2017 Score Sheet Design

[Peter Law]

Hi,

I've started thinking about scoring design for the 2017 game and have
created a candidate scoresheet in the usual arena.git repo [1];
there's a render attached.

It's currently a variant of last year's, keeping the same idea of
noting a token type using a letter. I think this worked well and
avoids filling up the sheet with many separate entry fields per token
type.

I'll adapt the scorer and create the proper compstate repo (with
scoring script) shortly.

Thanks,
Peter

[1] https://www.studentrobotics.org/cgit/arena.git/tree/comp17/

[Andrew Barrett-Sprot]

Hi Peter, 

The worst case is having 9 tokens in a single zone. I'd say the current 'Token' spaces are a bit tight for 9 clearly written capitals

I suggest making the areas larger, and maybe changing from a single constant underline to 9 spaced underscores (_________ to _ _ _ _ _ _ _ _ _), to force linesmen to space the letters legibly. It also makes it easier for them to check they've positioned all markers by just counting the spaces.

Thanks,

Andy B-S

--
You received this message because you are subscribed to the Google Groups "Student Robotics Development" group.
To unsubscribe from this group and stop receiving emails from it, send an email to srobo-devel...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Peter Law]

Hi Andy,

Thanks for looking at the design!

> The worst case is having 9 tokens in a single zone. I'd say the current
> 'Token' spaces are a bit tight for 9 clearly written capitals

I agree it's probably a bit tight for 9 letters, though would point
out that we've had these lines as they are for the last two years and
it hasn't been an issue (last year also had nine tokens IIRC; I can't
remember how many there were the year before). While I'd love to see
it, I doubt that any one team will manage to collect all nine tokens.

> I suggest making the areas larger, and maybe changing from a single constant
> underline to 9 spaced underscores (_________ to _ _ _ _ _ _ _ _ _), to force
> linesmen to space the letters legibly.

I agree that this hints that the linesmen should aim for one letter
per line, though also wonder whether that would result in them
squishing their letters to fit. IMO squishing is more likely with the
spaces than without. (Also, humans are usually capable of drawing
outside the lines if they run out of space).

I will definitely look at making the lines slightly longer though.

> It also makes it easier for them to
> check they've positioned all markers by just counting the spaces.

I'm not sure I follow. There will be 9 markers, but 36 "gaps" on a
completed scoresheet -- surely counting to 9 is easier?

Thanks,
Peter

[Andrew Barrett-Sprot]

Hi Peter,

On Tue, Jan 3, 2017 at 9:52 PM Peter Law <peter...@gmail.com> wrote:

> It also makes it easier for them to
> check they've positioned all markers by just counting the spaces.

I'm not sure I follow. There will be 9 markers, but 36 "gaps" on a
completed scoresheet -- surely counting to 9 is easier?

I was thinking in the (very likely) case that only 1 or 2 tokens are removed from the centre zone, the linesman can double check by counting the 1 or 2 blank spaces in the centre scoring zone, which would be faster and less error-prone than counting the filled spaces.

A A A B B B C _ _

is easier to count than 

A A A B B B C    

if you know there are 9 underlines

We're getting close to bikeshedding territory here though. the fact is the scoring sheet is already pretty good.

Thanks for handling this, by the way

Thanks,

Andy B-S

[Peter Law]

Andy wrote:
> I was thinking in the (very likely) case that only 1 or 2 tokens are removed
> from the centre zone, the linesman can double check by counting the 1 or 2
> blank spaces in the centre scoring zone, which would be faster and less
> error-prone than counting the filled spaces.
>
> A A A B B B C _ _
> is easier to count than
> A A A B B B C
> if you know there are 9 underlines

Ah, I see.
This somewhat relies on knowing that there are the right number of
slots, and feels rather convoluted to actually be useful. If anyone
has the capacity to user-test this I'd be interested in the results,
but I think this is super low priority really.

> We're getting close to bikeshedding territory here though. the fact is the
> scoring sheet is already pretty good.

I agree.

I previously wrote:
> I will definitely look at making the lines slightly longer though.

I had a look and found that making the lines noticeably longer results
in things getting somewhat squashed.

For both of the above I'm going just leave the form as it is for now.

Thanks,
Peter