zPower: Don't be a "Flier"

(Will I Am) #141

seems like the data for a given rider will show  in consistencies   that  expose those who are gaming the system or  are technically unable to setup their trainers

One example - whats up with these  guys  riding  at  5 or 6+ W/kg?    Lets  face facts  -  if they are  sustain this for  longer than  3 mins - there is 99.8% chance its bogus


(Michael Meysarosh) #142

Someone who is doing cat 3/2/1 racing would be able to maintain these kind of figures. But you would have to consider any them dedicated athletes and possibly sponsored professional, who could very well be using Zwift for training.

You can see some impressive times on Richmond and Wattopia that have efforts that rival or exceed the TDF field. I quickly checked out one and it calculated to a 6.5 W/kg for about 40 minutes. I think Lance on dope hit about 6.8-7 W/kg, so have some flagrant doping going around these virtual parts.


(Yancy Fry) #143

Dude, every cat 1/2/3 is going to have a power meter.


( 41, Godspeed ) #144

Would you be happy if…

  • When using a Controlled Trainer, you could enter a virtual world where only Controlled Trainers exist?  This assumes Zwift builds into the system multiple worlds with a selection screen - hopefully coming in some sort so we can select the location we want to ride in - something like, select site Watopia, Virginia, etc), then select world (Controlled Trainer only, zPower only, Open).  If you did’t have a Controlled Trainer, those options would not exist in the selections.
  • A filtering mechanism existed such that you could mask users to only Controlled Trainers, zPower, or Both.  This way, if you want to only see Controlled Trainers, only those users would be shown in your environment, on leader boards, etc.  But if you wanted a ride with a friend that uses zPower, you need to change the masking to allow Both in your view.

I would guess the first option, separate virtual sites/worlds would be easier to implement and require a lot less coding - pretty much the same model most video games use today for match making and dropping competitors into their own virtual world.  Could be as simple as running the environment X number of times and writing a simple lobby program for dropping folks into the world they want to ride in, of course, checking the system requirements against the hardware the user is using.

As many have noted, there is always being to be differences in hardware, resistance, air pressure, etc, that is going to make one person’s effort a bit more/less than another but if you didn’t want to ride in a world where someone could seriously manipulate the effort required, aka Controlled Trainer, you could spend the money and select a Controlled Trainer Only world to ride in.  And of course, for that group ride with the buddy that only has zPower, drop back to the lobby and drop into a world with them that allowed both zPower and Controlled Trainers.  Just keep in mind, unless everyone rides the same physical trainer, there are going to be differences in effort for the same results - you just want to minimize these if you can.  I could even see Zwift taking this to the extreme and creating a different world for each mug’s trainer - wouldn’t that impact the social aspect :-).



(Joel Dudgeon) #145

It appears Zwift will offer a tiered subscription model with smart-trainer subscribers being charged the most.  While I don’t totally understand why an electronically controlled resistance trainer should cost more than a power-meter setup, but if that’s the case, I’m going to want the option to only ride with power-meter/smart-trainer users.  If I’m going to have to pay a premium to ride with smart resistance, then a power-meter/smart-trainer only server should probably be an option for those subscription tiers.


(Johnny RCCSFC (D)) #146

I agree 110% with Joel… If I have to pay more … I’m not gonna like riding around with dumb trainers.


(Will I Am) #147

Michael - seriously - I was a Cat 1/2/3 - I  know the deal - you  are  sadly mistaken if you  think   the  typical CAt 1/2/3 can  Ramp up  to  6 W/KG  for a   30-40 min  effort


But  I  think you  get it based on  your  calculation  - there are not   too many  people that  can  hold  420- 460 watts for  30-40 mins -   and if you are someone who  can   chances are  you are not  riding on Zwift!


Would be   great to ban  the   cheater on zwift for  the  rest of us  mortals


(Ben Crone) #148

Will, I assume Michael was referring to your 5-6 W/kg for 3 minute comment. I agree with him 5-6 for 3 minutes isn’t unreasonable for a trained cyclist (I can do it and I’m nothing special).

I see your point though, and 40 minutes is obviously a different story.


Joel: Where did you hear about a tiered subscription? I haven’t seen any evidence of this and it seems quite unlikely.


(Stewart ) #149

The charge is going to be the same for everyone. They have described the different trainer types and the Zwift experience they can offer which may be confused with a pricing structure.


(Michael Meysarosh) #150

Yes, I am referring to the ability for a cat 1//2/3 to be able to maintain those power figures from around 3-5 minute ranges.

I brought up some of of those 30-40 minute examples that were taken from some of the best climbs by elite cyclists that won the TDF. Oh, there were doping when they accomplished those figures.

Look at the top of the lap charts on Strava for the two Zwift courses and you see numbers that high, which is why I stated we have some impressive virtual dopers.

I agree and actually made a comment earlier that Zwift events should be limited to power meters and smart trainers. I also highly disagree that the standard monthly fee should be different for any form of equipment, but commented that special events or event series can have a fee that would fund prizes for the top three competitors, and some for Zwift for their effort.


(Jason Buccellato) #151

Joel Dudgeon and  Michael Meysarosh, thanks for your comments.

Online gaming is competitive, and cycling can be extremely competitive. The beta is free currently, however, once we move to paid subscriptions, this zpower issue will become more pronounced. Software developers know well that entire gaming communities have abandoned certain titles due to unchecked online cheats, and I’m confident that Zwift has been educated on this matter.

We can thank, in part, adopters of power meters and wattage trainers for setting the stage for online software such as Zwift, Trainer Road, and CVT. These folks have invested in accuracy. The onus of making the case for accurate online racing environments is not their responsibility. Accuracy is the default goal.

Power users will move to whichever platform best accommodates their expectations. Trainer Road has online racing in development, and CVT already supports it, with a new software version in beta at the moment.

Another point worth making is that power equipment decreases in value quickly. It is new technology, and quickly changes. Ebay and your local craigs list are your friends if you cannot afford brand new, or care to just spend less.

At the end of the day, my view is that there will be no other option but to have separate leader boards. Zwift will have an enormous customer base. They can keep that base by caring for the expectations of their power users. 


(anon46748293) #152

So what is next? Receipts from doctor certifying that cyclist X has weight of Y. Prohibition of power meters that can be calibrated at home conditions (I can hang 0.7kg on a crank instead of 1kg) ? Or special software monitor that makes sure that I did not install some software filter that modifies every packet with power values flowing from that Garmin ANT+ stick? Or how about I will make cheap hardware ANT+ retranslator that will do the same?


(Jason Buccellato) #153

Kostya, are you suggesting that no controls should be enacted(including those addressing the most egregious issues) since all variables cannot be addressed?


(anon46748293) #154

There are situations where it is absolutely clear that this particular performance is out of whack and I agree that your software flags such cases. I am all for it. It at least shows cyclist that something is wrong. Maybe what you can also do is if you see cyclist performing at Cat 1 level and having no power meter you can send some personal hint just to that particular cyclist (like some colored performance icon indication the likeliness of such performance level).

It is obvious that you can not really do “official speed based competitions” due to inability to control external factors/cheating.

So I think it becomes more of a cultural question. Who rides Zwift and why. You can cultivate community of hard nosed a$$holes trying to measure who’s got a bigger one or you can encourage some tolerant society where people ride together for fun.

Which way your company goes or how it achieves that is way beyond my area of expertise. I am but a techie who just happened to love cycling. I am not a gamer though, I also do not give a flying hoot about someone passing me at warp speed. I suspect that I am probably not a good representation of a typical customer of yours.


(Jason Buccellato) #155

With winter approaching, the Zwift customer base size should not be an issue in creating two leader boards. The design of more maps will be a necessity in accommodating all of these subscription users in the near future.

Those who advocate for drug testing in the sport of cycling, are not hard nosed a$$holes. Those who recognize the real demand for more accurate indoor race/ride software should not be marginalized for seeing the need for such improvements. This is progress.


(anon46748293) #156

I have just expressed my own opinion. We do not have to agree :wink: I already told that I might not represent your typical customer.


( 41, Godspeed ) #157

I just saw a zPower rider climb Libby Hill at a minimum of 30 mph and then had another zPower rider fly by me as they coasted up a 10-14% grade; what a morale killer as I was putting out 400+ watts at 90 rpm and watched the zPower rider just coast by me and right up the hill without pedaling.  I guess I’m like some others that after working hard for an hour, it hurts to see someone blowing by with non-human capabilities.  I like being caught and pushed by those using controllable trainers, it pushes me and results in a better workout and eventually better performance.  


As mentioned prior, I would urge Zwift to consider implementing a lobby where users could select what course and what type of environment (zPower, Controllable Trainer, or both/open) which they would like to ride in.  Zwift could even let the user select a classification range they would like to compete within.  

Those users that want to compete (think race), and it looks like there are quite a number here in the beta program that have invested in a controllable trainer, would select the controllable trainer environment to ride in to compete, measure themselves, or like me train, get pushed, and know the field is level. Those with a controllable trainer could also elect to select the Open environment to ride in an environment which would be like the beta with all types of riders - perhaps ride with friends that don’t have controllable trainers and just want the social aspects Zwift offers.



(Will I Am) #158

That will kill the model -   I  agree with you J Lamb  -  what is  your  solution to the  obvious problem?


( 41, Godspeed ) #159

@Will I Am Adams - you are right. I wrote a comment and didn’t provide a suggestion - got lazy as I was just dismayed at the time and relied on the post from a few days ago.  So I pulled a component of that post forward and edited the post above to include a suggestion.  What do you think?  Would this work for you?


(Will I Am) #160

Good  solution - yes that would work for me  -  

I  also  think there will be  those with a smarttrainer who will game the system - it should be possible for  zwift to have an algorithm to sort t hem out - it  might  take a  red flag  to be  looked at  by a  human initially but - I’m  sorry - no one is  going to be  riding at  900 watts  for  extended periods of time  - let alone more than 30 seconds  etc…