|
Breakaways / Peloton AI - your experiences ?
|
| diacos |
Posted on 04-07-2013 09:19
|
Under 23
Posts: 56
Joined: 09-08-2011
PCM$: 200.00
|
Ok, so i will play a stage with an outsider like Demare. Let's see how it works. |
| |
|
|
| diacos |
Posted on 04-07-2013 10:06
|
Under 23
Posts: 56
Joined: 09-08-2011
PCM$: 200.00
|
Ok, so i played a stage with FDJ, my sprinter being Demare. The break is catch with 5k to go by Lotto and Argos. At that moment, there were 3 sprint trains: Lotto, Argos and FDJ. Omega Pharma is nowhere, and Cav takes Kittel wheel. FDJ cannot keep up with the pace, and with 2k to go, even Lotto train fails. Kittel and Cav sprint with 1.2k to go, Cav wins, Kittel second, Greiepl third and Demare on fourth place.
That was amazing and i can say that's the best PCM so far. Thank you guys for the tip, and btw nice database Jes! |
| |
|
|
| roturn |
Posted on 04-07-2013 10:24
|
Team Manager
Posts: 22246
Joined: 24-11-2007
PCM$: 3900.00
|
It`s not realistic 100% that other teams give up on a possible stage win or good result at least when they just have the 3rd or 4th best sprinter.
But on the other hand you must expect that you have to do most work when playing with Froome/Cavendish/Contador/Greipel/Sagan etc. |
| |
|
|
| diacos |
Posted on 04-07-2013 11:37
|
Under 23
Posts: 56
Joined: 09-08-2011
PCM$: 200.00
|
Yeah, it's not realistic 100% but with 18k to go i usually form my train of 9 riders with omega pharma and it's kind of hard for AI to counter that with other trains. But when i do the same thing with FDJ, they make trains because mine is not that powerful and sometimes i cannot keep up with the pace. So it's very important how strong are your riders. |
| |
|
|
| cio93 |
Posted on 04-07-2013 11:54
|
World Champion
Posts: 10845
Joined: 29-10-2007
PCM$: 500.00
|
roturn wrote:
It`s not realistic 100% that other teams give up on a possible stage win or good result at least when they just have the 3rd or 4th best sprinter.
But on the other hand you must expect that you have to do most work when playing with Froome/Cavendish/Contador/Greipel/Sagan etc.
Most work, indeed. But if I don't relay straight away, the whole peloton just doesn't give a damn and rides 20km/h from km 0.
Which, subsequently, leads to a densely packed front, and this in combination with the solid rider model sometimes leads to gaps of 15 minutes even if I decide to put a rider on relay from the start simply because it takes him some 20k to get to the front?
Plus to elaborate on the "you're in yellow, you have to work":
Yeah, but what about the second placed rider, who, together with his whole team, just gives up on that place in the last mountain stage of a GT where they allow a rider 12 minutes back in the GC to get a gap of 20 minutes?
|
| |
|
|
| Jokke97 |
Posted on 13-07-2013 01:01
|
Amateur
Posts: 15
Joined: 23-06-2012
PCM$: 200.00
|
On the Columbian nation championship the break won with over 50 min, nearly an hour. I had just one rider (Quintana) so I coudnt really do anything about it. The whole peleton was just chillin all the way except the 20 last kilometers. Henao, betancur and uran attacked. Maybe a bug?  |
| |
|
|
| cio93 |
Posted on 13-07-2013 01:03
|
World Champion
Posts: 10845
Joined: 29-10-2007
PCM$: 500.00
|
Small country NC's are fucked up since basically ever.
Got a 1-2-3-4-5-7-8-10 in Norway yesterday when noone gave a fuck that I attacked with all 8 riders just because both other full teams had one guy up there as well.
|
| |
|
|
| Lachi |
Posted on 13-07-2013 01:28
|
Grand Tour Champion
Posts: 8516
Joined: 29-06-2007
PCM$: 200.00
|
I would not call it bug because the AI is just not designed to handle races without real teams and/or or small pelotons. It is more a problem of the DBs which allow such races.
I don't play such NC races in 3D since like 5 years. It's a total waste of time.
|
| |
|
|
| Ian Butler |
Posted on 13-07-2013 08:30
|
Tour de France Champion
Posts: 21854
Joined: 01-05-2012
PCM$: 400.00
|
I play them because it's easy to win them 
I had the breakaway bug but it's entirely gone for some reason... |
| |
|
|
| Ad Bot |
Posted on 27-11-2024 15:33
|
Bot Agent
Posts: Countless
Joined: 23.11.09
|
|
| IP: None |
|
|
| cio93 |
Posted on 13-07-2013 14:05
|
World Champion
Posts: 10845
Joined: 29-10-2007
PCM$: 500.00
|
Ian Butler wrote:
I had the breakaway bug but it's entirely gone for some reason...
Nope, it's even worse for me. No chance to win any one-day race from a sprint.
|
| |
|
|
| Ian Butler |
Posted on 13-07-2013 18:29
|
Tour de France Champion
Posts: 21854
Joined: 01-05-2012
PCM$: 400.00
|
That's so odd. It seems so random! |
| |
|
|
| Stromeon |
Posted on 17-07-2013 12:02
|
Classics Specialist
Posts: 3507
Joined: 06-10-2012
PCM$: 200.00
|
Played through a Vuelta last week, and while admittedly most of the flat stages aren't that flat, which may mean that some of the teams like Argos would be unwilling to chase because Kittel might not get over the hills; the break won every time except the final stage; when Bos beat Cav, Greipel and Kittel  .
Playing with Saxo Bank with Contador, although for the majority of the tour I was in third behind Valverde and Nibali before a huge attack on the Angliru dislodged both of them  . Therefore in third place overall and with no sprinter in my team I felt I shouldn't need to chase during the flat stages.
In the whole Vuelta, only 6 stages were not won by breakaways, 5 of which were mountains where I buried my team to bring the breaks back.
Whilst it's nice that there's more variation in when the break gets caught - the 2nd stage, which was won by Valverde, the break was caught literally 500m from the line which was quite tense; the break just does seem to win too often.
What I notice quite a lot of the time is that the peloton brings the gap down to about 3 minutes with 30km to go, and then someone from the break attacks. The pace in the break suddenly accelerates, and they start relaying hard, whilst the peloton remain at the same speed. Therefore the break usually wins by about a minute, with a sprinting Cav/Greipel taking 4th/5th/6th (depending on how many people were in the break)
Sorry for that wall of text
|
| |
|
|
| BritPCMFan |
Posted on 17-07-2013 13:48
|
Stagiare
Posts: 245
Joined: 03-06-2013
PCM$: 200.00
|
I've had not a single problem.
Been playing on normal since I got the game about a week ago. Obviously, all testing on normal for obvious testing reasons.
I've done an entire GC TdF and multiple stages + half a Cav for green TdF runthrough.
Really, no problems with breaks. If I have wanted to bring it back and worked, usually it does. Once with OQS I tried to pull back a break on a very hilly stage and just had one guy manage to stay out till the line (although there was no timesplit by the line). If I don't work, then its really depends on what the other teams want to do. On my GC run through every sprint stage got pulled back by the sprinters team with me doing little work. Once I was in yellow, I pretty much got to decide what happened on lumpy no-sprint stages as they all left the work too me.
I think the main problem with break is potentially when its not a sprint stage and theres no clear GC leader, and probably don't defend their own positions enough. I had a break stay away on stage 3 with my GC run, but I was happy to let it go since I didn't want the yellow and its entirely fair to thing Saxo/Moviestar etc were of the same mind.
Personally, I think its a million times better and much truer to life. |
| |
|
|
| ShortsNL |
Posted on 17-07-2013 14:18
|
Breakaway Specialist
Posts: 898
Joined: 17-11-2011
PCM$: 200.00
|
Yesterday I did a TDF flat stage on my stream. The context:
-I wasn't a sprinter's favourite, playing with Vacansoleil
-I wasn't leading the GC.
A handful of guys formed the breakaway, with pretty weak riders. In such a case, you'd expect the sprinter's teams to take full responsibility to not let the gap grow too big and to catch the break in time. There are 5 sprinter's teams in the Tour after all.
They did chase, but at 10km the lead was still more than a minute and when the breakaway then attacked, 2 of them stayed away by only 10 meters, with the rest of the break blocking the sprinting peloton.
The result:
1. Breakaway rider.
2. Breakaway rider.
3. Van Poppel (only one of the two to not get blocked by the break)
4. Cavendish
5. Breakway rider.
6. Kittel
7-10 Breakaway riders.
The chances of a breakaway staying away on a completely flat sprinter's stage are pretty much <1% IRL. Therefore, this shouldn't have happened in-game.
I think that the AI starts chasing too late. If you have a sprinter's favourite then you might have been the one to pull it back in time but otherwise the peloton is too late with starting to chase.
You can see footage of what I am talking about on the Past Broadcasts page on my Twitch channel: www.twitch.tv/sho... |
| |
|
|
| cio93 |
Posted on 17-07-2013 14:36
|
World Champion
Posts: 10845
Joined: 29-10-2007
PCM$: 500.00
|
If it stays ahead, the break ALWAYS wins with 10 metres advantage to the sprinters.
I repeated a race 3 times, and after the same guy won every single time with a bike lenght to my sprinter, I just edited all his stats and stat limits to 50. Call that revenge 
Edited by cio93 on 17-07-2013 14:38
|
| |
|
|
| BritPCMFan |
Posted on 17-07-2013 14:59
|
Stagiare
Posts: 245
Joined: 03-06-2013
PCM$: 200.00
|
I've watched sprinters take control at 50% mark most of the time on sprint stages if I wasn't already doing so. I've even seen them take over at 50% to go when I was setting a very slow pace with a break 7'30 with 90k to go.
Also, lone stages might not be the best way to test the AI due to there being teams there with nothing to do. |
| |
|
|
| Ian Butler |
Posted on 17-07-2013 17:00
|
Tour de France Champion
Posts: 21854
Joined: 01-05-2012
PCM$: 400.00
|
cio93 wrote:
If it stays ahead, the break ALWAYS wins with 10 metres advantage to the sprinters.
I repeated a race 3 times, and after the same guy won every single time with a bike lenght to my sprinter, I just edited all his stats and stat limits to 50. Call that revenge
Poor guy I hope that wasn't Voigt  |
| |
|
|
| Lachi |
Posted on 20-07-2013 04:01
|
Grand Tour Champion
Posts: 8516
Joined: 29-06-2007
PCM$: 200.00
|
I am not sure if it belongs here but I have seen the weirdest race results ever.
I did not participate in the Tour Down Under and therefore simulated it. Jens Voigt won the race because on every hilly stage the peloton split into many small groups and Voigt always was in the first group.
For all these hilly stages, the time gaps were like on mountain stages. Small groups of about 5 riders arriving together and the peloton way behind.
Is this a common problem with simulation? If yes, it might be possible to win the Tour de France with Rodriguez if I simulate all hilly stage. |
| |
|
|
| Ramira |
Posted on 20-07-2013 04:47
|
Amateur
Posts: 16
Joined: 14-03-2008
PCM$: 200.00
|
Lachi wrote:Is this a common problem with simulation? If yes, it might be possible to win the Tour de France with Rodriguez if I simulate all hilly stage.
From what I noticed this always happens with quick sim hilly stages. Because of this I used to detailed sim the hilly stages that ended in a long flat bit (as they are so uninteresting) but now I'm forced to 3D every hilly stage or get unrealistic results, with gaps in stages that would lead to 40+ groups normally.
That being said it does seem the quick sim is a bit more realistic in 13, taking into account the profile of the stage to a certain extent. But that could just be my imagination. |
| |
|
|
| Daggen |
Posted on 29-07-2013 21:01
|
Sprinter
Posts: 1849
Joined: 08-07-2012
PCM$: 200.00
|
[rant]THE LATEST PATCH HASN'T DONE ANYTHING! I FIND IT EVEN WORSE THAN BEFORE! WHATEVER YOU DO DON'T DOWNLOAD IT!!![/rant] |
| |
|
