Concerns With Rule T1b

I completely agree with this.

By the way, this scenario would not warrant a DQ as you can only be QDed for SG3c violations if they occur in the last 30 seconds.

For events that already have a livestream, we wouldn’t be spending any additional resources. It is completely free. We are not asking for it to be required for video replays to be available at all events, just that if the capability is there, the referee should be allowed to look at the video if hey want to help them make decisions.

Some events have GPS strips on comp fields, some don’t. Some events have judges with robotics experience or industry experience or whatever, some get random people. Some events have experienced referees, some don’t. Some events have 10 qualifier matches, some have 6.

The goal should be to make each event as good as you can with the resources you have access to, not to limit the quality of all events because something isn’t possible at every event.

(Also, side note, having a nonprofit that owns the AV equipment that’s used at lots of different events run by different EPs is a great way to increase the quality of events while keeping costs shared and manageable)

If the problem statement Video Review is trying to address is:

As a competitor, I am concerned that the referees may not always see rules violations which leads me to believe that match outcomes are not always reflective of team skill

Then perhaps a more universal solution would be adding a clause to <G20>

b. . Match Affecting game rule issues.
i. Head Referee disables a Robot for a misinterpretation of a rule violation.
ii. Head Referee starts the Driver Controlled Period of the Match without determining the outcome of the Autonomous Period winner.
iii. The field is reset before a score is determined.

Add a fourth bullet point for something akin to (with wordsmithing - this is admittedly quite hard!):

iv. Head Referee determines that there is a potential rules violation for which they are unable to determine the application based on their observation of the match

Now we have a potential solution that works regardless of a venue’s technology infrastructure. Does this potentially take longer than what @AperatureLabs presents as the ideal workflow for VR? Yes.

If the ring is put there before the last 30 seconds, and then touches the platform in the last 30 seconds, does it not count?

It is not a violation.

Points “a”, “b”, and “c” apply to Robot actions during the last thirty (30) seconds of the Match:

Point “d” applies to Robot actions prior to the last thirty (30) seconds of the Match:
d. Placing a Mobile Goal on or under the opposing Alliance’s Platform, at any point during the Match is
considered a minor violation of this rule that, at a minimum, will result in a warning.

Sorry to revive the thread, but This happened at the competition I was at

We got disqualified because the opposing alliance said they were trying to balance a MOGO on the platform but we were in the way. The refs didn’t see the whole situation, so it came down to our word vs there’s. If they could see the replay they would see that they didn’t have a MOGO and didn’t pursue a MOGO untill 20 seconds after. The event was streamed, and this is their video, and it could have assisted a lot in the ruling.

This was a qualification match, so it really doesn’t matter, but is a good example of where it can help.

I am confused. I didn’t see any rule violations in this match. Your robot didn’t event enter the opponents home zone; did you by chance upload the wrong video?

oops, yes

should be fixed now

You were.

https://www.robotevents.com/VRC/2021-2022/QA/880

No they weren’t.

The picture clearly shows that the red alliance has no goals in their possession, they aren’t actively trying to score point.

With this thread has been brought back up and the February game manual update approaching, I was wondering if anyone from the GDC (@Jon_Jack) would be willing to say what they think about the stuff that has been said so far in this thread.

would look that there is physical contact by Blue and contacting Red (irregardless of possession of mobile goal possessions - interfering) and Red prevent from movement with other Blue team member in contact… … and mobile goals near platform that would could swing score …

Seriously - this is not a great example still image to illustrate situation in last 30 seconds - if red was moving and blue is all contact with both robot and platform …

Now ask me why I did not watch video - why? match is over.

No need to redo T1B

This is not an official answer/response, but there are somethings to consider regarding any changes to T1b…

The world’s major sports leagues (MLB/NFL/NBA/NHL/EPL/College Football) have, in my opinion, skewed the perception of how easy it is to do video replay.

  1. These are multi-billion dollar leagues with endless technology resources and sponsorships backing them.
  2. The TV contracts for these league (usually in the 9 or 10 figure range) require broadcasts with multiple 4K (or higher) camera angles. These camera angles allows whomever is reviewing the video to see a play from different angles. This may not seem as important, but I want you to think about reviewing a play in a game from one of these aforementioned leagues, while only using a full field camera angle. It’s not the same. Even when they show replays on TV, they’re cutting to multiple up close camera angles.
  3. The dynamics of the games played in these sports league are almost never the same dynamics seen in robotics competitions. In major sports leagues, the play focuses around a single game object and the play mostly follows that singular game object. In robot games, there are usually multiple game objects with multiple ways of scoring. This creates a contextualization nightmare.

Yep, we’re a robotics competition. Yep, we deal with technology. However, there is a limit to what you can ask EPs to do. Expecting every EP to have multiple high resolution cameras setup to an exact angle/areas of the field is, and streaming/archiving those feeds so they can be pulled up at a moment’s notice is, in my opinion, unreasonable. Even more so if it’s only going to be used during elims. Remember, EPs are pretty much volunteers putting on events with limited budgets.

If you think video replay won’t slow down events, consider this: Video replay was introduced to MLB in 2008, the average game time was 2:50. Since then, the average game time has climbed to 3:10 in 2021. There are some other factors involved here, but it still slows down the pace of the game. Other leagues have seen increases in average game times since they introduced video replay as well.

We actually piloted video review during Live Remote Worlds. Makes sense, right? We have video of every match, the video was archived on YouTube, easy… However, it was actually extremely difficult and time consuming to pull off. In my 10+ years of reffing events and handling disputes at VEX Worlds, I can say reviewing video was one of the toughest things to do. There are several reasons for this:

  • In general, match video is pretty crappy and I’m not just referring to the resolution either. Just the viewing angles of a lot of the videos made this difficult. Parts of the field weren’t visible. Drivers would get in the way. It was hard to discern how many objects were in a goal (even with CU’s relatively large objects).
  • As I mentioned above, you’re trying to made a judgement call with full field video. There’s no going to a different camera angle. There’s no zooming in without losing much clarity.
  • With Live Remote Worlds we were only reviewing scoring since there was no robot to robot interaction. However, even gauging if a robot was contacting an object was extremely difficult.
  • Each match took longer than you would expect to review - even if we were only checking scoring in auto for a specific match.

These are just some of the logistical concerns that have to be overcome. I would also like to point out that video review is barely available in high school athletics.

Lastly, sometimes it’s important to remember the big picture… The mission of the REC Foundation is:

The Robotics Education & Competition (REC) Foundation’s mission is to increase student interest and involvement in science, technology, engineering, and mathematics by engaging students in hands-on, affordable, and sustainable robotics engineering programs.

Nowhere in their mission statement does it mention finding the best robot in the world.

I almost feel like this makes the ability to have a video replay available more important. You have three referees with a limited amount of training expected to watch four robots without focus on a single game object. Especially with rules like SG3, it’s almost impossible for a normal relatively untrained referee to be able to get any degree of context for a violation in the heat of the moment, leading to either under penalizing or over penalizing because they lack the information required to make a good call. The number of times at the tournaments I’ve helped run this season the entire event has screeched to a halt for ten minutes because nobody can agree what just happened is absurd.

I’d also like to point out that it’s somewhat unfair to compare an event that happens in a 12’ x 12’ square to an event on a 360’ x 160’ field. At a state championship football game there’s gonna be multiple camera angles and live camera operators following the action because you need to move the camera to follow what’s happening. In VEX you can watch the match just fine from a single fixed camera angle and have a good idea what’s happening because the field is so small relative the things happening.

The full field view of a VEX match has a higher signal to noise ratio than even a moderately zoomed in view of a football field showing only about 15 yards of the field.

I think you’re missing my point. I wasn’t arguing that it was useless to have video replay. My point is that there are a number of technical and logistical issues that make video replay in our sport extremely difficult.

My point about multiple camera angles was that it’s easier to setup multiple camera angles when you really only need to focus on one game object moving around a field vs. multiple game objects/players scoring simultaneously. In addition, most cameras shooting a pro sports event have a camera operator behind them. This allows them to focus the camera on what is happening during the game.

Here’s an example of what I’m talking about from the NFL and NHL (sorry if there are any Cowboys fans reading). Notice how many angles whoever is doing the video review can use to determine if a call should be changed. Video review isn’t done with a single full field angle.

In fact, in recent years, professional sports leagues have moved to placing additional replay cameras inside the field of play (pylon cameras in the NFL, overhead goal cameras in the NHL, base cams in MLB, etc). These give reviewers even more angles to closely examine the play and make the right call.

I don’t disagree that a single camera is fine for spectating a VRC match. However, there are details that you just can’t pick up if you’re not there and/or don’t have a second perspective on what’s going on. Having done VRC video review at Live Remote Worlds last year where we only had a single camera angle, I can say that it’s extremely difficult to confidently change close calls without a second (or third) angle to review.

Again, my point isn’t that it’s useless. My point is that there are significant obstacles that need to be overcome before anything like this could be a reality.

Full disclaimer here, I have never been a ref, so take my opinion with a grain of salt.

It seems to me like for most (not all, but most) situations can be seen with only a single camera. It seems like most of the things needing to be reviewed this year are < SG3 > and pinning/trapping violations, both of which can be seen from the camera angle that Alec posted.

Now I will agree with Jon on the timing issue. Events run long and video replay would only make them run longer. I would propose having video replay at worlds qualifying events, and not allowing it for other events where the stakes may not be as high. Or perhaps you could say replay was only allowed during the elimination matches…

Just my thoughts, take it with a grain of salt…

Let’s pretend Tipping Point has a possession limit on the number of Rings your Robot can hold. One of the robots in the top right corner of the image you posted above is possessing more than they’re allowed, and this happens to be a match affecting violation.

Can you tell me, with 100% certainty (and don’t forget, within a reasonable amount of time, because this decision is now holding up the rest of the event) exactly how many rings the two robots in the top right corner are holding?

Time to start counting the number of visible rings :smiley:

Edit after counting all the rings I could see(and messing up at least twice) those robots should be holding 24 rings

The real point is that Video Review is a solution in search of a problem.

Define the problem first. Then consider many solutions and weigh the costs/benefits of each. Some may overlap.

If the problem is (and I may not be correctly identifying the problem):

As a competitor, I am concerned that the referees may not always see rules violations which leads me to believe that match outcomes are not always reflective of team skill

Then VR may improve this. There may be other, less technologically dependent solutions as well:

  1. Allow head refs more discretion in replaying matches
  2. Factor in referees ability to detect rules violations when designing the game
  3. Encourage and enable EPs to have multiple referees at each field

And probably several more others can think of.

Zero. They are holding zero.

Now, I assume you also mean rings in the corner that access is being blocked to and therefore “possessing”. Firstly, the red robot is being pinned by the blue robot which means that if a violation is occurring it’s a forced violation, so no penalty should be incurred.

But let’s say they’re on the same alliance. Then the answer is around ten. Which, unless I’m forgetting the rules of a game, is more than enough to DQ under any previous possession rule. And frankly, if you have designed a game where the possession limit is ten, no live referee is going to be able to count that quickly and determine whether it’s 9 which is totally fine, or 11 which is a DQ, so the problem is that the game rules are impossible to enforce.