GT5 Used Car Database [in progress]

  • Thread starter Famine
  • 1,863 comments
  • 428,844 views
Another alternative (though unlikely) is how unhappy people were with how long it took to find some cars so PD increased the frequency of those cars in this update.
 
Another alternative (though unlikely) is how unhappy people were with how long it took to find some cars so PD increased the frequency of those cars in this update.

i thought of that too. after i installed the update, i had a corvette c5-r and a altezza touring car. cars that only appeared twice in the old master list
 
i'm curious though. if it is an actual cycle like we think it is, do you think they made it even longer with the additions of new colors, or did they just modify the cars already in the cycle and then just reshuffled them? it's just interesting to me how we are still all at the same points in the cycle as before the update but all the cars are different

Well there are two thoughts I have on this. I'm sure there are other ways but these are the ones I though of.

  1. The list is generated by PD on their computers and implemented on the disk or in the update. Your initial list along with your current day would determine what cars you would see. This list cannot be infinite although it might be unreasonably long.
  2. The algorithm is contained in the game and the list is generated on your PS3. It would use parameters like current game day, car rarity, initial start point, etc to generate the hextet (of cars, ignoring mileage/colors/price) for a given day. The algorithm may be made in a way that we all end up seeing the same cars at one point or another, maybe since the initial point is used in the algorithm or maybe it is just a fluke we have going. This would also allow them to test more easily to find out if the lists are working how they want (since they don't have to worry about it being totally different for different consoles). This list would be infinite.
In both cases we know that the start point is generated on the PS3 based on its serial number (or some other unique identifier).
 
That is good news - too bad they had to switch the cars though :indiff:



Good idea about the save games - I loaded up a few from an account I was using to grab early UCD cars - so I've updated my spreadsheet to 46 days worth of data:

https://spreadsheets.google.com/pub?hl=en&hl=en&key=0AkKVFQSqKBCudF9QbWlHcWU2blpCYzJkNWJZcWpfLVE&output=html

OK...checked my backup account...new day 1 now matches pr0cy's car 5, day 22, starting with Nissan - SKYLINE GT-R V-spec (R33) '96....👍
 
Which is exactly the same place you were before.

I've also checked some various game saves and my day 1 matches the offset from pr0cy's days exactly as before.
 
I'm keen to find a way to analyse the new data succinctly and without spending a huge amount of time on it and so before I start I'm trying to think of useful ways to monitor packets of data.

Something I think might stand out in the new data set is if a car in a particular colour or of the same mileage reoccurs. With my previous method of data collection it was easy for me to see which cars were more common and so I have just looked at two cars; the Honda S2000 (US) '99 and the Honda Integra Type R (DC2) '98. Both of these cars appeared over twenty-five times each. From looking at the days they appeared there isn't any immediately obvious pattern however, if these cars, or any other, shows frequently in a particular colour then it might help decipher this thing.

I've just taken a look at SlipZtrEm's colour charts to see how many times a particular car has appeared in a different colour. From the two I stated above, there is only four or five colours listed. On the basis that this is likely to remain the case, I hope this might help.

I must say I'm really liking the way the new UCD unfolds; to see any of these cars in different colours is enough to make me smile after the repetitive nature of the previous cycle.
 
I don't have hard facts to prove this but my feeling is that race-rare cars are popping more frequently in my ucd now.
 
I don't have hard facts to prove this but my feeling is that race-rare cars are popping more frequently in my ucd now.

ina matter of less than 25 gamedays i have seen the #51 Oreca Viper twice... so yeah im agreeing with you; race cars are going to be appearing in the UCD more often now.
 
ina matter of less than 25 gamedays i have seen the #51 Oreca Viper twice... so yeah im agreeing with you; race cars are going to be appearing in the UCD more often now.

i saw the jag xjr-9 race car in a span of 3 days and i have seen a definite increase in touring and other race cars
 
Does anyone have any idea of how far this update shifted the database? I was at around day 1055 on the master list #2 prior to the update and now my UCD is not on either master list #1 or #2 anywhere. I shifted into unknown territory.
 
Does anyone have any idea of how far this update shifted the database? I was at around day 1055 on the master list #2 prior to the update and now my UCD is not on either master list #1 or #2 anywhere. I shifted into unknown territory.

It is looking like this is a completely new list so I don't expect us to see any overlap with the previous lists. :ouch:
 
I was so confused when I noticed nothing matched any database all of sudden. Good thing though is that on the new 2nd day I had the GT-R Concept LM Race Car that I've never seen before.
 
I downloaded the search-able UCD excel file but it think my excel program is too old to run the macro because I still cannot figure out which master list I have. Can someone who knows how to use the excel file do me the biggest favor ever and let me know which master list my UCD is, hopefully there is a match.

If you can, please please please PM me with which master list my UCD is. I would be incredibly grateful!

My cars in my UCD right now, from top left, to top right, to one cell lower left, over one right, one lower left, over one right etc...

Chrysler - Crossfire '04
Toyota - VITZ RS 1.5 '00
Toyota - MR2 1600 G-Limited Super Charger '86
Mitsubishi - Lancer Evolution VI GSR T.M. '00
Blitz - ER34 D1 SPEC '04
Daihatsu - MIRA TR-XX Avanzato R '97
Daihatsu - Copen Detachable Top '02
Nissan - SKYLINE GT-R V-spec N1 (R34) '99

Here is a link to the search-able UCD excel sheet from this thread:

https://docs.google.com/viewer?a=v&p...COKQ9MsG&hl=en
 
Last edited:
I agree with Fastas that the car color, price and mileage could be used as an additional marker for the the cycle.

Only time will tell but I think one should still be able to find their unique spot in the list with only the car names and order but the other info can only help. The issue I ran into was the time it took to record the exact color of the car was really making progress slow in the previous version.....

Yes, recording the colour does slow things down a little but, I'd rather be able to prove that the colour wasn't important that just assume it isn't. If after 100 days or so a NSX appears in Rosso Corsa Red then I think it becomes pointless. If the work SlipZtrEm has done on the colour chips proves to be correct then it helps all of us.

If, as it is looking, the cycle is following the same pattern as before, then we could spend another three months just logging the basics, which is useful but, doesn't tell us anything new in terms how it might work.

This is what I propose we collect,...for now at least.

(master list day)/Manufacturer/model+year/odometer/cost/PP/colour/username/user's day.

day order (as before?)
30 cars in view as days from the top in groups of six = 54321

Order within the day
6-5
4-3
2-1

D.
 
This is what I propose we collect,...for now at least.

I agree. With the original list I started recording all those (except PP, obviously) but eventually gave up when most of it was clearly irrelevant. While it was nice to be able to rush through without needing to record so much info, it does seem far more important this time.

You and I are on the same list or part of list. Where are you currently recording from? While it's nice to double-check lists, that seems like a luxury at the moment.
 
I agree. With the original list I started recording all those (except PP, obviously) but eventually gave up when most of it was clearly irrelevant. While it was nice to be able to rush through without needing to record so much info, it does seem far more important this time.

You and I are on the same list or part of list. Where are you currently recording from? While it's nice to double-check lists, that seems like a luxury at the moment.

I haven't really started yet. I cleared an account to look at day one and my main account is at/around day 1700.

I'm happy to start from day 1.

I don't want to get ahead of myself here but,...would it be easier to pick say a 50 day cluster i.e. I will do days 1-50 and in the mean time someone else can pick up days 51-100? maybe that's too complicated...
 
I don't want to get ahead of myself here but,...would it be easier to pick say a 50 day cluster i.e. I will do days 1-50 and in the mean time someone else can pick up days 51-100? maybe that's too complicated...

That's pretty much what I was hoping for. I've not got much data yet either - I have my days 1-21 with all data and 542-563 missing some mileages and prices. As far as I can tell those are from day 61 car 4 and 602 car 4 on TornadoAlley's master list #2. For the moment I'll carry on with what I'm doing - recording both of them slowly while I do various events - but I'll happily focus on one or the other to avoid overlapping.
 
what i have noticed now about the new UCD rotation is that the color of the cars changes. in the same 'UCD Day' two Corvette C5 Z06s, one yellow and one was machine silver. Both had different mileage, so obviously different prices and PP values. before that C5 Z06 was always silver. I have seen the 69 Ford GT-40 race car twice now, and both Oreca Vipers #51 several times and the #91 just once. I had never seen the #91 up until V1.06.
 
i'm going to be taking a break from recording cars for now. pd took some of the drive i had to record the days after they switched it up :indiff:
 
I've cycled through a new account to reach a save point from an older account and the cars and sequences match up. This means that the sequence of the used cars has been reshuffled but in terms of day progression, we should all be in the same place relative to each other now, as we were before.
.....

I was worried about this as well.^^^^

What if the UCD shuffle had broken the sequence of the list, and a new game would never match up its list to the list being generated by an old game that had previously been cycled and recorded?????

Thankfully, as Fastas has mentioned above, the old and new lists match up!

To test this I launched an old game that I had cycled thru 15 days and wrote down the cars being listed in the UCD after the V1.06 update.

I then started a new game (which will be my primary UCD tracking game) and recorded the cars up to day 15 and they start to match!!!!! Yeah!!!

I have now recorded the first 25 days in this new game, so I will put them on a spreadsheet shortly and send them along to TA.

Going forward.....I much prefer to just record the same basics as before (more time is spent driving rather than writing).

However, I guess I could be persuaded to record the colors, mileage and cost. These three pieces of information are useful for some since we have some car collectors who chase the low mileage cars and try to obtain all car colors.

How important is the PP value for purposes of this database?

Respectfully,
GTsail
 
Going forward.....I much prefer to just record the same basics as before (more time is spent driving rather than writing).

However, I guess I could be persuaded to record the colors, mileage and cost. These three pieces of information are useful for some since we have some car collectors who chase the low mileage cars and try to obtain all car colors.

How important is the PP value for purposes of this database?

Respectfully,
GTsail

I'm of two minds regarding what to record. I'm of the opinion that any information will be helpful for the database. The other stats are going to be helpful in potentially recognizing if there are any patterns to what is appearing and yes, finding any low mileage cars.

The PP value will be different for cars with different mileages so in a way it is telling us the same thing as the odometer with regard to different entries of the same car.

The issue that I foresee continuing with the google docs spreadsheet is that it is very limited to the total # of cells you can have in one document. I can either share a spreadsheet that must be downloaded through google docs which can contain all of the extra info (cost, odometer, color...) or I continue to keep a google spreadsheet that contains only the cars and their order.

It would be simpler for me to maintain one list that I upload a new version on a regular basis for downloading but if no one is going to use it then I don't want to go to the effort.
 
I think Color, Price, Mileage etc is overkill......

I think it is overkill to find where you are in the master list but I think it could be useful in looking into any patterns in the list as it is another piece of information. For example, is the same mileage always paired with a particular color?

I am going to incorporate any info that people contribute. If you only collect the same info as we had been doing before I think that is the most crucial data for the database. The other stats I am going to have a space reserved for but they will be left blank. Anyone who includes the extra info I will not throw it out.
 
I think Color, Price, Mileage etc is overkill......

It might well prove to be and I can understand that the more time spent recording it is less time for anything else.

I'm of the opinion though that without at least qualifying weather or not it is valuable information, it's too early to discard it. I'm probably more focused on finding a pattern to the cycle than just helping out GTP_JoeBloggs find a low-mileage MX-5 in pastel green.

As I posted earlier, a car that appeared over 25 times in the last dataset only had five different colour variables based upon the colour chips data and so, I'm curious to know if, for example, there are only five variants of that car but cycled five times.

I'm only prepared to collect so much data myself and if after a period of time or in such circumstance that cars start appearing in other manufacturer's colours then I will probably just stop rather than scale down the data collection.

I don't mind either way about the PP's btw.

D.
 
Which days are already mapped for the new database? I have some carefully collected data which I started logging on day 283 (thus, the data goes back to day 279) and stopped logging on day 287 (after which I got tired of stopping to log my UCD by hand after every race).

Should I post this information publicly or PM it to someone? If PM, who should I send it to?
 
Back