- 2
- Australia
That's what they need to remove
The WTC 600 in Tokyo done without exploits is genuinely one of the hardest races in single player in gt7 - share your setups, tips, and cars
- Thread starter The_It_Jojo
- 2,463 comments
- 343,008 views
- 1
- New Zealand
This is my first post in GTPlanet in more than a decade! I've been trying to understand the AI's response to different difficulty settings on this race lately, and I feel like I'm left more confused than when I'd started.
I recently got back into GT7 after a fairly long hiatus, largely inspired to after lurking this thread and seeing the different setups everyone used to beat this race. Since I'd been out for a while, I set my game difficulty to Easy to get back in the rhythm of this race. I was already quite surprised by how quick the AI was in this race on Easy. I had learned from scrolling through this thread that the AI's performance in this race on lower difficulty levels did jump significantly since I last played the game, but something still felt off - particularly seeing the AI sporadically match times like 2:08.xxx touted by top tuners like Praiano and other users (who I assumed would be running this race on hard). I think I'd seen either in this thread or reddit someone mention that Easy to Hard had about a 10 second variation in finish times, with easy usually finishing around 27:35 and 27:25 - but I often found the AI finishing under 27:30 even on easy.
After about 25 runs beating the AI with CRB consistently I decided I was ready to try the race on a harder difficulty, so I switched the difficulty to Normal. But on this attempt, I was surprised to find the 2nd place AI log probably the slowest time I'd seen on this race ever (27:38.454). Wanting to understand more, I logged the times from my next 18 attempts of this race, consisting of 6 sets of attempts on different difficulty levels with the same cars/tunes. (I know I'm hardly the fastest in this race, please don't judge :3)
Here's the full data:
And here's a summary of the median and mean times:
I'm not an expert on how the AI works in this game by any means, but I feel like I'm seeing barely any difference in the AI's pace for this race at different difficulty settings - especially compared to other races of this length where the AI is more than a minute slower on easy than on hard. But I can't ignore how much more consistent the AI was on hard, with no attempts where the FD3S pit twice. It's almost like the AI on this event always operates on hard pace but still operates with weaker decision making skills at lower levels - leading to more variation and mistakes? I'd be really keen to hear from anyone else.
I recently got back into GT7 after a fairly long hiatus, largely inspired to after lurking this thread and seeing the different setups everyone used to beat this race. Since I'd been out for a while, I set my game difficulty to Easy to get back in the rhythm of this race. I was already quite surprised by how quick the AI was in this race on Easy. I had learned from scrolling through this thread that the AI's performance in this race on lower difficulty levels did jump significantly since I last played the game, but something still felt off - particularly seeing the AI sporadically match times like 2:08.xxx touted by top tuners like Praiano and other users (who I assumed would be running this race on hard). I think I'd seen either in this thread or reddit someone mention that Easy to Hard had about a 10 second variation in finish times, with easy usually finishing around 27:35 and 27:25 - but I often found the AI finishing under 27:30 even on easy.
After about 25 runs beating the AI with CRB consistently I decided I was ready to try the race on a harder difficulty, so I switched the difficulty to Normal. But on this attempt, I was surprised to find the 2nd place AI log probably the slowest time I'd seen on this race ever (27:38.454). Wanting to understand more, I logged the times from my next 18 attempts of this race, consisting of 6 sets of attempts on different difficulty levels with the same cars/tunes. (I know I'm hardly the fastest in this race, please don't judge :3)
Here's the full data:
And here's a summary of the median and mean times:
I'm not an expert on how the AI works in this game by any means, but I feel like I'm seeing barely any difference in the AI's pace for this race at different difficulty settings - especially compared to other races of this length where the AI is more than a minute slower on easy than on hard. But I can't ignore how much more consistent the AI was on hard, with no attempts where the FD3S pit twice. It's almost like the AI on this event always operates on hard pace but still operates with weaker decision making skills at lower levels - leading to more variation and mistakes? I'd be really keen to hear from anyone else.
- 11,327
- Marin County
I find that the FD3S pits twice if you are running near it towards the end, regardless of the difficulty. The AI kind of latches onto you if you are nearby and will significantly increase it's pace to stay close to you, and when that happens the RX7 burns through its tires very quickly. The races that the FD3S pit twice, per your data, were the ones where your finishing time was very close to the RX-7's typical finishing time, implying it was near you and pushing towards the end. The ones where you were further ahead, it wasn't racing you.
I did a race today - Custom race 90 mins at La Sarthe (no chicane) with cars from my garage between 550-600PP. 0X tire wear, 1x fuel usage. Professional difficulty. I used my SLR @ 600PP. On SH tires I was running about 4:00 laps consistently and the 2nd place Ferrari F430 (599PP) was about 8 seconds back (amazingly with a higher top speed on the straight than the 620hp SLR, somehow). The Ferrari was running consistently in the 4:01 range or thereabouts per the replay. When I pit for fuel I decided to swap for SS tires and this put my lap times down to around 3:52. What do you know, as soon as I did that, the Ferrari also started running around 3:53 on the same tire. Keep in mind every other car in the field was 1 or even 2 laps down, and somehow the Ferrari was stuck to me like glue. The AI is many things, but consistent is not one of them.
I did a race today - Custom race 90 mins at La Sarthe (no chicane) with cars from my garage between 550-600PP. 0X tire wear, 1x fuel usage. Professional difficulty. I used my SLR @ 600PP. On SH tires I was running about 4:00 laps consistently and the 2nd place Ferrari F430 (599PP) was about 8 seconds back (amazingly with a higher top speed on the straight than the 620hp SLR, somehow). The Ferrari was running consistently in the 4:01 range or thereabouts per the replay. When I pit for fuel I decided to swap for SS tires and this put my lap times down to around 3:52. What do you know, as soon as I did that, the Ferrari also started running around 3:53 on the same tire. Keep in mind every other car in the field was 1 or even 2 laps down, and somehow the Ferrari was stuck to me like glue. The AI is many things, but consistent is not one of them.
Last edited:
- 1,513
- Germany
None of us is and everything is based on speculation and observation.
Like @Eunos_Cosmo said just before:
If you are very close to any AI car, this car will drive "more competitive" and it will result in a propable extra pit stop.
Then, as everyone should be aware off:
AI is absolutely not consistent in their own pace.
6 attempts per difficulty is barely enough to really see their delta of high and low times.
You could propable collect additional data from this thread to make a more informed statistic about it:
AI times by screenshots within this thread since februrary 23 (difficulty setting undefined) until Jan 17, 2024
27m + xs
05 - 1
10 - 2
12 - 2
15 - 2
16 - 1
17 - 1
18 - 2
19 - 2
20 - 4
21 - 4
22 - 1
23 - 2
24 - 2
25 - 7
26 - 4
27 - 9
28 - 8
29 - 5
30 - 4
31 - 9
32 - 6
33 - 12
34 - 5
35 - 8
36 - 8
37 - 5
38 - 6
39 - 7
40 - 4
41 - 3
42 - 2
43 - 3
46 - 2
48 - 1
49 - 1
53 - 1
Last edited:
