Concept, feature amd progress list

[zaqrack]

CONCEPT AND FEATURE LIST
Last update: July 5th

TRACKS
-3 races running simultaneously + permanent tracks
-If no one sends replay a day it's considered an inactive day.
-After 10 days of continous inactivity the track is scheduled for closure
-Inactivity could may be 7 days. We'll see how the system works.
-Track closing time: 5 QUIET days (after the inactive days)
-During quiet days you can see who sent replays but can't see the times.
-If a track is running for 90 days it gets permanent.

CARS
-All original Stunts cars (11) + 5 tuned/cheated ones will be available.
-Active tracks can not contain the same car twice, all cars must be different
-A track can't have a car assigned, which was used on any of the past 6 tracks.

TRACK UPLOAD
-Anyone can upload tracks
-Optionally you can upload a track picture. If you don't, the admin can do it later.
-At upload the track gets into the waiting list.
-At upload you can assign 1-3 cars for the track (with priorities)
-Uploader can select from an array of pre-set special rules (OWOOT,Specials from 2007,etc.)
-Tracks appear as active in the waiting list order.
-If all 3 assigned cars cannot be used for a while, the 2nd track from waiting list gets active.
-Only one track with special rule can run at the same time.
-Track uploader gets 5% time penalty on his own track IF he is in the top3.

REPLAYS
-method of uploading will be pretty much the same as until now
-replays are stored on the server and are available for download after track was closed.b
-most likely I'll be able to read transmission type and car color from the replay
-this also applies on automatic track validation. It's not my priority work.
-you can see a list of your uploaded replays
-you can set inactive your uploaded replays (useful when sent a wrong rpl)

STATS,EYECANDY,USEFUL STUFF
-When a track is scheduled for closure, every pipsqueak who sent a rpl on that track, gets a notification in email.
-The trend change of the leading time will be presented on a live graph
-All sent times are stored and browseable.
-Track stats
(track runtimes, no. of replays sent, popularity (=runtime*replays sent)
-Sent replay stats
(by pipsqueak, by track, by time)
-pipsqueak activity stats
(replays/track, replays/month, tracks uploaded)

RANK LIST
-Result is calculated from the past 6 races. Actives not included.
-Basic calculation: first: 10pt, 2nd:9pt, 3rd 8pt, linear, top10 gets points.
-Winning on a longer lasting track is worth more, determined by a multiplier
runtime multiplier
10 days 0.8
20 days 0.85
30 days 0.9
40 days 0.95
50 days 1
60 days 1.05
70 days 1.1
80 days 1.15
90 days 1.2
-pipsqueak with the most leading time gets a multiplier bonus for that track (1.2)
-pipsqueak with the highest activity gets a multipler bonus for that track (1.1)
-trend change of ranklist standings also available on live graph

SITE DESIGN
-optimized for 1024x768, no other resolutions available
-CSS based, maybe 2 selectable stylesheets (depends on time)
-2001-2007 archives of course available
-simplified menu system
-most important information (active tracks, runtimes, latest replays) directly on mainpage
-shoutbox stays

[Chulk]

WOW! Long (and very attractive I must say) list of features! I can't wait until October!

[Krys TOFF]

During quiet days you can see who sent replays but can't see the times.

Good way to put on pressure with sending a slower replay than current one, just to make other pipsqueaks crazy.

All original Stunts cars (11) + 5 tuned/cheated ones will be available.

Good to see that there will be places for other tuned cars than Melange, like the new cars made by Duplode and Mark. Or Skoda Kompressor, or Coconut car Coronet that I like too, or some of Zapper's cars that are quite good too.

A track can't have a car assigned, which was used on any of the past 6 tracks.

This may be difficult to manage is some tracks stay 90 days long as "active". I suppose that "permanent" tracks are excluded from this limitation.

Uploader can select from an array of pre-set special rules (OWOOT,Specials from 2007,etc.)

I also hope it will be possible to add new rule if needed, just by sending a mail to you Zak.

Only one track with special rule can run at the same time.

Nice idea too to let fans of pure "free Stunts racing" not be "excluded" like they may feel since the beginning of special rules at ZakStunts in January 2007.

Track uploader gets 5% time penalty on his own track IF he is in the top3.

Good idea too.

most likely I'll be able to read transmission type and car color from the replay.

If you develop a system like RPLInfo for tuned cars I'm sure all other Stunts competition managers would like to have it.

you can set inactive your uploaded replays (useful when sent a wrong rpl)

CTG will love this feature !

Basic calculation: first: 10pt, 2nd:9pt, 3rd 8pt, linear, top10 gets points.

You don't keep the idea of fractionnal points to classify pipsqueaks after the 10th place (0.01 for the last and growing by 0.01 up to the 11th place) ?

pipsqueak with the highest activity gets a multipler bonus for that track (1.1).

Nice idea for the runtime bonus for victories, but this bonus will always be for AbuRaf.

CSS based, maybe 2 selectable stylesheets (depends on time)

So maybe it will be possible to let other members create a css file for new website design and let it available for the others too ? I would like to create a Krys-mas css style for example.

[CTG]

Too difficult system. If it depends on me, every tracks will be permanent...

[DieselJoe]

Sounds like a great thing coming up!

Maybe the thing with additional special rules could also be automated, so that pipsqueaks would not have to send their suggestions to Zak. Should not be too difficult if I imagine it right, we simply needed a possibility to enter a description and maybe to upload further explanatory pictures.

[Argammon]

Hi pipsqueaks!
Please allow an oldbee to make some comments about the new concept  .
I wrote my ideas below Zak's ideas in italics.




TRACKS
-3 races running simultaneously + permanent tracks
Interesting idea. I just hope it won't happen that all 3 tracks end on the same week

-If no one sends replay a day it's considered an inactive day.
-After 10 days of continous inactivity the track is scheduled for closure
-Inactivity could may be 7 days. We'll see how the system works.
I would suggest to modify this rule slighty. I would suggest that the track is scheduled for closure if no new leading time replay has been send during 10 days. Imagine one pipsqueak is leading the scoreboard with a very good time, say 1 minute and 10 seconds. Then there is a small group of pipsqueaks between 1 minute and 18 seconds and 1 minute and 24 seconds. This small group could send many replays which only improve upon their last replay by 0.05 seconds to buy time. Maybe if they keep the track open for few more month they will find the trick of the top pipsqueak. I am not sure if we want this?

-Track closing time: 5 QUIET days (after the inactive days)
Yeah why not

-During quiet days you can see who sent replays but can't see the times.
Good idea! Could create more suspense

-If a track is running for 90 days it gets permanent.
Another option would be to seperate permanent and seasonal competition. For example, the tracks end definatly after 90 days and the pipsqueaks are awarded their points. Then pipsqueaks are view to improve in a seperate permanent competition... But not sure about this one

-All original Stunts cars (11) + 5 tuned/cheated ones will be available.
-Active tracks can not contain the same car twice, all cars must be different
-A track can't have a car assigned, which was used on any of the past 6 tracks.
First of all, I think all original cars should be used within a year. Then most of the cheat cars with the exception of the melange are really unpopular and unrealistic. Do they have to be used?

TRACK UPLOAD
-Anyone can upload tracks
-Optionally you can upload a track picture. If you don't, the admin can do it later.
-At upload the track gets into the waiting list.
-At upload you can assign 1-3 cars for the track (with priorities)
Will cars like the Lm002 ever be used in this case?

-Uploader can select from an array of pre-set special rules (OWOOT,Specials from 2007,etc.)
Special rules now and then are ok. But I really don't like most of them

-Tracks appear as active in the waiting list order.
-If all 3 assigned cars cannot be used for a while, the 2nd track from waiting list gets active.
-Only one track with special rule can run at the same time.
-Track uploader gets 5% time penalty on his own track IF he is in the top3.
Of course that should mean if he is in the top 3 he has to be at least 4th place with time penalty. It would be funny if the penalty would make him drop from 2nd to 5th place for example 

RANK LIST
-Result is calculated from the past 6 races. Actives not included.
-Basic calculation: first: 10pt, 2nd:9pt, 3rd 8pt, linear, top10 gets points.
-Winning on a longer lasting track is worth more, determined by a multiplier
runtime multiplier
10 days 0.8
20 days 0.85
30 days 0.9
40 days 0.95
50 days 1
60 days 1.05
70 days 1.1
80 days 1.15
90 days 1.2
Again, that allows some strategic options. For example if we race with the countach and I am sure none of the other pipsqueaks can touch me, I could artificially prolong the duration of the track to gain more points by sending 0.05 replays

-pipsqueak with the most leading time gets a multiplier bonus for that track (1.2)
-pipsqueak with the highest activity gets a multipler bonus for that track (1.1)
I am against both ideas and especially the second one. One reason is that the second rule can easily be exploited again by sending lot's of 0.05 replays. Second, many of us have a job now and thus never the chance to win those points. Also, chance is that those points go the leading time master anyways.
For sure, the multipliers are to strong.

-trend change of ranklist standings also available on live graph
I would also suggest that a winner should be announced after X tracks have been raced. That should be crucial to motivate pipsqueaks.


Ok, that's it for now. All in all an interesting new concept. But I think it has some flaws here and there.

[Krys TOFF]

most of the cheat cars with the exception of the melange are really unpopular and unrealistic. Do they have to be used ?

I disagree. Zapper's cars are good, new car by Mark and Duplode need more tests before I give a strong opinion but they seem promising. Coconut car Coronet handling is good too.
And regarding "realistic" (is powergear realistic ?), I think only a few cars are like that (those with powergear obtained immediately at 1st gear). Others, you may or may not like the handling of the car, but I think it's difficult in a game like Stunts to talk about "realistic" driving anyway.

Will cars like the Lm002 ever be used in this case?

Sure, Zak and I at least will design tracks for this car.

-Track uploader gets 5% time penalty on his own track IF he is in the top3.
Of course that should mean if he is in the top 3 he has to be at least 4th place with time penalty. It would be funny if the penalty would make him drop from 2nd to 5th place for example 

Argy has a point there : this % added should not make the concerned driver lower than 4th place IMO.
If 5% penalty keep him in top 3, then good for the driver. But if 5% penalty added put him out of podium, then he should not be lower than 4th.

Imagine tracks with very close times, 5% may put him out of top 10 !
Example : imagine that CTG designed current Pretoria track for USL event. All times at this moment are between 57"95 and 59"00, 3rd is CTG with 58"25. 5% addition would make his time slower than 1 minute, that is last place. Not fair...

For example if we race with the countach and I am sure none of the other pipsqueaks can touch me, I could artificially prolong the duration of the track to gain more points by sending 0.05 replays

Just for my own curiosity, can you test the track attached with this post ? Race it with countach during 1 week maximum without watching the replays made by during competition, and then publish your time and replay in the forum. I'm curious to see your best time and replay at this one.

-pipsqueak with the most leading time gets a multiplier bonus for that track (1.2)
-pipsqueak with the highest activity gets a multipler bonus for that track (1.1)
I am against both ideas and especially the second one. One reason is that the second rule can easily be exploited again by sending lot's of 0.05 replays. Second, many of us have a job now and thus never the chance to win those points. Also, chance is that those points go the leading time master anyways.
For sure, the multipliers are to strong.

That's right about 2nd multiplier : it will always be spamming CTG or crazy replay sender AbuRaf that will get this one, so maybe it should be decreased as Argy suggests.

[Argammon]

Ok some comments on krys comments are in order

I disagree. Zapper's cars are good, new car by Mark and Duplode need more tests before I give a strong opinion but they seem promising. Coconut car Coronet handling is good too.
And regarding "realistic" (is powergear realistic ?), I think only a few cars are like that (those with powergear obtained immediately at 1st gear). Others, you may or may not like the handling of the car, but I think it's difficult in a game like Stunts to talk about "realistic" driving anyway.

I agree that whether a car is good or not is mainly a matter of taste. But about realistics: If you compare, for example, the lada niva comback with the Porsche Carrera I would claim that the Carrera is at least somehow driving like a car while the niva steers like a space ship from a cheap arcade game.

Sure, Zak and I at least will design tracks for this car.

I thought it was like you always have to propose 3 cars as a track designer. Now, if you propose Lm002,GTO and Indy or Acura,Lm002, Carrera or 962,Lm002,Lancia or... or... the chance might be very slim that the lm002 gets the majority of the votes

Just for my own curiosity, can you test the track attached with this post ? Race it with countach during 1 week maximum without watching the replays made by during competition, and then publish your time and replay in the forum. I'm curious to see your best time and replay at this one.

Please allow me to mention that I didn't come up with that example to brag. I just tend to use examples to help others understand my arguments. I could as well have made the same example with Bonzai Joe and the Indy or whatever 
Furthermore, I am quite busy in real life and a 1 week contest doesn't really fit into my schedule right now. I hope you aren't angry if I abstain from that offer. Why did you make it anyways?

[Chulk]

-Track uploader gets 5% time penalty on his own track IF he is in the top3.
Of course that should mean if he is in the top 3 he has to be at least 4th place with time penalty. It would be funny if the penalty would make him drop from 2nd to 5th place for example 

Argy has a point there : this % added should not make the concerned driver lower than 4th place IMO.
If 5% penalty keep him in top 3, then good for the driver. But if 5% penalty added put him out of podium, then he should not be lower than 4th.

Imagine tracks with very close times, 5% may put him out of top 10 !
Example : imagine that CTG designed current Pretoria track for USL event. All times at this moment are between 57"95 and 59"00, 3rd is CTG with 58"25. 5% addition would make his time slower than 1 minute, that is last place. Not fair...

I like this rule! If you uploaded the track, then you have to play a good strategy and be 5% faster than 4th or race for the 4th place. It's your choice. Why should you have the chance to be on the podium if you're fast enough but not being 5 if you beat 5th pipsqueak by a small gap?
Imagina you upload the track and race 1:40.00 (100 secs.) and 5th or 6th did 1:40.50. You had a handicap but couldn't get an advantage out of it. 6th driver deserves to overtake you!

[Duplode]

On the issues Krys has not mentioned in Argammon's post, I particularly agree with declaring quiet final days after 10 days of no leader change (instead of no replay sent), to close "seasonal" competition for all tracks in 90 days (if the longer-lasting tracks are the best ones, they should have a winner declared! Also, 90 days might be a bit long), and to declare a "season" winner after X (21 maybe?) races. Adittionaly, be careful with the multipliers, they can be a majour source of imbalance. And overall, the planning seems great Zak, very promising!

[Krys TOFF]

I disagree. Zapper's cars are good, new car by Mark and Duplode need more tests before I give a strong opinion but they seem promising. Coconut car Coronet handling is good too.
And regarding "realistic" (is powergear realistic ?), I think only a few cars are like that (those with powergear obtained immediately at 1st gear). Others, you may or may not like the handling of the car, but I think it's difficult in a game like Stunts to talk about "realistic" driving anyway.

I agree that whether a car is good or not is mainly a matter of taste. But about realistics: If you compare, for example, the lada niva comback with the Porsche Carrera I would claim that the Carrera is at least somehow driving like a car while the niva steers like a space ship from a cheap arcade game.

Sure, Lada Niva comeback is one of those useless cars that get powergear from the start. No interest, I agree.

Sure, Zak and I at least will design tracks for this car.

I thought it was like you always have to propose 3 cars as a track designer. Now, if you propose Lm002,GTO and Indy or Acura,Lm002, Carrera or 962,Lm002,Lancia or... or... the chance might be very slim that the lm002 gets the majority of the votes

I considered that you can propose up to 3 cars, but only 1 if you prefer. And Zak precised that all 11 original cars will have to be used per season. Of course, there will be more indy races than LM002 races...

Just for my own curiosity, can you test the track attached with this post ? Race it with countach during 1 week maximum
without watching the replays made by during competition, and then publish your time and replay in the forum. I'm curious to see your best time and replay at this one.

Please allow me to mention that I didn't come up with that example to brag. I just tend to use examples to help others understand my arguments. I could as well have made the same example with Bonzai Joe and the Indy or whatever 
Furthermore, I am quite busy in real life and a 1 week contest doesn't really fit into my schedule right now. I hope you aren't angry if I abstain from that offer. Why did you make it anyways?

This track was used for USL special event last week : one week of race, and potential 53" theorical best time according to shortest times of each "sector".
You're not forced to race of course, but if you do and reach 53 seconds on this track (without watching replays already made to race like in a competition), then I'll agree to call you Countach master.

[zaqrack]

thank you for the high interest and activity!

I wanted to make this post before I go on holidays, so I can start coding with better ideas tuned by your comments.

my comments on yours:

-I like the idea of uploading own special rules. Shouldn't be hard to manage, I'll most likely do it.

-Fraction points: I don't think we'll need them. The number of pipsqueaks dropped significantly, and I know there will be always at least one track from the three active which will get lower attention. This is where slower drivers could earn points and train themselves!

-of course permanent tracks are excluded from the past 6. With permament tracks, the track is closed on the 90th day, and results are transported to the ranklist. Just because of the popularity it'll be available for driving and time comparison forever.

-I don't like the idea of closing the track after 10 days of change in leading time. Just check the LTB tables any year. Many times there are long-long gaps, while there is great activity between the mid-range drivers. I want to keep that.

-activity bonus: sorry I didn't explain it carefully. It's not the one sending the most replays, but someone sending many replays and improving a lot. Still it's very vulnerable (and hard to code). I'm considering not to implement this bonus.

-perhaps if the winner is sending another leading replay it shouldn't count as a track prolonging rpl. This would solve the problem Argammon mentioned at the multipliers.

-cars for tracks: we wont vote! Trak designer sets the 3 cars in a priority order, and the first one not used in the past 6 months will be used. You dont like the cars coming? Easy, send your own track and car!

-and CTG also has a point. We shouldn't overcomplicate the system. If something is too difficult to manage/code/understand we should think of some different way to solve.

[Argammon]

Some more inputs from me. Great work so far btw.

-I don't like the idea of closing the track after 10 days of change in leading time. Just check the LTB tables any year. Many times there are long-long gaps, while there is great activity between the mid-range drivers. I want to keep that.

I understand your opinion. Nonetheless it is easy to exploit. People might artificially prolong the race either to get some kind of advantage or just for crazyness or to annoy other people.
Imagine I would send 0.05 improvements on all 3 tracks every 5 days. ---> All tracks would always become permanent. Of course I won't do that but I guess you got my point.
I don't see a solution for this problem other than making the rules more complicated or to bite into the sour apple and use my idea.
A complicated idea could work like that:
Track is closed if: 1) No pipsqueak improved at least ? % during the last 10 games
                 or     2) No time ? % below the current leading time has been sent in
I'm not really happy with it but exploits shouldn't be that easy.

-cars for tracks: we wont vote! Trak designer sets the 3 cars in a priority order, and the first one not used in the past 6 months will be used. You dont like the cars coming? Easy, send your own track and car!

Then we can just propose 1 car as a track designer. I guess most of us can anticipate those coherences and work out which of the 3 cars would be used anyways 

-and CTG also has a point. We shouldn't overcomplicate the system. If something is too difficult to manage/code/understand we should think of some different way to solve.

I think the most important point by far would be to prevent exploits. Those can really ruin the fun for most pipsqueaks.
Another important point is that an overall winner should be found somehow. Whether it's after X races, years or somehow determined by a complicated dynamic formula? I don't know. But please think this over.

Keep up the good work

[Argammon]

I forgot to answer Krys.

First of all your track has many strait elements and is "to fast" in order to gain much time on your opponents. Furthermore, there aren't enough complicated elements so you can keep a high speed over the majority of the track.

The theoretical best time does not say much as you always have the situations like this:

Say the most efficient way to drive a track gives you this checkpoint times if you drive perfectly well:

A: 10 seconds
B: 10 seconds
C: 10 seconds
= 30 seconds

Now it might be possible to drive a longer way in A to gain more speed or something along that line. It helps you gain a better B section time but is senseless overall:

A: 20 seconds
B:   6 seconds
C:  10 seconds
= 36 seconds

---> best theoretical time according your math: 10+6+10 = 26
---> best theoretical time in reality                : 10+10+10= 30

Sorry maybe this was to much rambling . Just saying that maybe a 53 time on that track is impossible...

[Krys TOFF]

activity bonus: sorry I didn't explain it carefully. It's not the one sending the most replays, but someone sending many replays and improving a lot. Still it's very vulnerable (and hard to code). I'm considering not to implement this bonus.

The pipsqueak improving a lot ? So, send a 9"99"95 replay as 1st replay and you'll get it...
To be honest, I think it would be better not to implement this bonus.

Another important point is that an overall winner should be found somehow. Whether it's after X races, years or somehow determined by a complicated dynamic formula? I don't know. But please think this over.

I agree.

/Off-topic ON

I forgot to answer Krys.

First of all your track has many strait elements and is "to fast" in order to gain much time on your opponents. Furthermore, there aren't enough complicated elements so you can keep a high speed over the majority of the track.

The theoretical best time does not say much (...)

That's not my track, it was designed by CTG.
And you're right about theorical winning time, usually it's often/always impossible to reach.
Winning time for this track during competition was 56"xx. Now, knowing the replays and continuing the one with the best start, I reached 54"xx. As I'm not the best Countach pipsqueak and as this finish is still not perfect, normally 53"xx is possible. Do it from scratch in a week (don't watch the replays), and I'll call you Countach master.
/Off-topic OFF