oystein eker
Norway Unspecified sola

This is one of the most important features in the game, and made me jump of the fence and hit the purchase button.
I am not good in math, so maybe I am far off. Be free to shake your head and smile. In addition, I have not tried the game, just read the rules.
I see that the speed does matter, and not counting spaces. But if two cars are at speed 8 and within reach of finish line. Both cars have accumulated 17 turns. My point is that one car is just one space to go before reaching finish line, and the second car have 7 spaces to go.
They will both end up with 18 turns, and able to squeeze in the fianl 8 scale speed to improve the turn/speed ratio, not counting the last distance difference at all. Maybe it is not significant? Or too complex to calculate within the complexity game design? (The fixed stage length maybe adjusted for design effect calculation?).
......
Comment on final calculation:
The maximum seconds that can be obtained is 5 seconds, right? 0.09 x 60.
If this right, maybe a better way to calculate is to convert every 60 to next higher as +1.
Example:
2.65.93.
Read as 2 hour 65 minutes and 93 seconds
Calculate: 93 seconds = 1 minute and 33 seconds
You now have 2 hours 66 min. and 33 sec.
Calculate minutes the same way = 1 hour and 6 min.
Final:
3 hours 06 minutes 33 sec.
.
(Been around the same circles in my Formula Next Generation, but gave up converting to time and real life speed. You are free to steal/salvage/pick inspiration from it for future modifications. Link http://boardgamegeek.com/boardgame/94511/formulanextgenera...)
Really looking forward to race the vintage Mille Miglia!!!
Edit:
Forgot to say this is not meant as critic of the game. I know it has been tested for years, and I have not even tried the game. I am just asking/comments of curiosity. No doubt I will love the game as it is!!!


Darrell Hanning
United States Jacksonville Florida
We will meet at the Hour of Scampering.

Oystein, glad you got the game. But you should be aware, there's an even larger issue with calculating time than the one you discovered.
As it stands right now, cars can finish a segment in a different total number of turns. For instance, car A might take 14 turns to finish, while car B takes 15 turns to finish. However, number of turns does not currently factor into the total time computation  only average speed.
So, how is it that cars can take a different number of turns, and (worse still) how is it that a car can take more turns to finish a segment, yet still log the faster average speed, and thus the better time?
The problem lies in the curves. If you look at the typical curve on these track tiles, you'll see that there is often two and sometimes three times as many spaces around the outside lane, as there are on the inside lane.
This in itself is not unusual in a racing game. If we take a look at most any other system, such as Formula De, Motorchamp, Motorsportspiel, etc., we see the same thing. The difference lies in determining who wins. In the other games, the competitors all leave the start line at the same time, and whichever car crosses the finish line first (or maybe even by the most) wins. But in Legend, we find ourselves computing who wins, because the cars do not all start at the same time.
So, in the case of Car A versus Car B, we see that Car A left first, took the shortest line through all the curves, and ended up with an average speed of, for example, 6.2 spaces per turn. Car B, on the other hand, took the longest line through all the curves, and while he traveled a greater distance as a result, the difference in distance is not a factor in the time computation, and so his higher average speed (because he could take advantage of the higher speed limits on the outer paths of the curves) comes out to 7.1, for example. Once we compute out who had the better time for the segment, Car B looks to have beat Car A by several minutes.
So, one way to look at the problem is that we have too much of a difference in course length between travelling along the insides of the curves, versus travelling along the outsides of the curves. Another way to look at it is that we do not have enough straightline distance in each race segment, to make the curves a smaller (and more proportionally correct) fraction of the total distance travelled. And yet another way to look at it is that  regardless of the composition of spaces (curve and straight) of any given segment  we shouldn't base the time to complete the segment on average speed (but perhaps instead on number of turns taken).
So, I'm interested in getting your input on the subject, once you've run a couple of segment races.


oystein eker
Norway Unspecified sola

Hi Darrell.
Nice to be in touch with you again!(Have some gold) Thank you for a detailed answer. Have not got the game yet. Ordered it yesterday. Just read the manual a couple of times.


Aaron Thorp
United States Rochester New York

Oystein, Darrell,
Agreed there are two problems with the time calculation as you have pointed out: using the "sum of speeds" as distance traveled yields some very strange results and "discarding" the "excess" distance traveled on the last turn can swing the time for a stage one way or the other, particularly noticeable on a flying arrival.
For what it's worth, using the "easy pack" method of time calculation (simply count 4 minutes per turn) yields acceptably accurate results if you are playing all 15 stages. I just ran 1927 with a Lambda Lancia and pretty much tied the leading Lambda's time from the actual race.
And the game has NOT been tested for 4 years... it was researched for 4 years. In my opinion the game really needs additional development. That said, if played within the spirit of the game (don't count spaces, make decisions quickly, yes you will make mistakes) it has some elements that are very fun and I think it really captures the "feel" of driving fast better than any other game I've played.
Treat it like a roleplaying session and immerse yourself in the setting. That's where the value of this game is.
Aaron


oystein eker
Norway Unspecified sola

Darrell  have studied your issue a bit. Must be a way to work around it. Maybe the designer will chime in and confirm that this not a significant issue worth the work.
Edit: Aaron have sme good point before I hit the submit button. Thanks.


Darrell Hanning
United States Jacksonville Florida
We will meet at the Hour of Scampering.

aaron_roch wrote: Oystein, Darrell,
Agreed there are two problems with the time calculation as you have pointed out: using the "sum of speeds" as distance traveled yields some very strange results and "discarding" the "excess" distance traveled on the last turn can swing the time for a stage one way or the other, particularly noticeable on a flying arrival.
For what it's worth, using the "easy pack" method of time calculation (simply count 4 minutes per turn) yields acceptably accurate results if you are playing all 15 stages. I just ran 1927 with a Lambda Lancia and pretty much tied the leading Lambda's time from the actual race.
And the game has NOT been tested for 4 years... it was researched for 4 years. In my opinion the game really needs additional development. That said, if played within the spirit of the game (don't count spaces, make decisions quickly, yes you will make mistakes) it has some elements that are very fun and I think it really captures the "feel" of driving fast better than any other game I've played.
Treat it like a roleplaying session and immerse yourself in the setting. That's where the value of this game is.
Aaron
Four minutes per turn sounds about right to me  I was simply waiting to see what the designer had in mind, since he had previously stated he was working on an alternative method for determining time.
That still leaves a lot of cars finishing with the same time. Obviously, on those segments where you have a "flying finish", you could measure the difference in remaining movement points, and express the adjustment as (n/12)*4, where "12" is the highest possible speed, n is the speed on the last turn for the car, and 4 is the number of minutes represented by the last move. So, if for instance you finished the race in 14 moves, and you were travelling at a speed of 6 on the last turn, then the total time for the segment would be 54 minutes (13*4, plus ((6/12)*4)).
This, however, does not address those segments where you must cross the finish line at a speed of 3 or less. But I suspect that, on those segments, there is a greater deviation in number of turns taken to complete. (There's usually some jockeying around needed, to get to just short of the finish line the fastest, while leaving adequate braking distance, to get down to the maximum crossing speed.)


oystein eker
Norway Unspecified sola

Throwing in some ideas while waiting for the game:
1. Making the shorter track through a curve**:
Add the highest* circled number to the speed (for a better turn/speed ratio) when adding up during time calculation. Example: Speed 4 and highest* circled number 5.
Your chart write up looks like this:
4+5
Your speed will be 4. The final adding calculation will be 9. In this way you can lower number of turns and simultaneous increase scale speed.
Maybe your average speed will be too high with this?
You can counter this with by using the final turn number of spaces mentioned in initial post. Just subtract them from total number of speed.
Example: Last turn: Your speed is 8. With 7 spaces to go to enter the finish line. Subtract 7 from your total speed.
Example: Last turn: Your speed is 8. With 2 spaces to go to enter the finish line. Subtract 2 from your total speed.
Is worth trying variants of this  or is it a blind alley?
*edit
**Edit clarification: This is to address Darrell issue mentioned above with car A and B. The shorter track means the ideal line of maybe 3 spaces around the curve. You do not have to hit the circled 5 spaces. In short you travel at speed 4 through the curve of 3 spaces. You gain 9 speed points with this house rule.




Hello Everybody!
First of all let me thank you for pointing out some issues with the game: for us the continuous improvement is the key to the success.
I'd just wish to know you that thank to your notes and the unvaluable support of David Briel and Aaron Thorp we are working on some "bug fixes" about the race model.
In my personal opinion the biggest bug is the one about ideal line vs. external line race time: fixing that we could work out all the other in cascade. However, as said, this is my personal opinion and I want to share and work on this issues with other "free of mind" gamers.
My plan is to assess a basic fix I'm going to share with you in the next days so you can tell me your opinion about and/or test it yourself in order to share a feedback and develop it to a final bug fix to add in the upcoming new rules booklet.
P.S. Oystein, you final calc comment sounds pretty appropriate...I'll use it on rule booklet reprint and by the way, thank you for sharing your own work with us! It will surely be a source of future inspiration.
WRITE YOU ALL SOON!


Darrell Hanning
United States Jacksonville Florida
We will meet at the Hour of Scampering.

eker wrote: Throwing in some ideas while waiting for the game:
1. Making the shorter track through a curve**:
Add the highest* circled number to the speed (for a better turn/speed ratio) when adding up during time calculation. Example: Speed 4 and highest* circled number 5.
Your chart write up looks like this:
4+5
Your speed will be 4. The final adding calculation will be 9. In this way you can lower number of turns and simultaneous increase scale speed.
Maybe your average speed will be too high with this?
*edit
**Edit clarification: This is to address Darrell issue mentioned above with car A and B. The shorter track means the ideal line of maybe 3 spaces around the curve. You do not have to hit the circled 5 spaces. In short you travel at speed 4 through the curve of 3 spaces. You gain 9 speed points with this house rule.
So, if I understand you correctly:
If the maximum speed in a curve is "5", and one car (taking the shorter route) has a speed of "3" for the turn, he is actually logging "8" for his speed. But what is the other car (taking the longer route through the curve), at a speed of "5", recording for his speed for the turn?
I must be misunderstanding this, because that doesn't seem to help with the problem.


oystein eker
Norway Unspecified sola

Yes Darrell
My intention is to improve the turn/speed ratio for car A by adding more speed point, while keeping or improve chance to lower #turns. But of course this may be wrong...
If the main point is: If car B intentionally taking advance of a weakness in rules all the time and play against the spirit of the game, we can take a different approach to this.
No lack of ideas!!
Boy... I am looking forward to paint minis and race through Italy in an unreliable car!!


oystein eker
Norway Unspecified sola

Darrell
Can you give an approx estimate of gain of speedpoints by car B in an average curve vs car A?


Darrell Hanning
United States Jacksonville Florida
We will meet at the Hour of Scampering.

From this picture, you can see that the inside of this curve is 4 spaces (at a max speed of 4), and the outside is 6 spaces (at a max speed of 6). This is typical for the "midrange" curve  some are smaller (and slower) and some are larger (and faster).


oystein eker
Norway Unspecified sola

Going outside all the way means 6 speed points, and in addition be able to accelerate (2) up to speed 8. Downside is +2 spaces, and may end up with additional turn.
Going inside with a natural apex touch, you gain only 4 speed points and accel up to 6. Upside is 2 spaces with potential turn loss.
Going outside in all spaces due to unintentional high speed or just playing against the spirit of the game should be penalized, and encourage players to use a better raceline.
Right?
If I am correct, it should be an easy fix to this.




Oysten, this is my work direction (Aaron and David are working toward a different direction). My starting point is how I did design the game:
1 simulated speed = 18 km/h or 0,3km/min or 1,2 km/turn [considering 4 minutes per turn]. This means that each space on tiles approximately is 1,2 km long. If you divide the stage length in km for 1,2 km, you'll find the result approximates the number of spaces of that track.
Assuming this, if you ride the entire stage on the middle lane you'll achieve a "standard" racing time for that stage and the given car. If you use the ideal line, you'll be faster because you ride a shorter course AND achieve the +1 bonus on driving speed. The problem is on the outmost lane: I was thinking to assess a malus of 1 speed per space (or...each space counts double).
Take for example the picture above: my car is on the first space of the ideal line space in the middle lane, speed 6.
Case 1  Middle line: I slow down to 5 and move of 5 spaces
Case 2  Ideal Line: I stay at 6 and move of 6 spaces (don't care of damage on 4speed spaces now)
Case 3  External line: I stay at 6 and move of just 3 spaces on the outside line. Final speed = 3.
This is my idea and it must be tested (I see already some issues with this...probably the ideal line requires a bonus enhancement)
Aaron's system drops completely the sum of speeds and work on turns base only with fractional turns at the end of the stage: I wait Aaron to post the system proposal himself.
[small edit: about the ideal line, we could assume that the first ideal line space speed dictates the speed for the entire ideal line...In the case above this speed is 5 and you can assume the entire ideal line has a speed of 5 (instead the 4s on two spaces that actually reduce the speed if you don't want to make damage)...what do you think?]


Aaron Thorp
United States Rochester New York

JRC_Firewalker wrote: [small edit: about the ideal line, we could assume that the first ideal line space speed dictates the speed for the entire ideal line...In the case above this speed is 5 and you can assume the entire ideal line has a speed of 5 (instead the 4s on two spaces that actually reduce the speed if you don't want to make damage)...what do you think?]
Carlo, I like this better than the outside lane malus, as it is simpler. I'll post my other thoughts on time calculation later today.


oystein eker
Norway Unspecified sola

My guess is that the effect lies in the coarse granulation of the core system. Not necessary a design flaw. With this coarse granulation the game runs fast and easy. Inflating the game to finer granulation means longer playing time and bigger board (Inflating both speed and number of spaces).
What we need is a simple correction to this effect. I tried to calculate (I do have poor math skills) the benefit of using all outer spaces, and ends with approx 30%, based on picture. Not counting accel gain and possible turn loss.
My personal opinion. I am not a designer.
To make it simple and dirty when my game arrives, I will try this house rule:
"When using all outer spaces, fill in a speed box, subtracting 1/2 of one circled number."
Example (circled 6)  You fill in next speed box with 3.
To neutralize the minor space difference. Again an effect of coarse granulation. I will try this house rule:
"After final game turn (passing finish line), subtract spaces moved from current speed. Add this number to total speed."
Example: Speed 8. Last number of spaces moved 2. Add 6 to total speed. Speed 8. Last number of spaces moved 7. Add 1 to total speed.
I am not sure if this is correct, and work as intented. Please correct me.


Darrell Hanning
United States Jacksonville Florida
We will meet at the Hour of Scampering.

We can do better than this, and in fact be exact, if you're willing to do the math:
((AD/TM)*(TS/NT) * 18)/SD = Total time for Stage, where:
AD is Average Distance (distance calculated from counting the number of spaces down the middle lane of the stage)
TM is Total Movement (total speed minus those spaces remaining to be moved, if ending with a flying finish)
TS is Total Speed (total of all speeds recorded during the race)
NT is Number of Turns taken to complete the Stage
SD is the Stage Distance in kilometers
So, in the case of the car that sticks to the outer lane, running a higher average speed per turn, his adjusted average speed for the stage will be something less than one times his average speed per turn. If, for example, he travelled 100 spaces in a sector where the average distance is 80 spaces, then his adjusted average speed will be eighty percent of what he would record using the current system found in the rules.
In the case of a car that sticks to the inner lane, running a lower average speed per turn, his adjusted average speed for the stage will be something higher than one times his average speed per turn. If, for example, he travels 60 spaces where the average distance for the stage is 80 spaces, then his adjusted average speed will be one hundred and thirtythree percent of what he would record using the current system.
Mind you, these figures are for illustration only, and probably exaggerate the gains and losses made from actual usage of this formula.
I have some logs at home  I might calculate some of these out, to see how it changes the results, but I don't see how this formula could prove wrong. (But I'm open to someone pointing out errors!)


oystein eker
Norway Unspecified sola

Darrell.
In short  all you have to do is counting spaces prior to play, and then you use your normal logged data in the formula?
If you confirm, I will certainly try it!!!


Darrell Hanning
United States Jacksonville Florida
We will meet at the Hour of Scampering.

That's it. We need the space count for the middle lane of a Stage, before doing the calculation.


David Briel
United States Indianapolis Indiana
Grak !

The "Spaces" tab on this spreadsheet I made has the number of spaces per tile for ideal and longest path in the direction of the arrow. On very rare occasion there is a +/ 1 space difference running in the opposite direction. The "Stages" tab has more details on the actual stages based on those numbers.
https://docs.google.com/spreadsheet/ccc?key=0AmdkCoMo_OKDdG1...


Aaron Thorp
United States Rochester New York

My goals for revising the time calculation are two:
(1) Player who finishes in fewest turns wins, and (2) Players can compare times to historical times. Specially, this means that a good player who pushes their car and gets a little lucky should be able to equal the historical winning time using the same car.
In my view, the simplest system that accomplishes the above should be the preferred system.
The game is already scaled around 4 mins/turn, and simply multiplying game turns by 4 mins gives a good result that meets both goals as well any any method, and i'm sure that's why Carlo used it for the easy pack. In my view, the only problem with this simple method is that it yields too many ties and does not "feel" accurate.
But both of those things can be overcome while maintaining the very simple system. At the end of a stage, one often has "extra" movement left over i.e. a player will choose speed 8 but move only 2 spaces to cross the finish. This "extra" movement can be handled several ways I like Carlo's suggestion of giving the player a "bonus" on the beginning of the following stage. This "bonus" movement is also the tiebreaker for the stage & determines move order for the next stage.
SO on a Flying Arrival, our hero chooses speed 8, moves 2 to cross the finish and this completes his turn. At the beginning of the next stage, he gets 6 free movement he moves his car 6 spaces and does NOT count this movement as a turn. It is the 'remainder' of the movement from his last turn.
For a "Control Point", where we must slow to speed=3 and then do a standing start the next stage, we can simply discard the extra spaces saying that 1 or 2 extra spaces are irrelevant OR take the bonus spaces on the next stage and perform a standing start 1 or 2 spaces into the stage. I'm not sure which I prefer.
For the FINAL STAGE (Feltre  Brescia OR whatever the last stage is of the scenario being played): we calculate how MUCH of a turn the player used to get to the finish and adjust time accordingly. So if our hero finishes a scenario using 2 movement on a speed 8 turn, he has used 2/8 or 0.25 turns to finish. So we would take 0.25 x 4 minutes, and our hero's final turn counts as 1 minute.
We would end up with a time calculation very similar to that from the "easy pack" but with a few additions to overcome the minor problems of that system.
Aaron


Aaron Thorp
United States Rochester New York

DarrellKH wrote: We can do better than this, and in fact be exact, if you're willing to do the math:
((AD/TM)*(TS/NT) * 18)/SD = Total time for Stage, where:
AD is Average Distance (distance calculated from counting the number of spaces down the middle lane of the stage) TM is Total Movement (total speed minus those spaces remaining to be moved, if ending with a flying finish) TS is Total Speed (total of all speeds recorded during the race) NT is Number of Turns taken to complete the Stage SD is the Stage Distance in kilometers
Darrell, this is very elegant! I think you meant SD/((AD/TM)*(TS/NT) * 18) You can get very close to the AD by applying the game scale: 1.2 spaces per km.
Here's a sample comparison based on data from an earlier playtest. Feltre  Brescia 239km, Average Distance = 199 Sally takes the ideal line and pushes her car. 38 Turns, sum of speeds = 173, TM = 172 Giuseppe takes the outside line at max "safe" speed. 52 turns, sum of speeds = 253, TM = 252
Current Rules: Sally 2:54:59 Giuseppe 2:43:44 Easy Pack: Sally 2:32:00 Giuseppe 3:28:00 My "mod" to Easy Pack: Sally 2:30:40 Giuseppe 3:26:40 Darrell Method: Sally 2:31:15 Giuseppe 3:27:21
My tentative conclusion is that this calculation is accurate and elegant. I also came up with an alternate calculation that I thought was accurate (although not as elegant), but I think the 'easy pack' method of 4 minutes per turn is still best because it also meets all goals of time calculation AND is really simple to calculate, which in my eyes is a big benefit.
After all, we are measuring time, and the game already has a scale for time. Whether a player travels fast over more spaces or slowly over fewer spaces should be irrelevant whoever gets there faster (i.e. fewer turns) wins.
Aaron


oystein eker
Norway Unspecified sola

Gentlemen.
I am speachless. Just a silent wow over my lips. Crawling on my knees up to your feet,kisses them, and leave a humble sum of g gold.




Very good job gentlemen!
I'd say we could use Darrell's system as the new Time Calculation, and Aaron's as the Easy Pack one.
Do you all agree? Would you like to join me in testing the 2 systems so we can produce a rules' update?
I plan to make a new Time Sheet for the basic game to lead players in Darrell's calculations and a new Easy Pack rule set. The main game rules will be updated in the new booklet we'll print later this year.
No need to say you'll be all mentioned and credited for the update...


Aaron Thorp
United States Rochester New York

Carlo, I would absolutely like to test the two systems for a couple of different stages (and combinations of stages) using different cars & years.
Maybe you, Darrell, David have some games logged with enough detail to simply run the calculations?
Aaron



