Home Blog Page 398

Zwift Config File Graphics Tweaks: 2019

Michael Hanney
-
21
Zwift Config File Graphics Tweaks: 2019

This post describes the variables in the Zwift config files and how they affect the game’s graphics performance. This was originally written in 2016, and while the variables are still the same, the example hacks are for systems that are quite old now. All the more reason to try and squeeze some extra performance out of them!

A few important notes:

  • Use at your own risk. This comes with no guarantees or support from Zwift HQ and variables are subject to change without notice.
  • Also, be aware that changing contents, and permissions, of configuration files may affect the ability for the application update process to complete its actions (and might result in injury or death – joke)
  • None of the config file changes below will increase or decrease the polygon count or detail level in textures. The level of detail is set by the game engine based on your system’s graphics capabilities and can not be overridden. (Watch Ep 47- Zwiftcast Special: How The Alpe du Zwift Was Made on YouTube to learn more about the graphics engine and how the adaptive level of detail works.)

If you run into problems, revert to the original configs (below). If you still have problems, do a clean install.

Table of Contents

Default Configuration file contents

On Windows these are located under: C:\Program Files (x86)\Zwift\data\configs

On Mac OS they are located under: ~/Library/Application Support/Zwift/data/configs

basic.txt
res 1024x576(0x)
sres 512x512
aniso 4
set gSSAO=0
set gFXAA=1
set gSunRays=0
set gHeadlight=0
set gFoliagePercent=0.5
set gSimpleReflections=1
set gLODBias=2		medium.txt
res 1280x720(0x)
sres 1024x1024
aniso 4
set gSSAO=0
set gFXAA=1
set gSunRays=0
set gHeadlight=0
set gFoliagePercent=0.5
set gSimpleReflections=1
set gLODBias=1
high.txt
res 1280x720(0x)
sres 1024x1024
aniso 4
set gSSAO=1
set gFXAA=1		ultra.txt
res 1920x1080(0x)
sres 2048x2048
aniso 4
set gSSAO=1
set gFXAA=1

How to make a custom config file

First, you can not unlock Ultra by changing the config files. The ultra profile is set by the graphics engine and can not be overridden. You can, however, turn some features on or off (if your GPU supports it) and increase or decrease the strength of certain effects such as generated scenery (foliage) and headlamp strength.

Follow these steps to make a custom config file.

  1. backup C:\Program Files (x86)\Zwift\data\configs on Windows, or on Mac OS ~/Library/Application Support/Zwift/data/configs/
  2. find out which profile Zwift gives you by dropping your log.txt on http://zwiftalizer.com
    On Windows, the logs are located under: HOMEPATH%\Documents\Zwift\Logs
    On Mac OS, the logs are located under: ~/Documents/Zwift/Logs
  3. As an Administrator, Edit the contents of the config file that matches the profile your system was given. For example, a Radeon 7870 gets the High profile, so we need to edit C:\Program Files (x86)\Zwift\data\configs\high.txt

Example Custom Configs

Basic overrides for increased frames per second (FPS) on low-end systems

Edit basic.txt.

  • Lower shadow resolution
  • Lower foliage percent
  • Higher level of detail bias (less detailed models).

res 1024x576(0x)
sres 256x256
aniso 4
set gSSAO=0
set gFXAA=0
set gSunRays=0
set gHeadlight=0
set gFoliagePercent=0.1
set gSimpleReflections=1
set gLODBias=3

High downgrade to increase FPS for GTX 750 Ti and 2015-16 Alienware Alpha (Nvidia GTX 860M)

Edit high.txt.

  • Lower shadow resolution (512×512)
  • Basic lighting & shadows (gSSAO off)
  • Sunrays Off
  • Headlight Off
  • 30% normal auto-generated scenery (gFoliagePercent 0.3)
  • Simple reflections (gSimpleReflections on)
  • Lower polygon count models (gLODBias=2)

res 1280x720(0x)
sres 512x512
aniso 4
set gSSAO=0
set gFXAA=1
set gSunRays=0
set gHeadlight=0
set gFoliagePercent=0.3
set gSimpleReflections=1
set gLODBias=2
set gShowFPS=1

Basic overrides for increased realism on late 2013 iMac 21.5” 1920×1080 with Intel Iris Pro Graphics 1536 MB VRAM

Intel Iris Pro is a fairly decent integrated GPU. It can handle some increased resolution and lighting effects.

Edit basic.txt.

  • High world graphics resolution
  • Low shadow resolution
  • High-quality lighting and shadows (gSSAO on)
  • Sun Rays On
  • Headlight Off
  • 1⁄2 x normal auto-generated scenery (gFoliagePercent 50%)
  • Higher polygon count models (gLODBias=1.5)
  • 30-40 FPS at 1080

res 1280x720(0x)
sres 512x512
set gSSAO=1
set gFXAA=1
set gSunRays=1
set gHeadlight=0
set gFoliagePercent=0.5
set gSimpleReflections=1
set gLODBias=1.5
set gShowFPS=1

Very High, almost Ultra, for overclocked Nvidia GeForce 750 Ti, 860M, 950 or similar

Edit basic.txt or high.txt depending on which profile Zwift gives you by default.

  • High world graphics resolution
  • Very high shadow resolution
  • High-quality lighting and shadows (gSSAO on)
  • High-quality reflections (default in High and Ultra, gSimpleReflections off)
  • Headlight On (default in High and Ultra)
  • 1.5 x normal auto-generated scenery (gFoliagePercent 150%)
  • High polygon count models (default in High and Ultra, gLODBias=0)

res 1280x720(0x)
sres 1536x1536
set gSSAO=1
set gFXAA=1
set gSimpleReflections=0
set gSunRays=1
set gHeadlight=1
set gFoliagePercent=1.5
set gLODBias=0
set gShowFPS=1

Ultra downgrade to High to increase FPS by lowering shadows, lighting effects, and foliage for Nvidia GeForce 960M (Mobile/Laptop) or similar.

Only do this if you are getting the Ultra profile, but would prefer higher FPS to image realism.

Edit ultra.txt.

  • Highest world graphics resolution (ultra)
  • High shadow resolution (not ultra)
  • NOT High-quality lighting and shadows (gSSAO off)
  • NOT High-quality reflections (gSimpleReflections on)
  • SunRays On (default in High and Ultra)
  • Headlight On (default in High and Ultra)
  • 1⁄2 normal auto-generated scenery (gFoliagePercent 50%)
  • Medium polygon count level of detail bias (gLODBias=1)

res 1920x1080(0x)
sres 1024x1024
set gSSAO=0
set gFXAA=1
set gSimpleReflections=1
set gSunRays=1
set gHeadlight=1
set gFoliagePercent=0.5
set gLODBias=1
set gShowFPS=1

Ultra override for medium to high-end Radeon and Nvidia cards that don’t get Ultra by default.

Edit basic.txt or high.txt depending on which profile Zwift gives you by default.

  • Highest world graphics resolution
  • Highest shadow resolution
  • High-quality lighting and shadows (gSSAO on)
  • High-quality reflections (default in High and Ultra, gSimpleReflections off)
  • Headlight On (default in High and Ultra)
  • 2 x normal auto-generated scenery (gFoliagePercent 200%)
  • High polygon count models (default in High and Ultra, gLODBias=0)

res 1920x1080(0x)
sres 2048x2048
set gSSAO=1
set gFXAA=1
set gSimpleReflections=0
set gSunRays=1
set gHeadlight=1
set gFoliagePercent=2
set gLODBias=0
set gShowFPS=1

It is debatable whether you will notice the difference of ultra res and sres on a 1080 display.

If you can’t tell, then revert the first two lines to:
res 1280x720(0x)
sres 1024x1024

Experiment with gFoliagePercent to increase auto-generated scenery.

Stop increasing it when your FPS drops below what you are comfortable viewing.

Ultra 1440 and 4K resolution option override

Do this if you have an ultra display but do not have the option in-game to select 1440 or 4K and are confident your GPU can run at the resolution you choose.

Edit prefs.config in %HOMEPATH%\Documents\Zwift\prefs.xml

Change the USER_RESOLUTION_PREF to your preferred display resolution

1440 wide quad high definition (WQHD – 4 times as many pixels as 720 HD)

<USER_RESOLUTION_PREF>2560x1440</USER_RESOLUTION_PREF>

2160 ultra high definition (4K – 4 times as many pixels as 1080 FHD)

<USER_RESOLUTION_PREF>3840×2160</USER_RESOLUTION_PREF>

Variable definitions

Textures resolution

Sets the graphics resolution/detail of the world.

Examples

basic: res 1024x576(0x)

medium, high: res 1280x720(0x)

ultra: res 1920x1080(0x)

Shadow resolution

Sets the resolution of the shadows generated by all objects.

Might be a bit more dependent on CPU than GPU.

Higher numbers increase realism but decrease FPS.

Examples

very low: sres 256x256

low: sres 512x512

high: sres 1024x1024

very high: sres 1536x1536

ultra: sres 2048x2048

Screen space ambient occlusion

Enables or disables higher quality light and shade – an object casting shade on itself, why the corners of a room appear darker, and that sort of thing. The effect is subtle.

Dependent on GPU.

On increases realism but decreases FPS.

Examples

On (High and Ultra): set gSSAO=1

Off: set gSSAO=0

Fast approximate anti-aliasing

Enables or disables smoothing of model edges and diagonal lines. The jaggies.

On increases realism but decreases FPS

Examples

On: set gFXAA=1

Off: set gFXAA=0

Frames per second indicator

Show the real-time FPS in top left corner

Examples

On: set gShowFPS=1

Off: set gShowFPS=0

Sun Rays

Enables or disables fancy sun rays – very cool effect in the dust and tree lined routes. Also affects moon rays.

On increases realism but decreases FPS.

Examples

On: set gSunRays=1

Off: set gSunRays=0

Headlight

Enable headlight on bike.

On increase realism but decrease FPS.

Examples

On: set gHeadlight=1

Off: set gHeadlight=0

Foliage Percent

Sets the level of auto-generated scenery – trees, grass, flowers, fruit, underwater kelp, etc.

This is a very effective way to manipulate the environment to suit your preferred FPS.

A number between 0 and 1 is normal.

Higher number increases scenery density / realism but decrease FPS.

Lower number decreases scenery density / realism but increases FPS.

Examples: * 0 = 0% * 0.25 = 25% * 0.5 = 50% * 0.75 = 75% * 1 – 100% * 5 = 500% – be very careful!

Examples

low: set gFoliagePercent=0.25

normal: set gFoliagePercent=1

extreme: set gFoliagePercent=5

Simple Reflections

Enables or disables simple reflections on water, glass, metal, carbon, cobbles, etc.

Off increases realism but decreases FPS.

On: set gSimpleReflections=1

Off: set gSimpleReflections=0

Level of detail bias

Not too sure of how this variable works.

Appears to draw models with fewer polygons when value is set to 2 or higher.

Bike riders look very low fi, and may even vanish when gLODBias=5.

Examples

very low detail: set gLODBias=2

low detail: set gLODBias=1

normal: set gLODBias=0

Heads Up Display

Enables or disables the heads up display overlays.

Set to 0 for making cool videos free from rider names and other info.

Examples

On (default): set g_bDrawHud=1

Off: set g_bDrawHud=0

ANISO

Improves the clarity and crispness of textured objects in game.

Appears to be deprecated.

Speed Tests: Zwift’s Fastest* Wheels for Climbing

Eric Schlange
-
38
Speed Tests: Zwift’s Fastest* Wheels for Climbing

Important: this post contains performance data which was invalidated in Zwift’s mid-March 2025 update. Learn about those changes and get updated performance data >

Your choice of wheelset influences your Zwift speed significantly since it impacts both weight and aerodynamics. And Zwift’s release of the Drop Shop has given us access to lots of wheelsets: 39 now, with more on the way!

Wheel choice for flat/mixed courses has already been discussed. What about climbing? When aerodynamics matter less and weight matters more, which wheelset should you choose?

We’ve been hard at work in our lab testing wheels. Here’s a post summarizing some of our findings.

Want a look at the complete data?
Check out our Zwift Speed Test Charts for Wheels >

*First, a Disclaimer

The “which bike frame or wheelset is fastest” question isn’t as easy to answer as you might think. Each piece of equipment has unique weight and aerodynamic (CdA) values, making them perform in ways that mirror outdoor physics. Lighter items perform better on climbs, while more aerodynamic items perform better on the flats. Differences between options will also be magnified or minimized by your own power levels, drafting abilities, and more.

Your best Zwift setup, more than ever, depends on your specific situation.

About the Results

Below you will find the ten fastest wheelsets for longer, steeper climbs. This would include big Zwift mountains like Alpe du Zwift, the Epic KOM, or Innsbruck’s KOM. This isn’t an exact science, but we’re pretty confident in our findings thus far.

This list is based on our recent speed tests of 100+ rides up Alpe du Zwift, a 12.2km, 8.5% monster of a climb. We tested all wheelsets using the same power levels, rider weight, and bike frame. This made it easy to spot which wheelsets performed best. We then ranked those wheelsets based on their climbing performance, using their flat/mixed course performance as a secondary factor when needed.

Your Winners

#1 & #2: Lightweight Meilenstein and Roval Alpinist CLX

Lightweight Meilenstein: Cannot be purchased, must be won via the Alpe prize wheel
Roval Alpinist CLX: 634,300 Drops to purchase, must be Level 38+

The Lightweight Meilenstein wheels are super-premium IRL, and their price reflects that. Luckily for us, we can get them free by just riding up the Alpe! Problem is, it may take one try… or 25. Nobody really knows how often the spinner lands on the Meilensteins, but they’re definitely a prize worth winning.

The new climbers from Roval were released in mid-2021, matching the Meilensteins climb time. Read more about the Alpinist CLX wheels >

Both wheelsets turn in essentially the same flat test times as well (we clocked the Lightweights at 0.5s faster over ~50 minutes), so your choice really comes down to preference and what’s in your garage.

#3: DT Swiss ARC 62

400,000 Drops to purchase, must be Level 40+

These hoops deliver strong climbing performance and solid aero performance, making them strong all-arounders. In fact, they rank at the top of our combined percentile rankings (see bottom chart here).

#4 & #5: ENVE SES 3.4 and FFWD RYOT55

ENVE 3.4: 200,000 Drops to purchase, must be Level 27+
FFWD RYOT55: 136,100 Drops to purchase, must be Level 18+

Just 6 seconds behind the Meilensteins, these wheels come at a low price and level unlock.

#6: CADEX 65

223,900 Drops to purchase, must be Level 33+

#7: GIANT SLR 0

165,100 Drops to purchase, must be Level 21+

Slightly lacking in aerodynamics compared to the top competition, they definitely have the light weight necessary to turn in a fast Alpe climb time. The SLR 0 wheels tested just a half second faster than the CADEX 65’s up the Alpe, but the CADEX 65’s are much faster on flat ground, so we gave them the edge.

#8 ZIPP 454

204,300 Drops to purchase, must be Level 30+

These climb at the same speed as the 353 NSW below, but are faster on the flats so they get the edge in our rankings. Because you have to get to the Alpe before you can climb it! A solid all-around wheel, the Zipp 454s rank second overall in our combined percentile rankings (see bottom chart here).

#9: Zipp 353 NSW

168,800 Drops to purchase, must be Level 24+

Zipp released the IRL 353 NSWs in late April 2021, billing them as the lightest tubeless wheelset they’ve ever made. They climb at the same speed as the 454’s, but are a bit slower on the flats.

#10: CADEX 36

216,900 Drops to purchase, must be Level 29+

These climb just a half-second slower than the highly ranked DT Swiss ARC 62 wheels, but the problem is they are much slower on flat/mixed terrain, so we placed them here. While they’re the third fastest wheelset on this list for pure climbs, their aero performance is so bad that it’s hard to recommend them over the other wheelsets above which are (at worst) just 1s slower up the Alpe!

What’s the Time Gap?

Only 7 seconds separate the Meilensteins from the Bora Ultra 50’s up the Alpe at 300 watts with a 75kg rider on a Zwift Aero frame.

Again, your best Zwift race setup depends on our specific situation, so do your research and choose wisely. Feel free to sift through our detailed charts!

Share Your Thoughts

Hopefully the mountain goats found this post useful. Got questions or comments? Post below!

Changelog

  • December 11, 2023: updated the list based on current performance data.
  • December 6, 2023: updated prices, level unlocks, and text to account for Zwift’s Drop Shop reorg.
  • September 19, 2021: changed the Meilensteins and Alpinist wheels to be tied for top slots based on fresh tests. Added CADEX 36 in #10 slot.
  • July 26, 2021: inserted the Roval Alpinist CLX wheels into the #1 slot.
  • Apr 22, 2021: inserted the new Zipp 353 NSWs into the 2nd slot.
  • Aug 11, 2020: inserted the new DT Swiss ARC 62s in the 3rd slot.
  • Nov 8, 2019: Moved Bora 50s down two slots to #6. Removed 7, 8, and 9 because their flat performance is so bad, and you can get better climbing wheelsets at lower levels and prices.

Important note: this post contains speed test results for Zwift frames or wheels. These results may change over time, and a bike's performance relative to others may also change. We don't always revise posts when performance rankings change, but we do keep current, master versions of our speed test results which are always available. See the frame charts, wheel charts, and Tron vs Top Performers for current performance data.

Speed Tests: Zwift’s Fastest* Wheelsets

Eric Schlange
-
28
Speed Tests: Zwift’s Fastest* Wheelsets

Speed Tests: Zwift’s Fastest* Wheelsets

Important: this post contains performance data which was invalidated in Zwift’s mid-March 2025 update. Learn about those changes and get updated performance data >

Your choice of wheelset influences your Zwift speed much more than your choice of bike frame. And Zwift’s release of the Drop Shop has given us access to a pile of new wheelsets: 28 in total!

The big question for racers is: which wheels are the best?

We’ve been hard at work in our lab testing wheels here’s a post summarizing some of our findings.

Want a look at the complete data?
Check out our Zwift Speed Test Charts for Wheels >

*First, a Disclaimer

The “which wheel is fastest” question isn’t as easy to answer as you might think. You see, each wheelset has unique weight and aerodynamic (CdA) values, making them perform in ways that mirror outdoor physics. Some wheels perform better on climbs, while others perform better on the flats. Differences between wheelsets will also be magnified or minimized by your own power levels, drafting abilities, and more.

Your best Zwift wheelset, more than ever, depends on your specific situation.

About the Results

Below you will find the five fastest wheelsets in game for flat and mixed courses. “Flat and mixed” covers most routes on Zwift, with the exception of anything including major climbs like Alpe du Zwift, the Epic KOM, or Innsbruck’s KOM. This isn’t an exact science, but we’re pretty confident in our findings thus far.

This list is based on our recent speed tests using the very flat Tempus Fugit route.

Your Winners

#1: DT Swiss ARC 1100 DICUT DISC

655,600 Drops to purchase, must be Level 39+

Released in Zwift’s August 2021 update, the new disc wheel from DT Swiss is head and shoulders above the competition, turning in a flat test time 6.5 seconds faster than the Zipp 858/Super9. Read more about this wheelset >

#2: Zipp 858/Super9

550,000 Drops to purchase, must be Level 37+

Released with the Fuego Flats expansion, this wheelset just barely edges out the 808/Super 9 below. And how about those colors? This is a fast wheelset… but also one of the heaviest, so don’t use it for long climbs.

#3: Zipp 808/Super9

450,000 Drops to purchase, must be Level 34+

The original disc wheelset. Beta users got it, then Zwift finally made it available a few years later to everyone. Like the 858/Super9, this is a fast but heavy wheelset, so it’s not recommended for long climbs. This wheelset is just 2 seconds slower than the 858/Super9 over our ~50 flat test.

#4: Zipp 858

300,000 Drops to purchase, must be Level 31+

The innovative sawtooth design of this wheel is definitely eye-catching, and at least in game it makes for some nice speeds. It’s the fastest non-disc wheelset in game, barely edging out the ENVE 7.8.

#5: ENVE SES 7.8

278,900 Drops to purchase, must be Level 36+

ENVE’s deep dishes are nice and zippy. Generally, deeper the dish, the faster and heavier they are. But that changes when we look at the ENVE 7.8 vs the 8.9, where the 7.8 is 1s faster on the flats, and significantly faster on the climbs.

#6: ENVE SES 8.9

250,000 Drops to purchase, must be Level 28+

Even though the 8.9’s don’t climb well, they’re still pretty fast on the flats, so they’ve earned their spot here.

#7: Zipp 808

200,000 Drops to purchase, must be Level 25+

Originally the best all-around wheelset, the 808’s have slipped down several notches with the arrival of newer wheelsets. It is the easiest wheelset to attain on this list, though, so you may want to pick it up if you don’t have the Drops or ability to score the wheelsets above.

What’s the Time Gap?

The 6 wheelsets above are separated by 17.5 seconds on a ~50-minute, 2-lap test of Tempus Fugit at 300 watts with a 75kg rider. See our complete data charts for details >

The next fastest wheelset, the Zipp 454, is 6 seconds behind the slowest wheel on this list. So the wheelsets above are truly in a group by themselves when it comes to speed on flat/mixed courses.

About the Tron

This is what everyone’s asking: is the Zwift Concept 1 (Tron) still the fastest bike?

The Tron bike is one unit–you can’t separate its wheels from its frame. So we can’t test the wheels for their performance like we can all other wheelsets in Zwift. But we have created a handy chart comparing the Tron bike to other top performing frames+wheelsets in Zwift >

Share Your Thoughts

We hope all you Zwifting fanatics found this post useful. Got questions or comments? Post below!

Changelog

Important note: this post contains speed test results for Zwift frames or wheels. These results may change over time, and a bike's performance relative to others may also change. We don't always revise posts when performance rankings change, but we do keep current, master versions of our speed test results which are always available. See the frame charts, wheel charts, and Tron vs Top Performers for current performance data.

Dede Griesbauer on Longevity in Endurance Sports (Zwift PowerUp Tri Podcast #17)

Eric Schlange
-
0
Dede Griesbauer on Longevity in Endurance Sports (Zwift PowerUp Tri Podcast #17)

Dede Griesbauer on Longevity in Endurance Sports (Zwift PowerUp Tri Podcast #17)

Have you noticed that many elite endurance athletes are getting fitter and faster as they get older? Some are peaking in their late 40s.  Multi-time Ironman champ and Kona top 10 finisher Dede Griesbauer sat down to chat with us about this recent phenomenon. With an athletic career that will reach its 4th decade this January, Dede has some great insights. 

Lots of recent race news, too, as well as Jordan’s quick take on CBD and its place in triathlon in this episode.

About the Podcast:

The Zwift PowerUp Tri Podcast is hosted by former pro triathletes Matt Lieto and Jordan Rapp, lending their in-depth knowledge of the multisport to the Zwift Tri audience.

Give this Guy a Ride On: Michael Knudsen Riding 5000km on Zwift

Eric Schlange
-
0
Give this Guy a Ride On: Michael Knudsen Riding 5000km on Zwift

Give this Guy a Ride On: Michael Knudsen Riding 5000km on Zwift

Michael Knudsen is not your typical cyclist.

At first glance he may just look like another strong rider: 31 years old, 71kg, with an FTP around 324 watts putting him in the ~4.6w/kg range (see ZwiftPower profile). He’s a level 50 Zwifter with over 150 Zwift races under his belt.

But look a little closer and you’ll find this Copenhagen resident isn’t just another strong rider. He’s an ultra cyclist, putting in massively long rides, including vEveresting on Zwift. And he’s not just another ultra cyclist–he’s an exceptionally gifted one, finishing 2nd in the ridiculously grueling 2018 Red Bull Trans-Siberian Extreme (“You cannot train for 9100km,” Michael said, “It will always be in your brain–this was for sure the mentally hardest race I have ever done.“)

And right now, Michael is set up on Zwift at the massive Field’s shopping mall in Copenhagen, with the goal of riding the Race Across America (RAAM) distance of 5000km within 10 days.

He’s got about 850km to go, and figures he will finish around 10am local time Thursday–right on schedule.

Reasons

Why is he doing it? Here’s what his partner Kathrin Fuhrer said on Facebook:

Fluid in the legs? 9 days of zwift. + 4000 km
  • This is our passion, cycling and crazy stuff is just us
  • We want to get the word out for “the real” Race across America next year 2020 where Michael will take part
  • We want to motivate people to ride with us and engage them to ride their longest ride ever as well as learn a bit more about ultracycling and what is needed for it
  • To achieve this goal going to America for RAAM next year, we need to collect some sponsor money as the whole thing will cost around $35,000USD

Challenges

I asked Michael what the toughest parts of his epic Zwift have been so far. Here’s what he had to say:

Knudsen’s Garmin at the time of this post

Toughest part has been being in the mall on the trainer. Fixed cycling position for 240 hours. Living in air-conditioned air and getting dry blood in the nose due to the dry air.

All the uncontrolled elements in a big mall. Like music at night. Long walk to toilet. Sometimes up on 2nd floor. Shower in gym on 3rd floor.

Not seeing the sky or breathing fresh air for 10 straight days. Not sleeping in your own bed.
Constantly having thousands of people around you.

It’s difficult to fathom how mentally demanding it is to sit still on a trainer in a mall for 10 days, only walking the route from the toilet and back to the bike.

Not seeing the sky at all also plays a little with you over time.

But it hasn’t been all bad, of course. Michael also said:

Other than that it has been an amazing event and the Zwift riders from all over the world have been outstanding. Really incredible. I basically never ride alone.

Pushing Limits

Obviously, a ride like this pushes a person’s limits to the edge. But what Michael didn’t mention above is that his rides are also pushing Zwift’s limits, with the app struggling when rides get over 1000km in length, according to one Facebook comment I saw from Michael.

Just try to look at the details of his first activity in Zwift Companion… here’s what you’ll get:

Ride with Michael

Zwifters have joined Michael on every bit of his RAAM ride, lending him a wheel and chatting to keep him company. You can join him as well! Just look him up on Zwift Companion, follow him, then choose to “Ride With” him in Watopia.

If nothing else, give him a ride on!

Watch His Ride

Zwifting Multi Streamers has been featuring Michael’s ride group on and off during the attempt, so check their Facebook page for a live feed.

Are Zwift’s Tarmac Speeds Realistic?

Eric Schlange
-
9
Are Zwift’s Tarmac Speeds Realistic?

Are Zwift’s Tarmac Speeds Realistic?

When you compare your Zwift speeds to outdoor speeds, which is faster?

Zwifters tend to find their speed in-game is faster than their speed outdoors, typically by a margin of 1-4km/hr for solo rides. This can be disconcerting since many riders set fitness goals based on miles ridden per week.

There are several possible reasons why your outdoor speeds could be slower than your Zwift speeds:

  • Zwift’s algorithms for calculating speed are incorrect.
  • Zwift’s calculations include assumptions (such as your position on the handlebars or the condition of the road) that do not reflect your outdoor rides.
  • Outdoor rides are slowed in ways Zwift rides are not.
  • Zwift gives you more opportunities to draft.
  • Your power readings in Zwift are not accurate, leading to unrealistically high speeds. (This is a separate topic which we won’t deal with here.)

Let’s look at these possibilities in detail and figure out what’s really happening.

Possibility #1: Zwift’s Algorithms for Calculating Speed are Incorrect

Formulas for calculating bike speed are well-established and reliable (see some details about that here). Zwift simply feeds a variety of variables in such as:

  • Rider and bike weight
  • Rider CdA
  • Aero settings for equipment
  • Tire rolling resistance
  • Gradient
  • Air density

into an established equation for speed, and the correct speed gets spit out.

I highly doubt ZwiftHQ’s game physics programmers have made any major errors in this area, so I’m going to dismiss this as an issue.

Possibility #2: Zwift’s Calculations Include Assumptions Which Do Not Reflect Your Outdoor Rides

Do you spend most of your time on the flats, on the hoods, or in the drops? How high is your seat, and how tucked in are your elbows, knees, and head? How smooth is the pavement?

There are other factors that affect your speed apart from the basics of road grade, power, and rider weight and height. Zwift has to make some assumptions here, and it’s fair to say those assumptions are quite generous.

One example: all Zwift tarmac has the Crr of new tarmac. How often do we get to ride on new pavement outdoors? It rolls fast.

We ran some speed tests on the Fuego Flats forward sprint with different rider weight and watt variables. (No, we didn’t run the numbers for climbs or group rides–we wanted to keep it simple.) Rider height was set to 175cm, the average male height in the US. We used timings from a basic Zwift Carbon frame/32mm wheels setup as well as a very fast setup of the Specialized Venge S-Works frame/Zipp 858 wheels.

We compared those numbers to two different online bike calculators, setting them to 25 degrees Celsius, 50m elevation, 8kg bike weight, riding in the drops, and default values otherwise. Here are the results:

  • 300 watts, 100kg rider
    Zwift Carbon, 32mm carbon wheels: 38kph
    Specialized Venge S-Works, Zipp 858 wheels: 38.9kph
    BikeCalculator.com: 39.16kph
    Kreuzotter.de result: 37.1kph
  • 225 watts, 75kg rider
    Zwift Carbon, 32mm carbon wheels: 35.8kph
    Specialized Venge S-Works, Zipp 858 wheels: 37.2kph
    BikeCalculator.com: 35.77kph
    Kreuzotter.de result: 34.8kph
  • 150 watts, 75kg rider
    Zwift Carbon, 32mm carbon wheels: 30.8kph
    Specialized Venge S-Works, Zipp 858 wheels: 31.4kph
    BikeCalculator.com: 30.55kph
    Kreuzotter.de result: 30kph
  • 150 watts, 50kg rider
    Zwift Carbon, 32mm carbon wheels: 33.7kph
    Specialized Venge S-Works, Zipp 858 wheels: 34.4kph
    BikeCalculator.com: 31.44kph
    Kreuzotter.de result: 31.7kph

As you can see, Zwift speeds are generally within 2kph of the bike calculators, with the greatest difference being at the 150 watts/50kg rider test where the difference between one calculator and our fast test setup was 2.96kph.

It’s worth noting that small changes in the bike calculators (such as using tubulars instead of clinchers, or a slightly improved CdA, or a higher elevation or temperature) are enough to equalize the speeds.

Additionally, the bike calculators don’t let you set which bike frame or wheelset you’re using, and we know those impact speed outdoors just like they do in Zwift!

Overall, I think it’s safe to say Zwift makes some slightly generous assumptions about your posture and environment which do not reflect a typical outdoor ride experience. The difference these assumptions make in speed is reflected in the numbers above.

Possibility #3: Outdoor Rides are Slowed in Ways Zwift Rides are Not

Stopping at traffic lights, braking to avoid vehicles or other obstacles, slowing into turns on steep descents–these are all are common occurrences outdoors which never happen in Zwift.

Then there’s wind, which doesn’t exist in Zwift but plays a major factor outdoors, rarely helping but often hindering our speed. And there’s coasting, which we do all the time outdoors, but rarely on Zwift.

Consequently, your Zwift average speeds will logically always be higher, even for similar effort levels.

Possibility #4: Zwift Gives You More Opportunities to Draft

Drafting in Zwift, like drafting outdoors, results in a power savings of ~25% or more. If you are comparing your outdoor solo ride speeds with Zwift speeds, the draft effect alone will make a noticeable difference since you often benefit from some draft effect in Zwift even if you’re not taking part in a group ride.

Conclusion

While a Zwift ride will generally be faster than an outdoor ride of similar effort, the differences are minimal and not enough to invalidate including Zwift miles in your training metrics.

Pro tip: if you’re really looking to take your fitness to the next level, start training with power and TSS, with less focus on distance. Here are two top books on this subject:

The Cyclist’s Training Bible
Training and Racing with a Power Meter, new 3rd Edition

Six Day Special with Mark Cavendish, Elia Viviani, Katie Archibald and more (Zwift PowerUp Cycling Podcast #38)

Eric Schlange
-
0
Six Day Special with Mark Cavendish, Elia Viviani, Katie Archibald and more (Zwift PowerUp Cycling Podcast #38)

Six Day Special with Mark Cavendish, Elia Viviani, Katie Archibald and more (Zwift PowerUp Cycling Podcast #38)

In a 6 Day Special we let Events Specialist Dani Rowe loose on track centre at the London Velopark – where she won her Olympic gold medal in 2012. She discusses all things track, goals, and cheat meals with some of the biggest names in the game – Mark Cavendish, Elia Viviani and Katie Archibald, just to name a few.

Dani grabbed Cavendish just as he’d come off the track (you can still hear the action going on in the background!) and she cornered Barbieri in a treatment room to talk girl power, so it’s a jam-packed episode bursting at the seams with stories, ambition, and winners!

About the Podcast

The Zwift PowerUp Cycling Podcast features host Matt Rowe (Rowe & King), with regular co-hosts Greg Henderson (Coach Hendy) and Kev Poulton (Powerhouse Cycling). The hosts share tips on how road races can get faster in-game and outdoors.

All About Zwift’s Cannondale SuperSix EVO Frame

Eric Schlange
-
3
All About Zwift’s Cannondale SuperSix EVO Frame

All About Zwift’s Cannondale SuperSix EVO Frame

Zwift added the Cannondale SuperSix EVO bike frame to the Drop Shop this week, and Cannondale fans have been justifiably pestering us with messages ever since. “It’s the only bike in Zwift with 4 stars for weight and 3 for aero! It must be fast… how does it perform?”

We’ve been hard at work putting test data together. It hasn’t been easy, because much of past test data was invalidated when Zwift modified the rolling resistance of various surfaces, making it so we can’t compare new test runs to old. But we still made it happen… for science!

Fast On the Flats, Faster Up the Climbs

The SuperSix EVO is billed as an “aero climbing bike”–that is, Cannondale is trying to find the holy grail of cycling–a stiff, fast aero bike that is also featherlight so it climbs well. Some say they’ve achieved that with the newest SuperSix EVO, rolled out in the summer of 2019.

Kudos to Zwift, who have set the bike up in-game to perform the way Cannondale intended–it’s fast on the flats, and faster up the climbs.

Seeing Stars

And it is the only frame in the Drop Shop combining a 4-star weight rating (the best possible) with a 3-star aero rating. You may think this would make the SuperSix EVO the fastest climbing frame, since the other top climbers are rated 4 stars for weight and only 2 for aero.

But you’d be wrong.

Test Results

While the SuperSix EVO slots in at #6 on the list of best climbing frames, it is still 8 seconds behind the King of the Alpe, the Specialized Tarmac Pro, on a ~50-minute climbing effort. So it’s a strong climber, but not the strongest.

And while it’s pretty fast on flat roads, it is still ~8 seconds behind the four fastest frames over a ~1-hour effort. (Those four fast frames are the Tron, Felt AR, Cervelo S5, and Specialized Venge S-Works.)

This is a perfect illustration of how Zwift’s 4-star grading scale, while simple, falls short in certain instances.

For example, the Cannondale EVO frame is rated 4-stars for weight and 2 for aero, so you would think it would be slower up the Alpe than the new SuperSix EVO. But no! If we were able to look at the precise numbers Zwift has used for these frames’ weight and CdA, we would see that the SuperSix EVO is slightly heavier than the standard EVO, even though both are rated at 4 stars.

Chameleon Paint Scheme

You can unlock the EF Education First Pro Cycling Team paint scheme if you finish the Giro de Rigo. But what about the default paint on this frame? To be honest, we’re still trying to figure out how it works, because it’s unlike any other colorway on Zwift! We can’t figure out if it’s a wacky bug, or a crazy sort of reflective paint. Check out these in-game shots, where the frame is clearly “mimicking” its surroundings:

The Verdict

If you don’t want to mess around with a “climbing rig” and an “aero rig,” you could do worse than having the Cannondale SuperSix EVO as your only frame.

Keep in mind you’ll need to be at level 36+ and have at least 816,500 Drops available to make the purchase.

The SuperSix EVO won’t beat the “specialized” rigs on their own turf, but it’s pretty darn close. If you’re looking for an all-around frame with a bent towards climbing, this is a smart choice.

Your Thoughts

Have you picked up the SuperSix EVO yet? What other frames would you like to see in game?

Important note: this post contains speed test results for Zwift frames or wheels. These results may change over time, and a bike's performance relative to others may also change. We don't always revise posts when performance rankings change, but we do keep current, master versions of our speed test results which are always available. See the frame charts, wheel charts, and Tron vs Top Performers for current performance data.

Zwiftcast Episode 71 – Rumble in the Jungle, Whither the Rivals and all the latest Zwifty chat

Simon Schofield
-
0
Zwiftcast Episode 71 – Rumble in the Jungle, Whither the Rivals and all the latest Zwifty chat

Zwiftcast Episode 71 – Rumble in the Jungle, Whither the Rivals and all the latest Zwifty chat

Simon, Shane and Nathan reconvene to discuss and analyse all the latest Zwift news and chat.

The podcasters get things going with a new take on the age-old Kickr vs Neo debate, updated to the KickrBike vs NeoBike dilemma. If you have 3k plus to spend on a smart bike this will be of interest to you… Shane is one of the very few people in the world with saddle time on both, so what’s his verdict?

The Zwiftcasters move on to consider the demise of VirtuoGo, one of the rivals to Zwift. It has just announced that its number of subscribers is too low to be sustainable so the issue for discussion is: can any of the rivals live with the well-funded, well-established, heavily-promoted Zwift?

Next, it’s the Rumble in the Jungle–no significant game update would be complete without some Zwifters having a bit of a moan. Is the long-ish lists of complaints about the increase in rolling resistance justified? The podcasters try to put it all in perspective.

Meanwhile, Simon has a problem with the stand-up/sit-down threshold changing to 80rpm… has Zwift got this one right?

Along with a few more tidbits of Zwift stuff – including a review of the Keep Everyone Together Group Ride functionality by experienced Zwifter Christian Nistri – this is a busy edition of the podcast for Zwifters.

We hope you enjoy listening.

Zwift Update 1.0.41943 Released

Eric Schlange
-
1
Zwift Update 1.0.41943 Released

Zwift Update 1.0.41943 Released

Zwift released a minor update two nights ago, mostly to support upcoming special events like Movember and Giro de Rigo. But it also includes a new bike frame (always fun) and some interesting notes from the Game Master himself, Jon Mayfield.

Reworked “Lead-In” System for Events

Events have long had issues with displaying accurate distances, due to how Zwift was computing lap distances. They may also display confusing or inaccurate information about the “lead-in” portion of the event. Notes for this update included this blurb:

Reworked ‘lead-in’ system for events. This should improve the lead-in countdown before lap based events, as well as resolve some of the other discrepancies we saw in the user interface during those events.

We haven’t been able to test these updates much, but hopefully some or all of these issues have been resolved. (That said, we’ve already seen some Zwifters complaining about strange distances on their Strava rides vs Zwift rides, so it looks like Zwift may have fixed some bugs but unearthed some others.)

Tweaked Meetup Speeds

Zwift released the “Keep Everyone Together” option earlier this month, and while feedback has generally been very positive, word has quickly spread that speeds in these meetups are higher than they should be. This has resulted in some riders hosting meetups on their own, just to get a boosted speed. (It may seem silly, but most are probably doing it to prove a point.)

Zwift has adjusted these speeds down in this update. Here are some of the speeds we saw in today’s test:

  • 37kph when both riders held ~200 watts (before the change this was 45kph)
  • 45kph when both riders held ~400 watts (before the change this was 55kph)
  • 39kph with one rider at ~200 watts and the other at ~400 watts (before the change this was 50kph)

Compare these speeds to a simple test we ran outside of the meetup:

  • 40kph when both riders held ~200 watts
  • 47kph when both riders held ~400 watts

So we’re now moving a bit slower in the Meetups than we would be in a free ride. We’re closer to the free ride speed than we were previously, but some riders are already complaining that it feels a bit slow.

New Cannondale SuperSix EVO

New bike day! This frame has been released to support Sunday’s Giro de Rigo event, where riders can unlock the Education First Pro Cycling Team paint scheme for this new bike.

Of course, you’ve got to buy the bike first! It requires you to be at level 36+, and costs 816,500 Drops.

How quick is it in-game? We’ll post about this soon, but for now we’ll say that it looks like decent performer on the flats, but it really does best on the climbs.

Other Changes

This update included several other changes/fixes as well:

  • Fixed screen-size issue due to an Apple change in tvOS 13 on certain Apple TV’s, when used with 720p televisions
  • Fixed WatchOS 6 support
  • Fixed crash on iOS 13.1 + iPhones (this was bugging a lot of folks!)
  • Added additional variant of new user tutorial (making it easier for the newbies is always good)

Comments from Jon

Jon included some interesting comments on two topics. First, regarding recent rolling resistance changes:

I also wanted to touch on the rolling resistance changes that we made earlier in the month. Up until recently road and TT bikes went unrealistically fast on rough surfaces. As we start to think about adding other bike types such as gravel, cx, mountain bikes, fat bikes, etc in the future it made sense to start taking into consideration surface roughness vs tire/bike type. We don’t allow drafting on TT bikes. Some bikes are kilograms heavier than others and are bad for climbing. And now, road bikes can’t pretend dirt is tarmac. Dirt now acts more like dirt.

Now, with that out of the way, much of the negative feedback from some of our most outspoken Zwifters really stemmed from the fact the transition wasn’t handled very gracefully. We did a poor job of ensuring everybody got that specific update at the same time causing group rides to get blown apart as well as races being won unfairly. We’ll now be building some simple restrictions to events to ensure a more level playing field for the future (ie, requirements on equipment, up-to-date software, etc). We regret that some of the competitiveness of some group rides and races was ruined. As we continue to deprecate the older clients this issue should become mostly resolved. Additionally we’ll soon try and ensure everybody riding on events specifically on the jungle circuit will be doing so on the same class of bike (ie, MTB).

And secondly, regarding the Repack Ridge FutureWorks steering test:

Finally, I want to thank the community for all the feedback on the steering test we launched earlier in the month. While the early skeptics were certainly vocal about it in the “Whats new” post previous to this one, in the end we’ve had almost 20,000 test rides fully completed start to finish and overwhelmingly positive feedback. While it was meant to be a low volume test, I think it could end up as a permanent addition to Watopia in the future. Who knows – maybe a whole network of single track trails can spread through the wilderness of Watopia one day.

Read Jon’s full update notes >

1...397398399...526Page 398 of 526