Zwift Climb Portal [2023]

not sure that answers why there are only 2 climbs available at any single point

Because “many people find it more fun to be doing something among a populated road of other cyclists”

Same reason why they provide the gradient scaling to increase participation. Same reason for only 3 worlds available at a time.

Of course many people disagree with the reasoning which is fine, I’m not trying to defend it.

Question: How do I get access to the Climb Portal. I was on Zwift this morning, tried to find the Portal in the different menus, nothing. Is there something I did not understand about the Climb Portal ?
Thx

The Climb Portal Experience

Both Watopia and France have Climb Portals. These can be accessed from the routes menu of the Watopia or France maps, or at the bottom of the homescreen, where you’ll see an option for the month’s featured climb (located in France) and the day’s featured climb (located in Watopia).

Thank you Ian
Since the last update, impossible to scale, so everything is now so small on my tv screen.
Anyway now I know.

2 Likes

I rode the col de la couillole. The col had sections of 18% and even 25%. Afterwards I wanted to look up the proportion of the steep parts, and it turned out that IRL there were no such steep parts anywhere. I thought it was a fun climb to ride, don’t get me wrong. But isn’t the purpose of the ClimbPortal to simulate a climb as it is in real life? If it is not the actual profile, you can just as well randomly generate a climb and give it any name. But why pretend it’s a certain climb if the profile is wrong?

I suspect they just took some elevation data quickly. I can’t imagine they had someone riding the real climbs as well as the portal climb to check each one. (someone will come along to say they actually did I suppose).

1 Like

yes that’s what it looks like

1 Like

Thank you very much for the clarification :pray::grin:

BĂČĂłX

This is where Fulgaz is a lot better, the gradients are usually far more accurate. The people submitting the climbs put a lot of effort into getting them accurate. Plus you have the real scenery -better than looking a bland, empty void.

Today I rode Passo Pennes (Penserjoch) which I’ve also ridden in real life. Very well done and no sudden unexpected spikes of 25% gradient. It was just like riding the real road.

1 Like

I second this! I’m training for the Mt Washington hill climb and it would be amazing to have it in the climb portal

1 Like

Are these climbs even vaguely accurate? I looked at the one for a few days time, trolls ladder, and the zwift version looked nothing like the veloviewer segment
similar for a few others.

I’d have thought replicating an existing segment was the easiest way of doing it?


if zwift has one infuriating corporate cultural habit, it is zwiftifying things in a way that makes them less good


2 Likes

Maybe that’s why they haven’t bothered publishing maps or graphics beyond 80’s block colors that match the gradient?

Personally, i don’t mind the colourful cartoony approach. Its ultra fast for adding new climbs etc. But I don’t understand why they aren’t vaguely accurate 3d representations when these are available. Maybe some climbs don’t have accurate data available?

1 Like

Do we have a portal climb request thread?

Apologies if we do, I’d like to suggest Road To Hell from Denbigh in North Wales to Llyn Brenig.

~6.9 miles and ~1209 feet elevation gain. Lots of gradient changes, including a short descent after the nastiest part that hits ~20% (veloviewer details are close but not accurate IMO).

Well worth doing in real life too.

~33mins at ~3.5W/Kg (I can only dream of matching that these days).

1 Like

I assume they’re being built from topographic data. If the resolution isn’t good enough to clearly define the roadway separately from the hillside you can get all kinds of crazy elevation changes as the overlaying the road route on low res data will have it snaking literally up and down the hill.

For example, here’s a shaded relief image of Pla d’Adet:

Compared to a section of Trollstigen:

Now you might have to license high quality data for these climbs which I wouldn’t fault them for avoiding, personally, considering the unfortunately low participation that the climb portal sees.

2 Likes

Which is probably because they are sometimes inaccurate - and with generic bland graphics very boring. Even a linked video of the real thing like Rouvy/Fulgaz would do is better - plus Fulgaz had very accurate elevation details.

Personally I’m after more climbs to link up with ADZ as in real life, the Galibier and Sarenne climbs along with the balcony road between Sarenne and real life Huez village. Also the other different sides of Alpe d’huez.

And then the loop at top of ADZ becomes a small village as in real life.

2 Likes

I don’t know the zwift spaghetti code but RGT showed it could be done with their magic roads. We don’t need animals and mythical beasts running around or logging crews trying to crush the riders. Given the advancements over the last 10-15 years since Jarvis and Watopia were first pitched, you’d think it be possible to have generic landscape scenes. Maybe rushing out a new climb, but after it’s been public for 6m-1yr
seems like the easiest way to provide the opportunity. Also given the Nintendo graphics, it’s not like they are Forza, COD or GTA.

2 Likes

Just asked ChatGPT

Model Resolution Vertical Accuracy Usage Conditions
Copernicus DEM 30 m / 90 m ±4 m Free to use
SRTM DEM 30 m / 90 m ±16 m Free to use
ASTER GDEM 30 m ±20 m Free to use
1 Like