[FRC Blog] The 2016 Season, by the Numbers 2

[GeeTwo]

Originally Posted by catmanjake

What does the 3114 total mean? Is that how many teams participated in an event this year?

That sounds about right. The sad truth is that half of the FRC teams that ever existed do not exist any more. The good news is that the longer a team manages to continue, the better its chances are to continue to continue.

[ATannahill]

Originally Posted by GeeTwo

That sounds about right. The sad truth is that half of the FRC teams that ever existed do not exist any more. The good news is that the longer a team manages to continue, the better its chances are to continue to continue.

I wouldn't go quite with half, I'd guesstimate at a third. FIRST will usually leave a gap between the last rookie of year x and the first rookie of year x+1. I know one year they skipped every other number for a certain range. This allows for if a team splits or if there is another reason to use a number from a previous year without having two teams have had the same number at a different time.

[EricH]

Originally Posted by rtfgnow

I wouldn't go quite with half, I'd guesstimate at a third. FIRST will usually leave a gap between the last rookie of year x and the first rookie of year x+1. I know one year they skipped every other number for a certain range. This allows for if a team splits or if there is another reason to use a number from a previous year without having two teams have had the same number at a different time.

And of the teams that do fold, sometimes they'll come back under a rookie number. Or under their own number for a season or two (FRC23). That can drive the team numbers up with respect to teams that have competed.

[cue Mark McLeod with the real numbers... wait, CD doesn't work that way]

[notmattlythgoe]

2363 Triple Helix should be row 861.

[marshall]

Originally Posted by notmattlythgoe

2363 Triple Helix should be row 861.

900 The Zebracorns are highly likely to be row 355.

[Greg Needel]

As a Supplier we really appreciate that FIRST released this data.


There are a couple of insights that we can gain from looking at this, which help us to make some choices about our future products and their uses.


We had some interesting thoughts, specifically related to motor controllers

The number of total motor controllers reported used is probably lower than it actually is, mainly due to the number of teams that use Y cables to power their PWM motor controllers.

The types of PWM motor controllers reported likely has some errors in it, as the report out is based on library choice in code and all of them are the same (victor, talon, SPARK, etc). The total number of SPARK's used for example is considerably less than the number that are "in the wild" which means either people used different libraries in their code or a significant portion of them are used in practice bots or for testing ( I suspect a bit of both)

There are 8 teams that show having zero motor controllers.....

The average number of controllers per robot is 7.25



I am 99% sure line 1089 is 2848

[MechEng83]

I was slightly thrown off by the lack of "encoders" considering we had 6 incremental and 1 absolute encoder, but I'm 95% sure that line 592 is us.
Most of our encoders were plugged directly into the TalonSRXs.

Not a lot of 2006 Rookie year C++ teams with no pneumatics. We definitely had more than 7 Talons on our CAN interface, though I don't remember 10.

[Greg Needel]

Originally Posted by MechEng83

I was slightly thrown off by the lack of "encoders" considering we had 6 incremental and 1 absolute encoder, but I'm 95% sure that line 592 is us.
Most of our encoders were plugged directly into the TalonSRXs.

Not a lot of 2006 Rookie year C++ teams with no pneumatics. We definitely had more than 7 Talons on our CAN interface, though I don't remember 10.

This leads to another question about the data....when was it taken? If this was from the first time robots connected for the season it would be a lot different than if it was from a team's last match. Maybe it is an average or maximum for all of the categories.

[marshall]

Originally Posted by Greg Needel

This leads to another question about the data....when was it taken? If this was from the first time robots connected for the season it would be a lot different than if it was from a team's last match. Maybe it is an average or maximum for all of the categories.

It was based on team number and I'm 93.2% sure it is based on the last time the team connected to the field so you are seeing the output from the final run.

I know it was based on team number and the number of controllers we have listed only happened for 1 match at NC DCMP.

[GeeTwo]

Originally Posted by Greg Needel

This leads to another question about the data....when was it taken? If this was from the first time robots connected for the season it would be a lot different than if it was from a team's last match. Maybe it is an average or maximum for all of the categories.

It's in the original [re]post - the largest for counted objects and latest for language and framework:

Originally Posted by OP

- Counted objects and TRUE/FALSE show the largest number of any given object used in any one match (i.e. if a Robot had 2 Encoders in match 1 and 3 Encoders in match 7, 3 will be reported). Language and Framework report what was used in the last recorded match.

[AlexD744]

Originally Posted by Greg Needel

There are 8 teams that show having zero motor controllers.....

I can see this happening if they only had a drive base and initialized it with the RobotDrive constructor that takes in port numbers instead of motor controller objects.

[Alan Anderson]

Originally Posted by AlexD744

I can see this happening if they only had a drive base and initialized it with the RobotDrive constructor that takes in port numbers instead of motor controller objects.

I encountered more than one LabVIEW-using team that used Drive for all of their motors, not just drivebase ones.

[Cel Skeggs]

Originally Posted by AlexD744

I can see this happening if they only had a drive base and initialized it with the RobotDrive constructor that takes in port numbers instead of motor controller objects.

There are some other explanations. At least two of the teams with no reported motor controllers are due to our framework - we don't report usage data currently, due to it not being a high priority for us compared to the other goals we could work towards. See rows 525, which is our team, and row 1951, which is another team that we know uses our software. These rows pretty much only report the language.

I could definitely understand it if something similar happened with another few projects.