maybe that is the solution then. Pacer bots set pace not power.
pace set as the average speed for the route they’re on.
Not always, because the robopacers can zoom away when they get the benefit of drafting a lot of fast riders. Setting a maximum speed they can reach would prevent that.
Constance can go from 41km/h to 48, even 50km/h very quickly in the area just after the banner on Tempus fugit (where the sprint segment is), that’s a primarily flat section. And will maintain that speed for about 1km (sometimes more). But you end up with multiple surges along the route.
Capping the maximum speed to say 43-44km/h would solve that. Robopacer will just not exceed that speed. That stops the surging and keeps the pace more steady.
OK I think we are just using different terminology. In the context of Zwift RoboPacers, pace means power. Maybe that is the wrong way of expressing it but that’s how it is. You are talking about speed. Let’s talk about speed and power. With dynamic “pacing” power varies by terrain and is perfectly constant on flat roads, while speed varies based on pack dynamics and other stuff.
1 Like
Doing the same power?
1 Like
I already explained where this was happening, so you can watch the Constance group and see for yourself.
So is this a yes or a no?
It’s a yes Gerrie.
I don’t have the big issue with the current pack dynamics that Chris and Paul seem to. My experience is that the big jumps in pace are from movement within the pack, and when you get dropped is usually from riders in front of you in the pack dropping wheels when you aren’t really paying attention.
I think the pace partner groups with current pack dynamics are fine - though I still think they were better with steady pacing. As Chris said - the original problem that people were leaving the pace bots behind on the climb is still there.
They were 1000% better with steady pacing. If you are light then just slow on the climb, the same can’t be said for a heavy rider.
I still don’t really see why both can’t co-exist
why can’t you have dynamic power groups and static power groups
if the group sizes are two small maybe alternate between them on different days or have them in different worlds
too much complexity and poor UI/UX for low information/new users?
You could argue that the static pace is actually easier for users to understand, as we have seen recent complaints about the bots constantly displaying different power on the homescreen and users don’t know that they are dynamic.
fair point!
in that case, just change between the two without any warning/explanation! there is a 50% chance people will get the one they want at any given time
One suggestion floated before would be to introduce a couple more RP rides that are instead structured as 24/7 group workouts – e.g. one at zone2, one at zone3. If you want to tweak the effort, then adjust your workout bias %
There are two aspects that result in variability within the pace partners, one is the dynamic pacing of the robo pacer itself over varied terrain, and the second is the amount the robo pacer is drafting. I think it would be worth revisiting putting the pace partners on TT bikes to reduce variability due to the robo pacer drafting. That will not result in a fully steady pace, but it will be more predictable than it is today - we at least won’t get the complaints about people pulling the pacers.
The fact that they always have one pacer on a flatter route helps a lot with the dynamic pacing aspect of it. Folks who don’t want dynamic pacing pick the flatter routes, and I expect if they were on TT bikes the flatter routes would be a reasonable steady pace even with the lesser attended bots.
The issue with putting the pacers on TT bikes is the power you will need to keep up with the pacer while in the draft will be significantly less than what the pacer itself is doing on the TT bike with no drafting benefit, so there’s some UX issue about explaining to people what your actual effort will be for any given pacer.
This varying pace even happens on the flat routes, but it’s to do with the way the robopacers can zoom ahead with the draft when there are lines of other riders. I said where that happens earlier, it’s easy to go and watch and see it.
The varying power output is never going to be changed so folks can forget about asking for that, we’ve seen the answer Zwift gave.
What would assist is giving upper speed limits thst the robopacer won’t exceed. This will prevent the group being stretched out or split apart so easily.
Well, again, putting them on TT bikes would help there.
Probably would make dropping the pacer a bigger problem, to the extent that it is a problem. Back to the fact that if you are not tracking the pacer and doing what they do, you may be choosing to ride away from it, and you might have to drop the pace a lot to not do that. People riding with pace partners without actually riding with the pace partner isn’t solved by these proposals unless the rider uses a TT bike as well.
Not sure why that would be a bigger issue if they are on a TT bike once you know which pacer is the one that suits your desired power.
The issue with the TT bike that I can see is trying to figure out the advertised pace of the rider who is in the pacer’s draft while the pacer gets no draft no matter the size of the group. Maybe if they advertised the steady state power you’d need to do in the pacer’s draft rather than the pacer’s power?
Because if you are in the draft and the pacer is not, you will not be doing the advertised power of the pacer. The pacer will be grinding away while you do much less. That’s the same problem as any TT bike rider joining a group ride and we all know what happens there. They have to work much harder to stay with the group. If Maria on a TT bike is doing 2.2 and you are in the draft, you’ll be doing ~1.9 or you will ride away from her. Probably less of an issue with Constance where there isn’t a big pack.
But that is exactly what I’m saying. The issue is how Zwift messages the average power you will need to ride with this bot who is on a TT bike assuming you are not on a TT bike.
I’m not sure what the best solution is to that, but for instance if they message the power you will need to keep up in the draft of the pacer rather than the pacer’s actual power (or something like that) then that might help.
Like instead of messaging “Coco is going at ‘x’ w/kg” they message “You will need to average ‘x’ watts on average to stay in Coco’s draft” or something like that.
This sounds like a suggestion that will be ignored by everyone who just does what they always do and pays no attention to the details. They see “Coco” and they expect it to be as it has been lately, and ignore all the numbers. If they were willing to pay attention to the pacer they wouldn’t have ridden away from it in any circumstance. I’m game for testing it though, bring it!