Pack Dynamics Test Events (December 2022)

It’s not really a dropout if a pedal PM is working as designed and only reports power, say, every 1.5 seconds at 40 rpm or every 2 seconds at 30 rpm. However, dropping power to zero for those one-second intervals in which there is no power reading (possible for any cadence below 60 rpm), Zwift would underreport power by a significant amount. Therefore, a small delay is needed in these cases, at least for low cadences. The current amount of time is grossly overdimensioned, though.

It’s a by-product of the mechanism that it also glosses over intermittent dropouts, but I presume it was not really made for that purpose.

1 Like

Yeah I’m hoping they have already started on a ground up Zwift 2.0, but I think this will be a sticky plaster job forever more.

Re: braking - thank you!!! I honestly felt like I was losing my mind with how quickly I was getting spit out of the back :joy:

1 Like

It’s called allometric scaling.

It’s absolutely real. It’s the ‘general’ rule of human performance in cycling.

A woman beating you or a smaller rider doesn’t make the math wrong. It means that their power vs CdA and other forces was greater than yours. On that day, they were better relative to you. In fact, due to them being smaller, it’s highly likely, that in w/kg they had to significantly stronger than you.

That is some serious n=1 math…

Off topic slightly but seeing as you mentioned it does this mean then that Zwift does not do any prediction to account for lag. More for the microbursting technique when a rider stops pedalling does Zwift instantly zero the riders power?

So the benefits of mirobursting is more down to the hardware layer e.g elite directo which is known to have a poor power taper holding watts higher than they should be.

Has anyone here heard more about whether server-side positioning is being tested soon?

1 Like

Knowing about this bug now and doing “group rides” with double draft like TdZ stages, it’s super noticeable and annoying! I always wondered why it was easier to coast in races than in e.g. RoboPacer rides. 1.33 can’t arrive soon enough. Never stop pedaling until then.

today might be the day?

1 Like

Well that would be a nice surprise. I was worried they wait until TdZ is over in 2nd half of Feb so they don’t break anything mid-way.

en que version de dynamic pack hay, o esta funcionando?
What version of dynamic pack is there, or is it working?

PD3 is what is generally in use at the moment.

An in-progress version of PD4 is included in the current game, but is only enabled for specific test events, and not for anything else.

1 Like

Only a few events are using PD4 as it still requires some work. Waiting on next came update towards the end of the month i think.

if you want to try it see events here. The at Home Cycling & Running Virtual Training App

I’m tempted to turn it on for some crit city events as well

1 Like


win you

I have a questions.

During last weeks ZRL TTT race I thought about how hard it is to maintain position in a team group. I think that the changes proposed for PD 4 should make it easier. This leads to my question. Has there been any TTT races done with PD4 to test it from that perspective. I really think that this should give good data on how fast groups are and how much movement there is between riders within a group.

1 Like

Any update on the development and maybe when new tests planned?

Probably on hold until 1.33 is out which will be no later than when Scotland stages for TdZ start.

Hopefully restarting next week :slight_smile: