Detector Building C [TRIAL]

User avatar
Umaroth
Member
Member
Posts: 222
Joined: February 10th, 2018, 8:51 pm
Division: C
State: CA
Location: Pretend hugging SIlverBreeze because social distancing
Has thanked: 3 times
Been thanked: 14 times

Re: Detector Building C [TRIAL]

Post by Umaroth » June 7th, 2019, 11:20 am

I competed in this event when it was a Division B trial event at SoCal States 2018, and I do have some thoughts on how it could be improved based on last year and what I heard this year.

The number one issue with the event in its current state is the lack of separation between the top and bottom teams. From what I can tell whenever this event is run, the top teams are all extremely close to each other, as are the bottom teams, with a huge separation between the two groups because of the heavily weighted build log. At the national level, where the top will most likely have nearly perfect scores with little to break ties, it may as well come down to luck. Granted, the only environmental factor that has been tested thus far has been temperature, but with an entire season to work on the detector, I doubt there would be anything preventing teams to nearly perfect every other factor.

My idea: make this event sort of a Swiss Army Knife do everything type device
Similar to how many events have topics that will only be tested at the State and National level, there would be different environmental factors that would be tested at each level. Not every factor will be tested each time, so, for example, and ES at one tournament could choose to test temperature and salinity, whereas another ES at another tournament could choose to test air pressure and wind speed. With this type of set up, competitors would have to prepare for every type of environmental condition that could be tested, but it is not guaranteed that all will be tested at once. While I could see cost getting in the way of this, it could be worked around by making the ones that are more likely to cost more only tested at the State or National level. By having the event set up this way, there would be much more to separate teams, and it would require teams to find more efficient ways to calibrate their devices and complete the tiebreaker quiz with so many things to test.

I hope the event committees will listen to people's suggestions for this event. Detector has the potential to be a great event provided that the rules are set out well.
Troy SciOly 2019-now
The Ambassador

Proud Kraemer SciOly Padre 2018-now
Dank Memes Area Homeschool Juggernaut 2018-now
Kraemer SciOly Pretty-Much Head Coach 2019-now
2020 Tryouts: Circuit, Code, Detector, DP, GeoMaps, Machines
States Events: HANDS ON DISEASE DETECTIVES

Umaroth's Userpage

User avatar
builderguy135
Member
Member
Posts: 659
Joined: September 8th, 2018, 12:24 pm
Division: C
State: NJ
Has thanked: 17 times
Been thanked: 12 times

Re: Detector Building C [TRIAL]

Post by builderguy135 » June 7th, 2019, 2:56 pm

Umaroth wrote:I competed in this event when it was a Division B trial event at SoCal States 2018, and I do have some thoughts on how it could be improved based on last year and what I heard this year.

The number one issue with the event in its current state is the lack of separation between the top and bottom teams. From what I can tell whenever this event is run, the top teams are all extremely close to each other, as are the bottom teams, with a huge separation between the two groups because of the heavily weighted build log. At the national level, where the top will most likely have nearly perfect scores with little to break ties, it may as well come down to luck. Granted, the only environmental factor that has been tested thus far has been temperature, but with an entire season to work on the detector, I doubt there would be anything preventing teams to nearly perfect every other factor.

My idea: make this event sort of a Swiss Army Knife do everything type device
Similar to how many events have topics that will only be tested at the State and National level, there would be different environmental factors that would be tested at each level. Not every factor will be tested each time, so, for example, and ES at one tournament could choose to test temperature and salinity, whereas another ES at another tournament could choose to test air pressure and wind speed. With this type of set up, competitors would have to prepare for every type of environmental condition that could be tested, but it is not guaranteed that all will be tested at once. While I could see cost getting in the way of this, it could be worked around by making the ones that are more likely to cost more only tested at the State or National level. By having the event set up this way, there would be much more to separate teams, and it would require teams to find more efficient ways to calibrate their devices and complete the tiebreaker quiz with so many things to test.

I hope the event committees will listen to people's suggestions for this event. Detector has the potential to be a great event provided that the rules are set out well.
No no no no no no no no please no no no no no no that's gonna take forever to build
HSN '22 & DMAH Co-Captain
6-time medalist in Nationals 2019
2020 Sounds MIT and Princeton Champion
2020 (and probably 2021 :( ) Events: PPP/Sounds/Wright/Code/PaRaSiToLoGy

User avatar
Unome
Moderator
Moderator
Posts: 4250
Joined: January 26th, 2014, 12:48 pm
Division: Grad
State: GA
Location: somewhere in the sciolyverse
Has thanked: 64 times
Been thanked: 29 times

Re: Detector Building C [TRIAL]

Post by Unome » June 7th, 2019, 3:01 pm

builderguy135 wrote:
Umaroth wrote:I competed in this event when it was a Division B trial event at SoCal States 2018, and I do have some thoughts on how it could be improved based on last year and what I heard this year.

The number one issue with the event in its current state is the lack of separation between the top and bottom teams. From what I can tell whenever this event is run, the top teams are all extremely close to each other, as are the bottom teams, with a huge separation between the two groups because of the heavily weighted build log. At the national level, where the top will most likely have nearly perfect scores with little to break ties, it may as well come down to luck. Granted, the only environmental factor that has been tested thus far has been temperature, but with an entire season to work on the detector, I doubt there would be anything preventing teams to nearly perfect every other factor.

My idea: make this event sort of a Swiss Army Knife do everything type device
Similar to how many events have topics that will only be tested at the State and National level, there would be different environmental factors that would be tested at each level. Not every factor will be tested each time, so, for example, and ES at one tournament could choose to test temperature and salinity, whereas another ES at another tournament could choose to test air pressure and wind speed. With this type of set up, competitors would have to prepare for every type of environmental condition that could be tested, but it is not guaranteed that all will be tested at once. While I could see cost getting in the way of this, it could be worked around by making the ones that are more likely to cost more only tested at the State or National level. By having the event set up this way, there would be much more to separate teams, and it would require teams to find more efficient ways to calibrate their devices and complete the tiebreaker quiz with so many things to test.

I hope the event committees will listen to people's suggestions for this event. Detector has the potential to be a great event provided that the rules are set out well.
No no no no no no no no please no no no no no no that's gonna take forever to build
If an event is replacing Mission, it kind of deserves to be tough.
Userpage
Chattahoochee High School Class of 2018
Georgia Tech Class of 2022

Opinions expressed on this site are not official; the only place for official rules changes and FAQs is soinc.org.

User avatar
pepperonipi
Wiki Moderator
Wiki Moderator
Posts: 153
Joined: January 21st, 2019, 11:38 am
Division: C
State: FL
Location: adding wiki templates lol
Has thanked: 36 times
Been thanked: 76 times

Re: Detector Building C [TRIAL]

Post by pepperonipi » June 7th, 2019, 7:19 pm

Umaroth wrote:I competed in this event when it was a Division B trial event at SoCal States 2018, and I do have some thoughts on how it could be improved based on last year and what I heard this year.

The number one issue with the event in its current state is the lack of separation between the top and bottom teams. From what I can tell whenever this event is run, the top teams are all extremely close to each other, as are the bottom teams, with a huge separation between the two groups because of the heavily weighted build log. At the national level, where the top will most likely have nearly perfect scores with little to break ties, it may as well come down to luck. Granted, the only environmental factor that has been tested thus far has been temperature, but with an entire season to work on the detector, I doubt there would be anything preventing teams to nearly perfect every other factor.

My idea: make this event sort of a Swiss Army Knife do everything type device
Similar to how many events have topics that will only be tested at the State and National level, there would be different environmental factors that would be tested at each level. Not every factor will be tested each time, so, for example, and ES at one tournament could choose to test temperature and salinity, whereas another ES at another tournament could choose to test air pressure and wind speed. With this type of set up, competitors would have to prepare for every type of environmental condition that could be tested, but it is not guaranteed that all will be tested at once. While I could see cost getting in the way of this, it could be worked around by making the ones that are more likely to cost more only tested at the State or National level. By having the event set up this way, there would be much more to separate teams, and it would require teams to find more efficient ways to calibrate their devices and complete the tiebreaker quiz with so many things to test.

I hope the event committees will listen to people's suggestions for this event. Detector has the potential to be a great event provided that the rules are set out well.
Definitely agree. I heard lots and lots of teams say that they got a perfect score in the device portion of the event when I was in the testing room at Nationals, which isn't really how an event should be.

Couldn't the rules also be changed so that there are more building restrictions? I think this would also make the event harder without adding too much cost for other sensors/parts in a Swiss Army Knife-like device. This could include restrictions such as: having to use certain parts (such as a third-party ADC) in the device's setup, requiring the device to fit within a very small size (such as within a 10-15cm cube), or requiring the device to be contained inside a waterproof case. There were already some this year which added some challenge to the event (such as requiring LED's or making sure the probe could withstand boiling temperatures for a set period of time), and I believe that adding more would just increase the difficulty of the event.
who made rainbows so pretty? 🥺🌈🌈🥺

Boca Raton Community High School
My Wiki Page | School Wiki Page | WikiProject Tournaments of States | Pi-Bot

2019: Code, Fermi, Thermo
2019 Trials: WTF (MIT), Detector (Nationals)
2020: Detector, Orni, Code (Substitution: Penn)
2020 Trials: Data Sci (Princeton)

User avatar
Umaroth
Member
Member
Posts: 222
Joined: February 10th, 2018, 8:51 pm
Division: C
State: CA
Location: Pretend hugging SIlverBreeze because social distancing
Has thanked: 3 times
Been thanked: 14 times

Re: Detector Building C [TRIAL]

Post by Umaroth » June 7th, 2019, 8:03 pm

pepperonipi wrote:
Umaroth wrote:I competed in this event when it was a Division B trial event at SoCal States 2018, and I do have some thoughts on how it could be improved based on last year and what I heard this year.

The number one issue with the event in its current state is the lack of separation between the top and bottom teams. From what I can tell whenever this event is run, the top teams are all extremely close to each other, as are the bottom teams, with a huge separation between the two groups because of the heavily weighted build log. At the national level, where the top will most likely have nearly perfect scores with little to break ties, it may as well come down to luck. Granted, the only environmental factor that has been tested thus far has been temperature, but with an entire season to work on the detector, I doubt there would be anything preventing teams to nearly perfect every other factor.

My idea: make this event sort of a Swiss Army Knife do everything type device
Similar to how many events have topics that will only be tested at the State and National level, there would be different environmental factors that would be tested at each level. Not every factor will be tested each time, so, for example, and ES at one tournament could choose to test temperature and salinity, whereas another ES at another tournament could choose to test air pressure and wind speed. With this type of set up, competitors would have to prepare for every type of environmental condition that could be tested, but it is not guaranteed that all will be tested at once. While I could see cost getting in the way of this, it could be worked around by making the ones that are more likely to cost more only tested at the State or National level. By having the event set up this way, there would be much more to separate teams, and it would require teams to find more efficient ways to calibrate their devices and complete the tiebreaker quiz with so many things to test.

I hope the event committees will listen to people's suggestions for this event. Detector has the potential to be a great event provided that the rules are set out well.
Definitely agree. I heard lots and lots of teams say that they got a perfect score in the device portion of the event when I was in the testing room at Nationals, which isn't really how an event should be.

Couldn't the rules also be changed so that there are more building restrictions? I think this would also make the event harder without adding too much cost for other sensors/parts in a Swiss Army Knife-like device. This could include restrictions such as: having to use certain parts (such as a third-party ADC) in the device's setup, requiring the device to fit within a very small size (such as within a 10-15cm cube), or requiring the device to be contained inside a waterproof case. There were already some this year which added some challenge to the event (such as requiring LED's or making sure the probe could withstand boiling temperatures for a set period of time), and I believe that adding more would just increase the difficulty of the event.
A size bonus could also be a good idea, but not sure how it could be implemented well
Troy SciOly 2019-now
The Ambassador

Proud Kraemer SciOly Padre 2018-now
Dank Memes Area Homeschool Juggernaut 2018-now
Kraemer SciOly Pretty-Much Head Coach 2019-now
2020 Tryouts: Circuit, Code, Detector, DP, GeoMaps, Machines
States Events: HANDS ON DISEASE DETECTIVES

Umaroth's Userpage

chessbucket
Member
Member
Posts: 23
Joined: September 5th, 2018, 4:34 pm
Has thanked: 0
Been thanked: 0

Re: Detector Building C [TRIAL]

Post by chessbucket » July 7th, 2019, 2:02 pm

Has anyone tried a Kalman Filter or nueral network?
2020: MIT Gravity Vehicle 2nd
2020: MIT Machines 6th

2019: Nationals Mission Possible 2nd
2019: NC Mission Possible 2nd
2019: MIT Mission Possible 4th

2018: NC Mission Possible 1st

shrewdPanther46
Member
Member
Posts: 437
Joined: October 9th, 2017, 6:25 pm
Division: C
State: NJ
Has thanked: 0
Been thanked: 0

Re: Detector Building C [TRIAL]

Post by shrewdPanther46 » July 8th, 2019, 4:29 pm

chessbucket wrote:Has anyone tried a Kalman Filter or nueral network?
highly doubt it

Post Reply

Return to “2019 Build Events”

Who is online

Users browsing this forum: No registered users and 1 guest