a second! That win was on Deep Forest -- not a really short track. So i would say the Prelude is the slower car.

So my proposal for the final list is:

1. Alfa Romeo Giulia Sprint GTA 1600 '65
2. Toyota Celica XX 2800GT '81
3. Mazda MX-5 (NA) '89
4. Nissan Skyline 2000GT-R (KPGC110) '73
5. Toyota 2000GT '67
6. Honda CR-X del Sol SiR '92
7. Nissan Fairlady 240ZG (HS30) '71
8. Audi quattro '82
9. Mazda Savanna RX-7 GT-Limited (FC) '85
10. BMW 2002 Turbo '73
11. Honda Prelude Si VTEC '91
12. Chevrolet Corvette Coupe (C2) '63
13. Buick GNX '87
14. Toyota Supra 3.0GT Turbo A '88

All in all this will be a lucky draw list -- but all lists are in a way -- shuffle racing at it's best!

 

[snowgt]

Hmm. I understand up to a point that you're concerned about cars that are hard to drive. But, once you do decide that you want to have also some of those cars on the list, I don't understand how you hope to avoid that the less experienced drivers will ever get into them by moving them towards the end of the list.

How does it make anything better, if you rank a car that's faster and easier to drive than its competitor (quattro vs 2002) higher (closer to the top) on the list...? Whoever finished 10th in the previous race will not be thrilled that he will get a car he will struggle with even against his immediate opponents from the last race (and those are the ones he will most likely have to fight against in the next race). A mediocre driver (let's just assume someone finishing 10th is for this example), even when he drives almost flawlessly for his skill, will not be able to keep up with somebody with the same skill in the quattro, never mind that he is much less likely to get all out of the BMW to begin with.

Actually, a car that's hard to drive should be the one ranked towards the top of the list by your logic, because you keep arguing people make more mistakes in them.

In shuffle mixer races you simply cannot exclude the possibility that someone will possibly be assigned a car that's difficult to drive in the next race. That's also the beauty of it. It's a bit random, but you make it unfair, if you give someone a double disadvantage.

For parity races, you are allowed to choose between 2 cars anyway, so that shouldn't be a problem either. You just have to take care that there's not 2 really hard to drive cars next to each other.

As a last note I have to say that the BMW 2002 and the RX-7 are by no means anywhere near as difficult to drive as maybe an MR-S was or an A310 or Stratos is, so those worries are, in my opinion, a little unfounded.

 

[Patrick8308]

You are correct @snowgt but I guess you are forgetting one point and maybe @watermelon punch was aiming that.

The lower a car is positioned on a list the lower the chance this car will be driven at all. Remember that most of the time it's 6-9 people racing. So if hard to drive cars are moved towards the end of a list there's a big chance they are missed out in the usual races. The result is the racers who finished 7th or 8th in the last race will get an easy to drive but still kind of fast car for the next race.
Of course this all doesn't work if 10-16 racers would attend an event. But this seems to happen rather seldom.

 

[snowgt]

You are correct @snowgt but I guess you are forgetting one point and maybe @watermelon punch was aiming that.

The lower a car is positioned on a list the lower the chance this car will be driven at all. Remember that most of the time it's 6-9 people racing. So if hard to drive cars are moved towards the end of a list there's a big chance they are missed out in the usual races. The result is the racers who finished 7th or 8th in the last race will get an easy to drive but still kind of fast car for the next race.
Of course this all doesn't work if 10-16 racers would attend an event. But this seems to happen rather seldom.

Well, if the aim is that a car is not driven at all, it shouldn't be on the list (rather than out of the order determined by testing times).

In the end, the list has to work with all possible amounts of participants, and not sort of break down when more than 9 people take part.

I know that in an ideal world not only would the cars be faster the higher the ranking in a list, but they would also be at least as easy to drive as the slower ones on the list, because the theory says that the less skilled drivers are more likely to get them assigned. But we all know it doesn't work like that. There's really fast cars that are easy to drive and rather slow cars that are quite a handful to keep on the road. But that distribution is more or less random and, in my opinion, we have to live with that and not try fixing something with causing another problem.

Assuming that one particular driver will be haunted by 1 car on 1 list being at 1 specific position in the list, sounds to me like a lot of if's. The shuffle system is specifically designed to be robust against that kind of thing and to provide variety and the number of different cars everyone gets to drive is proof for that.

 

[tarnheld]

Some more data tidbits:

Here are the gaps for all tracks and tests of the Vette vs Prelude, negative means Prelude is faster:

Tsukuba: -0.261
Deep Forest: -0.267 -0.229
Laguna Seca: 0.473 0.153
Suzuka: 1.416 0.777
Motegi East: 1.227
Rome Reverse: 1.463
Monza NC: 1.99

So Prelude is only faster on tighter tracks, and only by a small factor.

And here they are for the 2002/quattro/Savannah cluster (negative means first car is faster):

2002 vs Savannah

Deep Forest -0.517
Suzuka       2.114
Tsukuba     -0.003 -0.081
Motegi East  0.595
Suzuka East  0.319
Laguna Seca -0.267  0.227

Savannah vs quattro

Deep Forest -0.12
Suzuka      -0.744
Tsukuba     -0.475 -0.413
Laguna Seca  0.904  0.112
Motegi East -0.425
Suzuka East  1.006 -0.723

quattro vs 2002

Deep Forest  0.637
Suzuka      -1.370
Laguna Seca -1.131 0.155
Tsukuba      0.478 0.494
Motegi East -0.17

Looks like rock-paper-scissor to me.

EDIT: Regarding the chance of getting a car:

Don't forget we have Shuffle and Parity races:

For shuffle the chance looks something like this:

The bump comes from the number of racers, it is mostly less than full grid. There are some specialties to consider here, as new racers entering late get #3 per default, so this number is special. In any shuffle the number of participants is special too, as it's more likely for any racer to get this number, if he makes a serious mistake and ends up last. But this number is essential random too, so we cannot take this into account too.

In Parity, it's mostly determined by racer funkyness, but the chance for any random driver looks like this:

It's more skewed for higher car numbers.

So basically we should look for an overall list fairness.

 

[Mires]

I plan to make an ultimate online test for myself, including all cars on all tracks. Lol. Not sure if or when i will be able to finish it, but i can post the results in parts if it will help you guys in future decisions for another car lists..

Silly idea, time consuming, but hope worth

 

[tarnheld]

Regarding the Savannah/quattro/2002 cluster, i would propose the ordering:

Savannah
quattro
2002

Reasoning:

After intense

on the data above (i always wanted to use that smilie ), it seems that the Savannah is a bit faster than the quattro, quattro vs 2002 seems undecided with a little edge of the quattro. Savannah seems also faster than 2002. The sandwich of the quattro also helps in Parity races, so you will have the choice of the tamer or more exciting ride for the whole list:

1. Alfa Romeo Giulia Sprint GTA 1600 '65
2. Toyota Celica XX 2800GT '81
3. Mazda MX-5 (NA) '89
4. Nissan Skyline 2000GT-R (KPGC110) '73
5. Toyota 2000GT '67
6. Honda CR-X del Sol SiR '92
7. Nissan Fairlady 240ZG (HS30) '71
8. BMW 2002 Turbo '73
9. Audi quattro '82
10. Mazda Savanna RX-7 GT-Limited (FC) '85
11. Honda Prelude Si VTEC '91
12. Chevrolet Corvette Coupe (C2) '63
13. Buick GNX '87
14. Toyota Supra 3.0GT Turbo A '88

Coincidentally this is the ordering from the spreadsheet when i remove the 4 controversial TT times reintroduced above.

Numbers don't lie...

EDIT: sorry, i fell for the inverse list logic, faster savannah and slower 2002 have to be switched, changed the list according to that.

 

[snowgt]

Some more data tidbits:

Here are the gaps for all tracks and tests of the Vette vs Prelude, negative means Prelude is faster:

Tsukuba: -0.261
Deep Forest: -0.267 -0.229
Laguna Seca: 0.473 0.153
Suzuka: 1.416 0.777
Motegi East: 1.227
Rome Reverse: 1.463
Monza NC: 1.99

So Prelude is only faster on tighter tracks, and only by a small factor.

And here they are for the 2002/quattro/Savannah cluster (negative means first car is faster):

2002 vs Savannah

Deep Forest -0.517
Suzuka       2.114
Tsukuba     -0.003 -0.081
Motegi East  0.595
Suzuka East  0.319
Laguna Seca -0.267  0.227

Savannah vs quattro

Deep Forest -0.12
Suzuka      -0.744
Tsukuba     -0.475 -0.413
Laguna Seca  0.904  0.112
Motegi East -0.425
Suzuka East  1.006 -0.723

quattro vs 2002

Deep Forest  0.637
Suzuka      -1.370
Laguna Seca -1.131 0.155
Tsukuba      0.478 0.494
Motegi East -0.17

Looks like rock-paper-scissor to me.

EDIT: Regarding the chance of getting a car:

Don't forget we have Shuffle and Parity races:

For shuffle the chance looks something like this:

View attachment 333295
The bump comes from the number of racers, it is mostly less than full grid. There are some specialties to consider here, as new racers entering late get #3 per default, so this number is special. In any shuffle the number of participants is special too, as it's more likely for any racer to get this number, if he makes a serious mistake and ends up last. But this number is essential random too, so we cannot take this into account too.

In Parity, it's mostly determined by racer funkyness, but the chance for any random driver looks like this:

View attachment 333300
It's more skewed for higher car numbers.

So basically we should look for an overall list fairness.

I agree with almost everything you say. One observation about the testing times though: the used tracks for this list show a tendency for slow, twisty tracks (except for Suzuka). Now, you can argue that this is the sort of track that these cars will mainly be raced on, but I'm not sure that is necessarily true. People don't always pick a car/track combination that a certain list might be best for. Quite on the contrary, they often like going against type and pick e.g. Low Power on Le Mans. So, the list ordering should actually reflect some middle ground after all and gaps opening up on a higher speed circuit are a noteworthy indicator of which car is actually going to produce better results easier (because speed on straights can usually be achieved by everyone, but corner speeds are where it's harder to find the limit). We're basically having a 5:1 bias towards the slower tracks at the moment, which worries me a bit.

I have also started to work on a classification system for tracks. Maybe we can derive some weighting or so out of that, if it should prove to work.

I like your remarks about chances of being assigned a certain car (and the chart thereof). As I said though, for parity racing the car ranking of problematic cars is more or less secondary as you're free to choose between 2 cars. That again makes the system very robust against possible driver/car allergies.