Obtaining Better B-Spec Stats ??

[SportWagon]

That 99.9% win ratio sort of spoils the look of the Status Page a little.

But I do like the way you achieved exactly 100,000 A-spec points. I finally got past 10,000 A-spec points. I had half-a-mind to take a picture with both scores 10,000, but unfortunately when I was somewhere above 9800 A-spec I entered the Shelby Manufacturer's Races with a Series 1; difficult not to get lots of A-spec points that way, and so my first score above 10,000 was well above it.

 

[gt4falcon]

I am at 9985/100/99/99 I must have missed a track some where along the way. I got board trying to found it so I have been doing some a-spec racing. But with Super Speedway find I am going to try that track and see what happens when I get home after work. I’ll let you know what happens.

Tried the Super Speedway with the 4 classes of cars and there modifications and I got 1 over all point with the Lupo. I am now at 9986/100/99/99.

I strongly suspect you are missing a pocket of floating battle points. When your skills changed to 98, did they change together? Could you estimate the difference based on the changes to 97, 96 and 95? If this difference is only 1 or 2 points, then you probably missed an entire track - 13 points of even course and battle skill.

I did keep some records, they are not the best but I did keep some. I did not keep the results when doing the 4 classes and their modification doing Nurburgring. Once I got to 100 machine points I went on to do the 58 tracks with the 4 classes. That is when I started to keep records. To my understanding there are 13 points a track, so as long as I got 13 points I went on to the next. Sometimes I would pass all of the cars in the 1st lap and then quit the race. If I got 13 points I went on. If I came in 5th or 4th and got 9 points I run it again and get another 4 points. However there are times where I only got 12 points and could not get any more. Back to your question when my skills change they did not change together. At first it was a few points between them but down the line they seemed to change at a greater difference. (Example: 8905/100/89/87 then 8918/100/89/89).

Track - Track Points (9996/100/99/99, or 9986/100/99/99): This happened to me on my clean up game. I didn't keep precise enough records so I forgot to race one track in one direction and didn't clean up the 4 battle points on the track. I went back to the races just before and after I went to bed, when I was most likely to have made a mistake, and got lucky and found the lost track after only a few races.

I think I am in the same boat. I will look at my records again to see if I can found it. If not I think I will have to start over and race each track with the 4 classes until I found it. Wish me luck.

 

[Orion_SR]

As I stated in that original post, I merely copied that information from two of the posts found in the original 500 post thread at GameFaqs. If I remember correctly, Orion_SR started that thread, but most of the posts were somewhat confusing and contradictory in nature.

I didn't start the Original B-spec Thread on GameFAQs.com, that honor goes to FallenOne. I joined in shortly after it started when I was able to report that a machine skill of 100 was possible, and offered the strategy of using practice mode with random cars on random tracks as a method for collecting B-spec points. (It soon became obvious that battle skill was not improved with this method.) However, I did spend a lot of time "hosting" the thread - bumping it with half-baked theories, and setting a pseudo-scientific tone that helped generate interest and participation. A side effect of the success of this thread was that the GameFAQs board was completely free of the usual criticism of B-spec as "the lazy way." However, as you mentioned, a lot of the information in that thread is completely wrong. You got the key information needed - without many of the early misconceptions. It took quite a while to finally stop misinformation from spreading through the threads a GameFAQs. This board missed most of those bad ideas.

If anyone would like to check out the Original B-spec Thread, it has been preserved at the following link. This is suggested reading only for people with new B-spec theories that are not based on the JP Method (check here first - lots of good examples of what doesn't work), or someone who really wanted to dig deeper into B-spec Theory and needs a complete reference.

http://andypsionfan.users.btopenworld.com/GT4/

 

[Orion_SR]

Tried the Super Speedway with the 4 classes of cars and there modifications and I got 1 over all point with the Lupo. I am now at 9986/100/99/99.
I did keep some records, they are not the best but I did keep some. I did not keep the results when doing the 4 classes and their modification doing Nurburgring. Once I got to 100 machine points I went on to do the 58 tracks with the 4 classes. That is when I started to keep records. To my understanding there are 13 points a track, so as long as I got 13 points I went on to the next. Sometimes I would pass all of the cars in the 1st lap and then quit the race. If I got 13 points I went on. If I came in 5th or 4th and got 9 points I run it again and get another 4 points. However there are times where I only got 12 points and could not get any more. Back to your question when my skills change they did not change together. At first it was a few points between them but down the line they seemed to change at a greater difference. (Example: 8905/100/89/87 then 8918/100/89/89).

I think I am in the same boat. I will look at my records again to see if I can found it. If not I think I will have to start over and race each track with the 4 classes until I found it. Wish me luck.

Getting only 12 points occasionally is expected. 13 track, 9 course, and 4 T-battle are easy numbers to work with, but 12 and 27/29 track, 8 and 18/29 course, and 4 and 9/29 T-battle is more accurate. So sometimes it's 12 track (maybe 14, but I don't know why), 8 course and 5 T-battle, depending on the how the fractions round out. The 12's (and 14s) should show up in a regular pattern, but no one has offered data for a perfect collection of track points (7051-10000), so fraction theory still breaks down when applied to collecting course and T-battle points at the same time.

I wasn't clear when I described what I meant by skills changing together. This probably won't help unless you want to check your data using math logic to deduce what points are missing - I wouldn't worry about. During a perfect collection, the course and battle skills will always change after the same race. Usually people are missing some battle points so this skill changes a few races later. As the normal game collection is completed, more stray battle points are collected (only 4 or 5 T-battle, because you already got the 8 or 9 course points in another race) and the next battle skill change happens only a couple races after the course skill - progress. So anyway, if someone tells you they are doing a good job on battle skill because it only changes a few points after course skill, they're right. That's tough to do in a normal game.

9986/100/99/99 - 14 more points. Are they floating points, or track points? I suspect that at least one point is a track point, otherwise course should be 100 - in theory. However, no one has ever posted a score of 9999/100/100/99. 1 missing floating battle point may not be possible, or still score as 9999/100/99/99. If you decide to go after the track points first, you may want to use the alternate set of cars. I'll repeat the names I assigned to the cars. I keep hoping they will be adopted as standard descriptions.

Lupo A: VW Lupo 1.4 default
Miata A: Mazda MX-5 1800 default
Miata B: Miata + supercharger
Miata C: Mitata + turbo 3
RSC A: Toyota RSC Rally Raid default
RSC B: RSC + medium racing tires
350Z A: Nissan 350Z Concept LM default
350Z B: 350Z + Turbo 4

JP track tuning (I used same cars without the racing tires for the Lupo or Miata)
Lupo R: Lupo A, + medium racing tires
Miata R: Miata B, + medium racing tires
RSC R: RSC B
350Z R: 350Z A

If you are searching for points by repeating the tracks, and aren't sure if you have all of the F-battle points, then use the alternate tuning set to repeat the races, just in case they find some lost floating points that were not collected the first time. So if you used Lupo A, Miata B, RSC B, and 350Z A the first time. Then use Miata A, Miata C, RSC A, and 350Z B when you repeat the tracks. These cars should earn the same track points, but might find lost floating points. I still suspect you are missing some floating points, but you may as well run the courses to find any stray track points. If you get to 99xx/100/100/99, please let us know.

Oh, yeah. Good luck.

 

[SportWagon]

Getting only 12 points occasionally is expected. 13 track, 9 course, and 4 T-battle are easy numbers to work with, but 12 and 27/29 track, 8 and 18/29 course, and 4 and 9/29 T-battle is more accurate. So sometimes it's 12 track (maybe 14, but I don't know why), 8 course and 5 T-battle, depending on the how the fractions round out. The 12's (and 14s) should show up in a regular pattern, but no one has offered data for a perfect collection of track points (7051-10000), so fraction theory still breaks down when applied to collecting course and T-battle points at the same time.

If some of the totals are maintained internally somehow, then perhaps 4 and 9/29 sometimes "rounds to" (completes) 5, while 8 and 18/29 also rounds to 9.

I.e.

0:0 and 0/29 - truncates to 0, rounds to 0
1:4 and 9/29 - truncates to 4(+4), rounds to 4(+4)
2:8 and 18/29 - truncates to 8(+4), rounds to 9(+5)
3:12 and 27/29 - truncates to 12(+4), rounds to 13(+4)
4:17 and 7/29 - truncates to 17(+5), rounds to 17(+4)

So, taken by themselves, you get increments of 4 or 5.

Similarly, 8 and 18/29 gives increments of 8 or 9. If the game reckons it has simultaneously had an increment of 9 plus and increment of 5, perhaps that gives the 14's. I.e. perhaps the course and T-battle totals are maintained in fractions, and reckoned as whole numbers to be added to the skills total independently.

 

[Orion_SR]

If some of the totals are maintained internally somehow, then perhaps 4 and 9/29 sometimes "rounds to" (completes) 5, while 8 and 18/29 also rounds to 9.

If the game reckons it has simultaneously had an increment of 9 plus and increment of 5, perhaps that gives the 14's. I.e. perhaps the course and T-battle totals are maintained in fractions, and reckoned as whole numbers to be added to the skills total independently.

You may be correct. However, I'm fairly sure that no numbers are "rounded up." I suspect that the fractions of different types of points may not always add to whole level points as expected. Whole points of each type appear required to build skills for instance. Occasionally, both course and battle get large enough remainders to score 9 course and 5 battle on the same race, and that's when we see 14s popping up. In order to solve this puzzle, a data set with a perfect collection from 7051 thru 8500, would be helpful.

I think this just rephrases your argument, without considering rounding up as a possibility. Completing the whole number for that skill point appears to be a more accurate description.

The 7375, 7750, 8125, 8500, 8875, 9250, and 9625 collection milestones should have no extra fractions for any type of point. For people making systematic collections from 0 to 7051 thru 10000, this is a place to stop and clean up dropped fractions before moving on.

 

[SportWagon]

Truncating instead of rounding merely changes when in the sequence you will get the "extra increment".

My records
http://www.geocities.com/gt2toxs/gt/logs/gt4-game1.txt (serious B-spec'ing starts about day 1111, July 8, 2005) indicate that, after I had got 100 machine skill points, I never got more than 13 points with one car at one track. I.e. all such >13 scores were before I had collected all machine points (and therefore, I gather, I still had floating points available to collect).

Although I had 100 machine points, I was, while working on completion, apparently missing 2 T-Battle points for the 350Z category at the Super Speedway. Those summary records infer I must have previously Bspec'd some event at the Super Speedway with that class of car, but it must have been part of a series, not a single race. (And therefore I don't log it individually). I A-spec'd that Enduro, and also that race in the P/GTW series. The event was possibly fresh in my memory, since my initial B-spec'ing with the 350Z at the Super Speedway was sort of shoddy. (A single race only).

 

[Orion_SR]

Sometimes I would pass all of the cars in the 1st lap and then quit the race.

I have suggested that quitting just after passing the leader, and about one lap, was an effective way to earn floating points on the Nurburgring. But, my early experience suggests that finishing was usually important to earn T-battle points on most tracks. However, when I did pass everyone too soon and quit early, I was occasionally surprised to find all 13 points earned. Anyway, just to be clear, I would not recommend quitting on purpose as a method for earning track points.

If you know there are course points on the track, then quitting early can quickly show if that car will collect them. But if 0 points are earned, you haven't really tested for the battle points until you finish a close race.

 

[Orion_SR]

My records indicate that, after I had got 100 machine skill points, I never got more than 13 points with one car at one track. I.e. all such >13 scores were before I had collected all machine points (and therefore, I gather, I still had floating points available to collect).

Okay, I found a pattern in the JP data set. Check the link below for reference.

http://www.geocities.jp/j8ba7yrg/gt4/7_10000.html

Whole numbers occur at certain milestone scores. These are listed below, followed by the score in the JP data, the line number for that race, and the points earned for the race.

Milestone, JP score, Race #, points earned
7000 (hidden on the Ring)
7375, 7375, #63, +14
7750, 7750, #92, +14
8125, 8124, #121,+13
8500, 8499, #150, +13
8875, 8874, #179, +13
9250, 9249, #208, +13
9625, 9624, #237, +13
10000, 9999, #266, +13

First, notice that #92 was the last time a score of 14 was reported, and race #63 was the only other score of 14. Once a fraction was dropped, the 14s dissappeared. They appear to be an indication that all points are collected - whole points of each type. Also, we have enough data to find the pattern of 12s and 14s, perhaps someone could find a way to make that fit with fraction theory.

I think I was also finally able to find where the JP author dropped his point. Everything was going fine up to race #92. But a fraction was dropped before race #121. Race #96 seems to match the description of race #267 (10000th point) and fits in this range. Race #121 appears to be the 1st place to detect that it was dropped.

Based on the points gained between race #63 and #121, the pattern of 12s, 13s, and 14 for a perfect collection appears as follows. Do you think you can make the math fit the pattern? Anyway, this pattern is the same as predicted by the JP author's prediction table.

http://www.geocities.jp/j8ba7yrg/gt4/3000_1.html

12 pts, (12 x 13 pts), 12 pts, 12 x 13 pts, 12 pts, 13 pts, 14 pts - should repeat every 29 races.

 

[SportWagon]

Using two totals, truncating then adding, I get 8 +14's per cycle of 29.
Using two totals, rounding then adding, I get 4 +14's per cycle of 29.

I even verified that if you add the sum of the two numbers (rounding or truncating) you get only +12's and +13's.

So I've not yet found a scheme which gives exactly one +14 per cycle.

 

[HugeF1]

A quick q. Does "Bob" use NOS?

 

[FoolKiller]

A quick q. Does "Bob" use NOS?

No. He could use it sometimes but he would probably hit it while in a turn or something silly like that.

 

[Orion_SR]

I was considering cautioning people about staying on the topic of increasing B-spec stats, but I never found a good beginner's B-spec reference on this board for some of the basics of directing the B-driver. If good references are found, it might be helpful to collect links to the sources and possibly include them on the B-spec hot link thread started by SportWagon.

https://www.gtplanet.net/forum/showthread.php?t=69405

The following link contains some information on Pace and Overtake theory. It seems to reflect the current best strategies. The pace section could use a little expanding for lower settings.

https://www.gtplanet.net/forum/showpost.php?p=1687606&postcount=33

What else? I don't monitor the board much anymore so I forget the common questions.

Winning is not important, but the races must be close.
AI-drivers don't use NOS.
The little red/orange arrows indicate a faster time in the previous sector.
Triple Speed always resets to normal during a pit stop.
B-spec driver's will turn on the Pit icon when the fuel is low or tires are worn.
Hit Circle to force the B-driver to pit, or to cancel the B-driver's request to pit.
Sarthe is sometimes glitchy - check this thread for more info https://www.gtplanet.net/forum/showthread.php?t=65361
The Wall in NY is always a problem. Tune the car for the turn, or try a different car.

Here's an idea for explaining how fractions might get lost. Let's say I run a race and come in 5th and collect 11 points. I figure the game penalized me for the poor finish and only gave me half the T-battle points. 4 and 9/29 T-battle divided by 2 would be about 2 and 4/29th. So I repeat the race for the stubborn points and do a little better, but not well enough, so the game only coughs up the other half of the points. So now I've collected 4 and 8/29th T-battle points, 1/29 short of the total for that track. I have no evidence that the game works in this manner. As far as I can tell, there may be 4 and 18/58 T-battle points on each track.

SportWagon, thanks for working through the numbers for the point progressions. I am a little frustrated to have this much data and still not be able to find the pattern.

 

[SportWagon]

A while ago I Bspec'd the last series of the Professional Hall (GT World Championship), and lost several races but won the series.

I repeated that series last weekend, B-spec'ing after qualifying as well as I could, usually first. B-spec won all races except for Super Speedway (I did that one) and Hong Kong (notes not available here yet; qualified 4th, finished 3rd, I think).

The overpowered Minolta Toyota seemed to drift wide in the central hairpin. I think perhaps I should win a new one and not put the cage in. Trying less power might be reasonable, too.

I guided it around New York by using pace 1 for most of the straight preceding the bad corner, and bumping up to 5 after braking had finished; that seems to cause it to apex later and thereby hit the wall less hard and at a more appropriate angle, often missing it altogether.

It still had trouble at Mulsanne Hairpin as well, so I set the pace down there too. In fact, it needed help with the final chicane of Suzuka, too. Or, at least, it couldn't manage the chicane after doing the preceding section at pace >3 (it ended up in the kitty litter on exit).

 

[SportWagon]

Here is a sequence of B-spec point progression, through a cycle of 29.

12(+12) - 8 and 18/29, 4 and 9/29, 12 and 27/29
25(+13) - 17 and 7/29, 8 and 18/29, 25 and 25/29
38(+13) - 25 and 25/29, 12 and 27/29, 38 and 23/29
51(+13) - 34 and 14/29, 17 and 7/29, 51 and 21/29
64(+13) - 43 and 3/29, 21 and 16/29, 64 and 19/29
77(+13) - 51 and 21/29, 25 and 25/29, 77 and 17/29
90(+13) - 60 and 10/29, 30 and 5/29, 90 and 15/29
103(+13) - 68 and 28/29, 34 and 14/29, 103 and 13/29
116(+13) - 77 and 17/29, 38 and 23/29, 116 and 11/29
129(+13) - 86 and 6/29, 43 and 3/29, 129 and 9/29
142(+13) - 94 and 24/29, 47 and 12/29, 142 and 7/29
155(+13) - 103 and 13/29, 51 and 21/29, 155 and 5/29
168(+13) - 112 and 2/29, 56 and 1/29, 168 and 3/29
180(+12) - 120 and 20/29, 60 and 10/29, 181 and 1/29
193(+13) - 129 and 9/29, 64 and 19/29, 193 and 28/29
206(+13) - 137 and 27/29, 68 and 28/29, 206 and 26/29
219(+13) - 146 and 16/29, 73 and 8/29, 219 and 24/29
232(+13) - 155 and 5/29, 77 and 17/29, 232 and 22/29
245(+13) - 163 and 23/29, 81 and 26/29, 245 and 20/29
258(+13) - 172 and 12/29, 86 and 6/29, 258 and 18/29
271(+13) - 181 and 1/29, 90 and 15/29, 271 and 16/29
284(+13) - 189 and 19/29, 94 and 24/29, 284 and 14/29
297(+13) - 198 and 8/29, 99 and 4/29, 297 and 12/29
310(+13) - 206 and 26/29, 103 and 13/29, 310 and 10/29
323(+13) - 215 and 15/29, 107 and 22/29, 323 and 8/29
336(+13) - 224 and 4/29, 112 and 2/29, 336 and 6/29
348(+12) - 232 and 22/29, 116 and 11/29, 349 and 4/29
361(+13) - 241 and 11/29, 120 and 20/29, 362 and 2/29
375(+14) - 250 and 0/29, 125 and 0/29, 375 and 0/29

Column 1 - the total (relative to base value at cycle) as predicted by the Japanese author (the increment in parentheses is what he predicted)

The three whole numbers with a fraction are, Course points, T-battle points, and the sum of those two.

I'm can't find a way to resolve them.

 

[Orion_SR]

I can't find a way to resolve them.

For a minute there I thought you had it figured out, but closer examination shows that your table describes the point predictions based on observation (JP prediction table), what would be expected using fraction theory and addition, and that these are close, but different. Thank you for posting this data, maybe it will help someone else solve the puzzle.

With your table as a reference, I re-examined the JP collection data. I was hoping that I might be able to detect the minimum B-spec unit, but was unsuccessful. I was expecting that the pattern of 12s might shift a race earlier due to the dropped point. However, as noted in your table, the 12s are occuring one race sooner than expected anyway. So despited the dropped fraction, the pattern remains the same, except the 29th race scores +13 instead of +14, so the 1st race in the next sequence scores +13 instead of +12. The 12s at the 14th and 27th races are in the same position.

So, while these numbers and methods are fresh in mind, consider again that the minimum B-spec unit might not be 1/29th of a point. I have suggested that 1/58 would be a waste of space, and dropping 25/29th of a point would surely mess up the sequence of 12s and 13s. Since you did the math with 1/29th and it didn't work out, what would happen if you tried the calculations with 5/29th as the basic unit. Any rounding or truncating for either the course or T-battle points, must include the entire 5/29th of a point.

Hopefully, I have described this concept well enough, because it's still a little fuzzy in my mind. I couldn't keep the numbers straight long enough to even work out a method to attempt a solution. Besides, there was little interest in these ideas at the time.

 

[SportWagon]

Assuming there is no error in the 4 and 9/29 and 8 and 9/29 assumption, it's difficult to see what different the actual minimum unit would make, since both of those are multiples of 5/29 anyway (125/29 and 250/29, respectively). I.e. we're talking about maximized scores here.

That said, I'll try to think of something weird they might be doing in the process of conversion and rounding-off. Especially if the scores are being maintained to a larger total and scaled down to a nice-looking 10,000.

Last night I remembered again when I took my 9998/100/99/99 card, and got to 9999/100/99/99 by using only the 350Z at Super Speedway, and could not get to the full 10000 as if one of the other cars had been contributing small fractions. If I was really unluckily lucky, perhaps some small fractions actually came from other races I retried before going back to Super Speedway. Getting 9999/100/99/99 says something about something. Not sure what. I will try to double-check that is exactly what it got to. I think I posted it earlier in this thread.

Bah. I didn't report the skills totals. But I recall it as being 9999/100/99/99;
https://www.gtplanet.net/forum/showpost.php?p=1837859
Finding the above also led me again to...
https://www.gtplanet.net/forum/showpost.php?p=1838510
which I should review for possible ideas wrt the possible rounding algorithm.

 

[Orion_SR]

Assuming there is no error in the 4 and 9/29 and 8 and 9/29 assumption, it's difficult to see what different the actual minimum unit would make, since both of those are multiples of 5/29 anyway (125/29 and 250/29, respectively). I.e. we're talking about maximized scores here.

That said, I'll try to think of something weird they might be doing in the process of conversion and rounding-off. Especially if the scores are being maintained to a larger total and scaled down to a nice-looking 10,000.

I am very confident that the fractions I described accurately predict my point progression. Also, it just seems too obvious that 2000/232 and 1000/232 would yield these results. However, you are correct in questioning these assumptions. I wouldn't suggest spending a lot of time on the minimum B-spec unit question as a method for solving the track point progression puzzle. This idea was based on a hunch, but broke down whenever I tried to apply it.

The reason I thought this might be a possibility is based on skill weights for course points (100C) and T-battle points (100C, 33 and 1/3B). These weights predicted that the JP author's skills would change from 73 to 74 at race number #65 instead of #66. I saw a similar pattern when tracking my battle skill while collecting T-battle points only; 33 and a fraction larger than 1/3 was not enough to increment the skill. Translations of the JP notes seem to suggest that the author came to the same conclusion as I did, whole points are required to build skills. This may not be the case, but this is the clue that suggested that the B-spec level and skills are not simply a lense through which to view the B-spec units. There appears to be some sort of calculation to convert units to level points, and then level points to skills.

I think this is the last calculation that needs to be checked, then I'll feel a little more confident that everything is in order. Thank you for your efforts on the mathematics. I'm not sure if the whole point hypothesis for skill will help with predictions of track point progression, but it is a clue worth considering.

 

[Crunch_Monkey]

Hmmm, i'm thinking it would be abit easier to just email PD and ask them just how does their rating/score system work for B-spec. Maybe even send along a few links that lead to this fourm/threads to help show there really is a large interest in how this works by the people that are still playing GT4 on a regular basis. But then wheres the fun in that?

 

[SportWagon]

The closest thing I've seen to the divulgence of "secrets" of the GT series by developers is when the designer of the Nike 2022 (and Saleen S7) dropped by here. (And that wasn't really at all close to divulging secrets).

https://www.gtplanet.net/forum/showthread.php?t=57094&page=1&pp=30
https://www.gtplanet.net/forum/member.php?u=57727
https://www.gtplanet.net/forum/search.php?do=finduser&u=57727

I think the developers believe very strongly that the secrets should remain secrets.

 

[Orion_SR]

I think the developers believe very strongly that the secrets should remain secrets.

That would be my guess as well. Besides, I don't think there really is all that much interest in solving the math puzzle. Most people don't seem interested in playing a B-spec test game and trying for the perfect collection. Learning more about the details of how the points add up probably won't help anyone find their lost points. However, it seems reasonable to assume that anyone who collects 14 points every 29 tracks has collected all of their track points - maybe. If it wasn't for the problems with translation, I might be tempted to summit our current theory in the hope that they might set us straight on any misconceptions. However, I currently suspect the reason the numbers don't add up as expected is because the programmers deliberately tweaked the equations to hide their secrets.

 

[SportWagon]

I think it's possibly because things were tweaked to ensure that slightly less than 10,000 points doesn't view as 10,000.

 

[StevenDunn99]

A lot of these topics are getting quiet on GTPlanet, but better late than never I am now trying to maximise my bspec scores. Some great info here, thx to all.

I started my quest with c5500:55:55:55 (really wished I had written the exact numbers down now) in the bank from bspec'ing the enduros, 1000 miles and F1WC first time round.

Using JP techique at the Ring with the sub classes I got to 7557:99:75:75. I have had a few trips back to the Ring to get 100 machine pts with no luck, ocassionally I get a couple of pts, but that's it. I guess there are some floaters somewhere.

At World Circuits, completed with the Lupo, now using MX-5+s/c. Current total stands at 7835:99:78:78. Everything seems to be moving along nicely. I have 1x14pt race and 1x12 pt race, rest are 13pts. Battle pts have so far increased 3 races before Course pts, and course pts go up at just over the 100 Bspec pts, eg 7625, 7725, 7825. That also implies some missing Ring points but as I didn't start from scratch I'll think about that much later.

Technique: try to overtake nothing <15secs in, but sometimes that is difficult, and try to win or come 2nd by less than 5 secs, usually by sticking on the tail of the leader(s).

So far, seemingly so good.


Steven

 

[Orion_SR]

A lot of these topics are getting quiet on GTPlanet, but better late than never I am now trying to maximise my bspec scores. Some great info here, thx to all.

The thread may be slow, but I still monitor it for activity. It sounds like you have done your research, have a pretty good grasp of the details, and will make good progress toward 10000 B-spec points. I suppose it is possible for your missing floaters to show up while working the other tracks, but people generally don't collect many points with the standard JP tuning set if they can't be collected on Nurburgring. You'll probably need to experiment with some other cars and tunings to get the missing points. A few people have reported which cars and tuning they used to finally managed to collect stubborn floating points. Check the 10000 announcements and follow ups if you didn't find them earlier.

I find it interesting that you are collecting 14 point races without the usual pattern of 12s. Perhaps this is because the floating points haven't been maxed out yet. I suspect the 14s are a good sign, expecially if the pattern repeats every 29 races. If the pattern is consistent, I'd be interested in knowing the full sequence of a 29 race collection. Assume a gain of 14 points to be race 29 and the next race to be the beginning of the cycle.

Battle pts have so far increased 3 races before Course pts

Well done, it's difficult to keep the battle skill above the course skill. As a matter of convention, skill levels, or skill points, are often referred to as just "skill" to avoid confusion with all the other theories on what types of points there might be.

Your technique looks sound. I might emphasize repeating short tracks, especially Super Speedway. Don't worry too much about the passing rules. I suspect this may be more important for floating points than track points, but it's hard to tell. I ignored the passing rules whenever they were annoying, especially on short tracks. You'll have considerable more control of the races now that the Lupo is out of the way.

 

[StevenDunn99]

I find it interesting that you are collecting 14 point races without the usual pattern of 12s. Perhaps this is because the floating points haven't been maxed out yet. I suspect the 14s are a good sign, expecially if the pattern repeats every 29 races. If the pattern is consistent, I'd be interested in knowing the full sequence of a 29 race collection. Assume a gain of 14 points to be race 29 and the next race to be the beginning of the cycle.

Amongst my 13 pts races I now have the sequence of 12,14,12 then 17 x 13pts then 12,14,12 followed by 13s again. I have done the 12 pters 3 times with no gain. The pattern is scheduled to repeat when I use the RSC car at Suzuka East.

I am keeping a spreadsheet of progress, and currently at 7886:99:78:79 so expecting course skill points to go up in 2 or 3 races time.


Steven

 

[StevenDunn99]

I think the sequence will not be the same for me. As I had already used Bspec for some races, (Enduros, 1000 miles and F1 GTWC) I found eg that I got no points at Tsukuba whatever I used, so I probably already have the necessary points. I will find this elsewhere as I progress.

Note: when I used Bspec first time round I did use it intelligently especially early on, eg reducing to 3 for certain parts of certain tracks, overtaking was never immediate and usually involved pit stops, I never left the overtake on continuously, which was probably why I had, IMO, good Bspec starts before I started to go for max stats.


Steven

 

[Orion_SR]

I think the sequence will not be the same for me. As I had already used Bspec for some races, (Enduros, 1000 miles and F1 GTWC) I found eg that I got no points at Tsukuba whatever I used, so I probably already have the necessary points. I will find this elsewhere as I progress.

The pattern of +12s, +13s, and +14s is one of the major mysteries of fraction theory. I'm looking for new data from someone without all the machine points and making a track point collection. What I'm looking for is a pattern that repeats every 29 races and includes at least one +14. I'd like to compare this pattern with the one found by the Japanese author and listed below.

+12, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13,
+12, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13,
+12, +13, +14

If I understood your pattern correctly, it would look something like the pattern below, probably with the +12 in a different position.

+13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13,
+12, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13,
+13, +13, +14

If your pattern holds all the way to the end of the track point collection, or at least long enough to be confident in the sequence, it might shed some light on why the fractions don't add up. Also, it might be possible to predict where to start looking for dropped track points, even if all the floating points haven't been collected.

Ignore these courses when collecting track points.

• Rally Tracks
• Test Course
• Tsukuba Wet
• Suzuka West

 

[StevenDunn99]

Hi Orion_SR, no my pattern is (after doing the Ring races):

+13, +13, +12, +14, +12, +13 (17 times), +12, +14, +12, +13 (3 times so far).

I keep records (car type, diff, track, all bspec pt data, no of attempts) of everything so I can update further as I progress.

I am using the race list posted earlier in thread, which exclude those you mention above.


Steven

 

[Solid Lifters]

The pattern of +12s, +13s, and +14s is one of the major mysteries of fraction theory. I'm looking for new data from someone without all the machine points and making a track point collection. What I'm looking for is a pattern that repeats every 29 races and includes at least one +14. I'd like to compare this pattern with the one found by the Japanese author and listed below.

+12, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13,
+12, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13,
+12, +13, +14

If I understood your pattern correctly, it would look something like the pattern below, probably with the +12 in a different position.

+13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13,
+12, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13, +13,
+13, +13, +14

If your pattern holds all the way to the end of the track point collection, or at least long enough to be confident in the sequence, it might shed some light on why the fractions don't add up. Also, it might be possible to predict where to start looking for dropped track points, even if all the floating points haven't been collected.

Ignore these courses when collecting track points.

• Rally Tracks
• Test Course
• Tsukuba Wet
• Suzuka West

Interesting stuff, man. I can use a bit of help myself, because I don't quite get it yet. Here goes...

I have 8792 B-spec Skill Level, with 100 Machine Skill points, 87 Course Skill Points and 86 Battle Skill points. Why am I not getting any more B-spec Skill Level points when racing in the Family Cup races? Do I need to branch out to the rest of the game? I thought Battl Skill and Course Skill points come from the tracks, not the races. Yet, I raced in a Endurance race for 9 hours in B-spec mode and won points! Yet, I raced on that track more than once with different kinds of cars. So, what's up?

Can you tell me what I should do, like where to go and what to use, by just looking at my points? I figure you can and any help would be greatly appreciated.

 

[ANK]

Interesting stuff, man. I can use a bit of help myself, because I don't quite get it yet. Here goes...

I have 8792 B-spec Skill Level, with 100 Machine Skill points, 87 Course Skill Points and 86 Battle Skill points. Why am I not getting any more B-spec Skill Level points when racing in the Family Cup races? Do I need to branch out to the rest of the game? I thought Battl Skill and Course Skill points come from the tracks, not the races. Yet, I raced in a Endurance race for 9 hours in B-spec mode and won points! Yet, I raced on that track more than once with different kinds of cars. So, what's up?

Can you tell me what I should do, like where to go and what to use, by just looking at my points? I figure you can and any help would be greatly appreciated.

You have to use different cars/setups on most of the tracks in family cup. If you look back earlier in this thread there is a list of the different setups needed.