One “coming soon” feature from today’s Zwift announcement that has Zwifters talking is the “Climb Portal” – a new way to experience new climbs on Zwift.
While we don’t have detailed screenshots or gameplay videos, I’ve spoken to sources within Zwift to learn more about this exciting feature. Here’s what I know, along with some speculation and fun ideas!
How Climb Portals Will Work
Two portals will be available when the feature launches: one in Watopia (near the Volcano), the other in France (near the Castle).
You can access the portal as a route option from the home screen or navigate to the portal yourself while riding on the map. Once you begin your portal climb you’ll get a short warmup on flat ground before heading uphill.
A Curated Climb Collection
Zwift says the famous Col du Tourmalet and Col d’Aspin will be featured climbs in the portal, “both storied for their prominence in some of history’s most exciting bike races.”
Other climbs will be available as well, but it sounds like just one climb will be available in each portal (so two climbs total) at any given time. Zwift says this “will give the Zwift community the chance to enjoy the segments together.”
So Zwift is developing a curated library of famous climbs, then rotating them as the featured portal climbs on some sort of defined schedule.
About the Visuals
Zwift is doing something very different with the visuals for these climbs… but I’m not altogether sure what that is! Zwift’s Climb Portal FAQ includes this:
Q: What will the environment inside the portal look like? Will the climbs look like what they look like in real life?
A: Zwift has created a completely new visual experience that gives you the chance to focus on the climb, and enjoy the company of other Zwifters. We can’t wait for you to see it!
Clearly, Zwift are choosing their words carefully:
- New visual experience
- Gives you the chance to focus on the climb
- Enjoy the company of other Zwifters
Because Zwift’s current methods of creating new worlds/maps/roads seems to be quite time-intensive, my guess is that portal climbs won’t have the custom, detailed environmental visuals we’re used to seeing on climbs like Alpe du Zwift or the Epic KOM. While the climbs will be GPS-accurate, I bet the surrounding environment will be programmatically generated so Zwift’s art team doesn’t have to place each building, rock, tree, etc.
Zwift calls it a “new visual experience”, which may indicate something even more radical, like a Mario Cart’s Rainbow Roads!

Launch Date
Zwift is saying “this summer.” My guess is we’ll see Climb Portals roll out between June and August.
More Tidbits, Ideas, and Questions
The portal segments will not be available for club events or races, probably because of the way Zwift wants to curate the schedule.
There will certainly be some sort of in-game leaderboard for the featured portal climbs. But it would be really fun to have a web and/or Companion-based leaderboard for the featured climbs that any Zwifter could access.
Giving the current leader a virtual jersey would be a nice touch, and I assume that will happen.
How will this work on Strava? Will all the portal climbs originate from the same place in Watopia or France, then “spider” off from there if you were to overlay multiple climb activities on one Strava map? Or will entering the portal start a new activity, which then records using the IRL climb’s GPS coordinates?
Will we be able to turn around at the top and descend the climb, or is this one way? Are there turnarounds at the top and bottom so you can do hill repeats (vEveresting, anyone)?
I’d love to be able to spawn my own RoboPacer at the start of the climb, at a defined w/kg. Chasing a bot on screen is more fun than watching my wattage and trying to hold target power, so if I could just choose Col du Tourmalet, set my bot at 3.5w/kg, then start the climb that would be a fun challenge!
Questions or Comments?
Those are the details I know, and the questions I have. But what about you?
What climb(s) would you love to see on the Climb Portal? Got questions or other comments? Share below!
Sounds like they’re a little worried about RGT’s Magic Roads. A little healthy competition is a good thing!
Sounds a lot like Magic Roads, which I have used extensively. In about 5 minutes, I get a reasonably accurate topography with stock graphics (but with some “town signs”). This will be interesting.
I do hope they’re going all in on some easy to rollout fantasy landscape. It does sound like a copout from developing climbs like ADZ which was their best work to date, but this could mean we finally start getting new epic climbs faster which is personally the only thing I think the game needs route wise now. New maps just aren’t that exciting beyond the first ride, but epic climbs like ADZ are always a challenge and timeless. To be honest, some sort of fantasy landscape is going to visually way more interesting than anything else they could do… Read more »
Well, as bummed out I am about the rotation instead of permanent climbs, realistically that was bound to be the case. It enables Zwift to have a huge collection of mountains down the road (especially if these are relatively quickly implemented with the AI-generated scenery), not water down the quality of roads too much and it allows the art team to focus on maps that are not just for the handful of goats among us. Let’s not forget, in the grand scheme of things, how vastly unpopular our two huge climbs are compared to tempus fugit for example. And I… Read more »
Sa Calobra, Stelvio, Mt Teide south, Angliru for starters
I prefer to start with the famous Dutch and Belgium climbs from the classics … ,-)
Not really sure they are climbs as such more of a shorter steep rise or ridge not a long climb over several kilometres which the European climbs usually are. Woolonga Hill may be another.
Try to do a route there, and surely you will see them as climbs after that 😉
Stelvio would be a great addition, an awesome climb!!
3.5 w/kg?! Showoff… 🤣
Hahaha … now we know why Eric does this “12 week build me up” plan 😉🤣
is 3.5w/kg good or bad?!
Low end Cat B.
With a generic road, these long climbs will be torturous as there will be nothing to take your mind off the climb. I find Ventop to be really boring to climb as it doesn’t have anything interesting to look at and there are no turn by turn graphics like on the Alp.
I guess we’ll see.
Ventoux is a boring climb IRL once you exit the forest.
Google Street View and other entities like it, plus individuals’ photos uploaded to any number of mapping and photo services, give us a pretty good view of the landscapes for many/most/all of the famous climbs in the modern peloton. I wonder if Zwift’s figured out a way to use image-processing techniques (think machine learning), perhaps combined with high-resolution land-use data, to quickly generate virtual landscapes from photos.
That’s my hope. The ability to generate 3D world data from real world video/image/sensor data has progressed to the point that they, in theory, could replicate the real-world route remarkably accurately in -game. Obviously a lot of folks wouldn’t have the graphics processing ability to see it in full glory.
Sounds good. Magic roads on RGT i found disappointing but if the visuals are cool on these portal climbs this will be big. Segments instead of worlds means we can get on the climbs in the big races. Its the visuals that will make or break this i think. We need arches and crowds etc but if it feels samey and we see rock type 6 over and over again it will suck the fun away.
Yeah, it takes some imagination to enjoy Magic Roads. I use it for roads I know pretty well, and I know where I am from turns, topography and distance. It’s done in 5 minutes, and I can ride a familiar road even when the weather sucks. Guess I am not into the scenery as much (chewing stem too much–LOL).
interesting feature … hopefully not only for long climbs with 15+ km … for example “La Super Planche des Belles Filles” is around 7km long but hard & steep enough to be a big challenge for most of us
It’s going to be dino suits, isn’t it? (Just kidding).
Couple of thoughts — ‘Focus on the climb’ may be more above emphasising all the groovy new personal metrics and pacing data. Think more HUD type thing.
‘Enjoy the company’ could be anything from staccato video feeds of other people suffering to better chat or audio to better tracking of who’s where on the hill.
Doesn’t all need to be 3D world graphics — in fact it feels on some level like they are moving away from that for these portal zones
I am predicting around the 5th July 2023 for the portals to go live.. This year in the tour on 5th July travels from Tarbes to Le Cambasque above Cauterets. The finish climb is 16 kilometres long and averaging 5.4%. The riders tackle the Col du Tourmalet and Col d’Aspin on the run-in to Cauterets.
Would make sense that these 2 climbs being in the tour what better way to generate riders to do it then making it coincide
Good one. I hope you’re right!
That Cambasque climb is deceptive as the road up to Cauterets is really easy, then the final 5km to Cambasque is about 10% average. Quite a pretty view at the top, even if not as notable as Pont d’Espagne. I wonder if these climbs are an up and down thing or whether you can tag them together. For instance, it seems strange to do the Aspin and not the Tourmalet at the same time, or the Tourmalet and Luz Ardiden from the other side. Similarly you have the likes of the Sella Ronda in the Dolomites, which is lovely but… Read more »
Good call, could also be 29th of July, the women’s “TdF avec Zwift” does both climbs that day with a mountain top finish on the Tourmalet.
My guess is that they are finally incorporating stuff from the Gravel Mountain (https://zwiftinsider.com/gravel-mountain/) stuff they did a while ago, probably for the visuals. Also my interpretation of the official press release presentation is that multiple climbs will be offered at launch, with two of them being the aforementioned climbs. The specific wording in the press release is “At launch, there will be a rotation of several climbs available to Zwifters who ride through the portal, including two storied Alpine climbs…” Last point on your question on turnarounds: the release says “you’ll get a short warm-up on flat ground before… Read more »
I just hope this doesn’t mean Zwift will stop investing in map expansion, and in particular new “traditional” virtual climbs. Neither Makuri nor Scotland contain any new major climbs (yet), so that’s a bit worrisome. It would be really nice to have a 2nd way up to Chalet Reynard in the France map, and also to have a 400m (or more) vertical climb introduced to the Makuri map…
Looking forward to the launch. Endless possibilities if Zwift were to select popular climbs across all continents. Not sure if Zwift will have internal limits as to maximum distance/average time to complete the climbs, for example, Mt. Lemmon at almost 30 miles.
Both climbs happen to feature in stage 6 of the TdF this summer on July 6th, the first true mountain stage of the tour. I suspect the release will be sometime just before this IRL stage.
I hope they bring in a version of RGT magic roads, i.e. being able to upload your own local KOM route as a GPX file.
Don’t hold your breath!
I welcome this addition but I’m amazed at the enthusiasm for replicating long alpine or Pyrenean climbs. I’ve done most of the classics and you have to do them in real life. Ventoux is great IRL but it’s so boring in game. Remember the pros climb it in a train with peel offs. It will be the same with the tourmalet. Epic IRL but really in game with no train to follow. The Flanders climbs are great IRL but would be over too quickly in game. Zwift can’t provide replacements for RL, it is at its best as a platform… Read more »
I don’t think anyone views these climbs as suitable replacements for the real thing. No one would pass up an opportunity to do the real thing because they already did it on Zwift.
The fact that they are based on real life climbs isn’t really that important, but it definitely makes it easier for Zwift to roll out content if they have something to start off from. Plus marketing.
My enthusiasm is simply having more long climbs in the game.
This sounds really awesome and I can’t wait for it. But, Eric, your last paragraph confused me … is there a way to set your own RoboPacer?? Sorry if that’s a dumb question – I’m still fairly new on Zwift.
There is not a way to set your own RoboPacer – although you can sort of do that with HoloReplay (https://zwiftinsider.com/holoreplay/).
I was just suggesting the RoboPacer idea as something cool Zwift could implement in the Climb Portal.
Agreed, that would be a fun twist on setting a workout!
Sounds like a very fun feature that would totally fit with the more futuristic concept
Catching up to Computrainer capabilities; I forget what CT’s software was called. Anyway, a customizable pacer would be a great feature!
Would they be using video routes like they do in Rouvy and Elite systems.
Ie, someone videos the route and it plays to you at the speed you pedal at.
Having others on the route at same time is great.
I don’t understand this Climp Portal madness. For me, it’s completely unnecessary. If I have to climb without a 3D route, I prefer to do it for free from the Wahoo Roam Counter level :). I’m uploading a GPX route that I want and driving.
They will probably include the other side of the Aspin
“Enjoy the company of other Zwifters”. Based on recent events, I would say discord implementation for each climb.
“vEveresting, anyone”
well yes! =)
With Zwifts vagueness on the visual component of these climbs I immediately thought of everything that the latest AI is capable of. Maybe riders would see differences on multiple ascents. I’m guessing this would have to happen from the server side.
I would like Sa Calobra Mallorca/Spain, ikonic and popular, nice and har climb, 9,4km 7,5% , 670 meters elevation
https://www.cyclinglocations.com/sa-calobra-climb-mallorca/
Yay! More climbs!
Before Zwift I sat in my trainer staring at a blank wall and shifting gears occasionally in an attempt to mimic terrain. Indoor training was boring AF, miserable, and ineffective. Programs (games really) like Zwift and components (controllers) from Wahoo, Saris, etc. have improved indoor cycling so much that I actually look forward to winter. Consequently, I don’t really care what kind of scenery Zwift chooses to generate along its routes as long as they change things up a bit to avoid repetition. The long climbs have their place, and I’m glad to see Zwift adding more of them, but… Read more »
The ability to ride a climb with a pacer at a given w or w/kg is something I used to love to do on the old computrainer. It would be nice to have that in zwift
Haleakala and Muana Kea would be cool!
https://youtu.be/K4uJyiFZ_yc in progress