Zwifters nearby list ignores direction in Scotland

I noticed today in Scotland that the nearby riders list really shows riders nearby whether they are heading in the same direction as you or going the other way. The values shown seem correct so riders on the other side of a loop might be several minutes and kilometres behind, and they move from ahead of you to behind you once the avatars have met.

This seems like a bug to me, and even if it is so by design, it seems like a bad idea for a number of reasons: it’s different from all the other worlds, in events it makes it more difficult to see what the time gaps are, and in general when the world is more populated, it makes the list difficult to follow in general because the names move around too quickly.

It’s been noted. There are many other places across the world’s that do this.

General consensus is that there’s some strange issue with the code that triggers this to happen in random places.

1 Like

I spotted this in Glasgow a few weeks ago, and several people here said they’d seen it in other maps over the years. I agree, it’s distracting and not ideal.

Was this on one of the routes where you double-back over some of the same sections of road? This sounds like the same issue as when riding Tempus Fugit, where the riders are, proximity-wise, close to you, but route distance-wise not so much. Yet they are going the ‘same direction’ (on the route) as you are. I’m not sure Zwift can differentiate.

It happens in Glasgow. Is there a single route that includes both directions in Glasgow?

That’s the best explanation yet. Radio tower does it if i remember correctly. Although Alpe Du Zwift doesn’t.

1 Like

Parts of The Muckle Yin and Rolling Highlands routes cover the same stretch in both directions.

I just did the Muckle Yin–parts outside Glasgow go in opposite directions, but the route only includes Glasgow streets clockwise.

I originally noticed this after doing the ToZ Rolling Highlands stage, after which I diverted into Glasgow to sprint around. So I was off-route…but the route I was on before that didn’t have Glasgow in it at all. So my guess is that it’s not related to the route.

Hard to say without seeing how all of the coding works, of course. But, Glasgow is a pretty small piece of the Scotland map, so if there aren’t a lot of people riding might you see riders going the ‘other way’ in the list, even though they are not actually on the Glasgow roads?

Also: Is it possible to do a u-turn and ride Glasgow in reverse?

Muckle Yin was where I noticed this as well, plus freeriding afterwards on the flatter parts of the big loop, also on fairly straight sections. Haven’t noticed this in other worlds, but then again I haven’t been to the radio tower in ages since getting all the relevant badges…

The bug I’m more familiar with is the one where the system gets confused about who is behind and ahead of you when the road makes several 180-degree turns, such as on the little twisty climbs to the Italian whatever in Watopia, but I don’t think that one picks riders that are going the other way.

1 Like

I thought that was the general consensus about Zwift in general. I’ll get me coat.

4 Likes

Not something i could offer seen opinion on. I’ve zero knowledge in programming and i doubt anybody here is suitably entitled to find fault either.

If anybody was that good they’d have written their own version of Zwift.

This was happening on Innsbruck back in 2018.

Vincent said it was a bug.

But that was 4 months before my post Improve "Riders Nearby" list, so if it was a bug they didn’t seem in a hurry to fix it.

That it’s still happening more than 4 years later suggests they can’t fix it, or don’t think it’s important.

The riders nearby list needs a rework in all worlds.

By biggest annoyance is every time you go through junction you lose riders. I can understand that in free rides as you dont know which way a rider went but for events it would be nice to not have this issue.

1 Like

Surrey Hills is the one place that comes to mind that also had this problem but Zwift fixed it (at least on AppleTV).

It wasn’t people on the far side of the lap, no. You’d see the name ‘ahead’ of you on the list, someone doing far more w/kg than you, yet you were closing rapidly on them. Then you’d see their avatar pass you in the opposite direction, and their name on the list would move from ahead of you to behind you and continue to fall farther behind.

And yes, you can u-turn in Glasgow (in fact that short punchy KQOM has a KQOM in the opposite direction…although the sprint does not). But the riders I was seeing in the opposite direction were as described. They’d u-turned, but were on my list even though they were going in the opposite direction.

1 Like

Interesting. Maybe because there is no ‘reverse’ route Zwift still recognizes them as being on the same route (and direction) even after a u-turn.

My biggest annoyance are the fake “time” gaps in the first place. (If it’s really a distance, why not display it as a distance.) A side effect newly highlighted by the PR bots is that especially on uphill segments it’s impossible to tell how much faster or slower you actually are compared to your PR until you hit the finish line, it could for instance keep saying that you are 3 seconds ahead of the bot and then the actual time is 10 seconds faster or whatever.

Another thing zwift need to improve. I use sauce for most my races now partly for that reason the time gaps are far more accurate.

Talking of PRs the volcano kom is the most annoying it does not bring up kom info until half way up the hill most annoying when shooting for a PR.

2 Likes

I guess that could be happening, yeah. Zwift thinks they’re on the same route, just tearing around in reverse :slight_smile:

You mentioned Tempus Fugit before–that route also has one road segment that I don’t think exists on any official route in one particular direction. The right side of the top loop (where the arrow is pointing). I don’t think any route has that road segment running south to north. TF has you loop around clockwise and ride it south to north. Wonder if that might have something to do with it, or just a tenuous connection?

fugit loop