Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   General Forum (http://www.chiefdelphi.com/forums/forumdisplay.php?f=16)
-   -   Terrifying Karthik (http://www.chiefdelphi.com/forums/showthread.php?t=143460)

AdamHeard 10-02-2016 15:56

Re: Terrifying Karthik
 
Quote:

Originally Posted by Rangel(kf7fdb) (Post 1538149)
Different poll results on other threads also show 254 being screwed thanks to the Rock Wall change. I wouldn't quite factor a poll as a definite source. That being said, I think ignoring the low bar is a legitimate strategy/design. We probably would have done it too had we not had solutions to our strategy that allow us to remain short.

254 was pretty screwed by the rock wall change, cost them over a week I heard.

Had to switch from regular swerve to tank tread swerve.

BL0X3R 10-02-2016 16:21

Re: Terrifying Karthik
 
I think karthik is terrified because too many teams will be willingly sacrificing too much in order to be low-bar compliant. There might be teams that can only build a box on wheels plus a low-quality intake because of the constraints of the low bar.

Now don't get me wrong - a high quality, well driven box on wheels can do OK as a low first or second pick this year, but if an alliance is limited only to breaching then they shouldn't expect to make it to the semifinals. Likewise, an alliance without a shooter may be locked out of the finals completely, depending on the event.

That is what karthik is scared of - the amount of teams that think that a low-bar-traversing box-on-wheels will be good enough because of the value placed on defenses.

GaryVoshol 10-02-2016 16:23

Re: Terrifying Karthik
 
Quote:

Originally Posted by SM987 (Post 1537779)
//image clipped//

He meant to say "The results of this pole are terrifying."

To add insult to injury, I'm sure that's over 15 inches. FOUL!

Rangel(kf7fdb) 10-02-2016 16:23

Re: Terrifying Karthik
 
Quote:

Originally Posted by AdamHeard (Post 1538155)
254 was pretty screwed by the rock wall change, cost them over a week I heard.

Had to switch from regular swerve to twerk drive.

Fixed that for you. :]

XaulZan11 10-02-2016 16:32

Re: Terrifying Karthik
 
Quote:

Originally Posted by AdamHeard (Post 1538155)
254 was pretty screwed by the rock wall change, cost them over a week I heard.

Had to switch from regular swerve to tank tread swerve.

Sounds brutal. I hope they will have enough weight for their scissor lift climber...

Alex Chamberlin 10-02-2016 17:30

Re: Terrifying Karthik
 
My 2 cents:

The scouring potential of a good low bar bot significantly exceeds that of the non low bar bot causing the high ratio of low bar to other. That's why he believes the poll.

He thinks that if 90 % are low bar the best low bar will make runs while the other to sit around feeling useless.

Thats the gloomiest out look but terrifying is not the word I think he would choose.

So... more on the word terrifying.

It sounds like a low bar bot was a counter to some strategy he envisioned or his team worked off of.

Ginger Power 10-02-2016 17:39

Re: Terrifying Karthik
 
For those who are saying that Karthik may not have accounted for something, or that the high percentage of low bar bots will ruin his strategy... I highly doubt that. I don't know Karthik, but from listening to his speeches I am willing to bet he has a plan for just about everything. He probably knows the meta of the meta-game better than most people understand the basic aspects of the game.

It has been said multiple times, but I firmly believe that Karthik is concerned about other teams and their strategic decisions. It is strategically beneficial for many teams to ignore the low bar, and they're not doing that.

Obviously this is all speculation. I can't wait to hear what Karthik has to say on the matter at Champs.

evanperryg 10-02-2016 17:48

Re: Terrifying Karthik
 
Quote:

Originally Posted by Alex Chamberlin (Post 1538211)
My 2 cents:

The scouring potential of a good low bar bot significantly exceeds that of the non low bar bot causing the high ratio of low bar to other. He thinks that if 90 % are low bar the best low bar will make runs while the other to sit around feeling useless.

I doubt that a 3 cycler alliance, at least at low to middle levels of play, would be able to cycle fast enough for the low bar ability to become a significant advantage. Keep in mind a strong alliance will need both breach and tower points, so these teams will have to go through the other defenses for most of the match anyway, nullifying much of the low bar cycling strength.

Quote:

Originally Posted by AdamHeard (Post 1538155)
Had to switch from regular swerve to tank tread swerve.

I heard they also had to rework the drop-down mecanums, too.

GeeTwo 10-02-2016 17:54

Re: Terrifying Karthik
 
Quote:

Originally Posted by AdamHeard (Post 1538155)
254 was pretty screwed by the rock wall change, cost them over a week I heard.

Had to switch from regular swerve to tank tread swerve.

Quote:

Originally Posted by evanperryg (Post 1538218)
I doubt that a 3 cycler alliance, at least at low to middle levels of play, would be able to cycle fast enough for the low bar ability to become a significant advantage. Keep in mind a strong alliance will need both breach and tower points, so these teams will have to go through the other defenses for most of the match anyway, nullifying much of the low bar cycling strength.


I heard they also had to rework the drop-down mecanums, too.

Nonsense. They just had to add another couple of percent to the duration they ran the four big fans.:p

I don't know about Karthik, but the main thing that would be terrifying to me is if a significant percentage of robots are planning to only cross the outer works at the low bar. This would lead to easy defense, significant traffic jams, and the category C defenses (drawbridge and sally port) being in position 5 most matches.

jweston 10-02-2016 18:00

Re: Terrifying Karthik
 
My biggest concern is how low bar teams did their cost-benefit analysis. The poll suggests many teams felt there will be a huge speed pay off to going under the low bar. This speed pay off has to be high enough to outweigh the compromises made to effectiveness at goal shooting, breaching, climbing, and blocking.

The special thing these low bar bots bring to the table is the ability to ferry balls into the courtyard quickly, as well as breach the low bar. But points are scored on scoring goals, breaching, and climbing, not ferrying balls.

It gets interesting when you have so many teams making the same types of trade-offs. The more teams doing the low bar, the less of a premium there is to it. In the meantime, they must live with the design compromises made in order to make it under the low bar. This makes it more likely that there will be a premium on bots that can do well with goal shooting, other types of breaching, climbing and blocking.

Getting under the low bar is nice for an alliance to have, but it's not a killer feature. An alliance can score in all ways without a low bar bot. I'm not as sure that an all-low-bar alliance is as likely to be equally effective. There's just too many other aspects of the game that a low bar bot team might have underestimated in value or in difficulty. Time will tell.

evanperryg 10-02-2016 18:06

Re: Terrifying Karthik
 
Quote:

Originally Posted by GeeTwo (Post 1538223)
I don't know about Karthik, but the main thing that would be terrifying to me is if a significant percentage of robots are planning to only cross the outer works at the low bar. This would lead to easy defense, significant traffic jams, and the category C defenses (drawbridge and sally port) being in position 5 most matches.

This is a really good point. I hadn't even considered the number of teams making dedicated low bar shooters... I think it's time for a new poll.

EDIT: Here's the poll. Results are already a little frightening.

Chris is me 10-02-2016 18:21

Re: Terrifying Karthik
 
Quote:

Originally Posted by jweston (Post 1538224)
The special thing these low bar bots bring to the table is the ability to ferry balls into the courtyard quickly, as well as breach the low bar. But points are scored on scoring goals, breaching, and climbing, not ferrying balls.

You're overlooking a very big part of the game, and ultimately what pushed many teams toward a low bar oriented strategy. You need 8 balls scored to capture. If you want to reliably get that ranking point, especially at early events, you will want to do that by ferrying balls into the low goal. The low bar is probably the only reasonable shot of a single robot trying to stuff anywhere close to 8 balls in that goal every match. Trying to get an 8 ball throughput, even spread across the entire alliance, will require going to the secret passage early and often, and there's no faster way to it than the low bar.

If one of the ranking points was not predicated on scoring a high volume of balls I do not think the low bar would have been such a high priority for so many teams. But maybe I'm overestimating the depths of team strategic analysis, and people just decided to try and do everything.

Kingland093 10-02-2016 19:02

Re: Terrifying Karthik
 
Quote:

Originally Posted by jweston (Post 1538224)

The special thing these low bar bots bring to the table is the ability to ferry balls into the courtyard quickly, as well as breach the low bar. But points are scored on scoring goals, breaching, and climbing, not ferrying balls.

This is what we are trying to do. We always try to go for the supporting role as we just don't have the resources or the capabilities to be an all out offense robot. Unfortunately, I think a lot of teams this year have picked up on that strategy. (i suppose that's good for tall shooter robots who will have a lot of options come alliance selections)

I'm predicting that in the first few weeks of regionals, there will be several winning alliances that cheescaked a tall blocker onto one of the small support robots that will shut down the alliances of small shooters.

BotDesigner 10-02-2016 19:25

Re: Terrifying Karthik
 
My team is building a low bar low goal robot. We also have a 3ft long arm we use for defense manipulation as well as scaling. After seeing the polls about a week ago, we decided to mount a large plastic sheet to the arm for blocking shots. With all the short shooters I have a feeling we will mainly be using the arm for defense.

I don't know why Karthik is terrified, but my team could not be happier about the results of these polls.

jweston 10-02-2016 20:04

Re: Terrifying Karthik
 
Quote:

Originally Posted by Chris is me (Post 1538237)
You're overlooking a very big part of the game, and ultimately what pushed many teams toward a low bar oriented strategy. You need 8 balls scored to capture. If you want to reliably get that ranking point, especially at early events, you will want to do that by ferrying balls into the low goal. The low bar is probably the only reasonable shot of a single robot trying to stuff anywhere close to 8 balls in that goal every match. Trying to get an 8 ball throughput, even spread across the entire alliance, will require going to the secret passage early and often, and there's no faster way to it than the low bar.

I had taken the tower strength into account. There is 15s in auto and 135s in teleop. Let's say you can round trip a shot in 15s, probably faster doing the low bar. If you plan to capture or climb in the last 20s (i.e. you're going for that capture rank point), you can still get 8 shots in between auto and teleop. More shots are better of course, especially if you miss. My concern is if a design compromises the shooting accuracy even if it is low goal, more round trips are required to get the same effectiveness, potentially making the extra speed a wash.

Quote:

Originally Posted by Chris is me (Post 1538237)
If one of the ranking points was not predicated on scoring a high volume of balls I do not think the low bar would have been such a high priority for so many teams. But maybe I'm overestimating the depths of team strategic analysis, and people just decided to try and do everything.

Winning gets you two rank points. Breaching also gets a ranking point. The low bar gets you up to 10 points, it works as one of the four defenses needed to breach, and it cycles the ball faster. It's a nice piece to something bigger. If reality ends up reflecting this poll, low bar capable bots will be a very plentiful resource.

Of course, if a team can pull off the low bar and excellent shooting and/or a variety of defense breaching, that will be a very effective bot indeed.


All times are GMT -5. The time now is 17:03.

Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi