[robocup-msrs] CLARIFICATION: changing the number of Robudogplayers inthesimulator

[Tucker Balch]

Dear Nikos, Kyle,

To be fair to Kyle, I believe he received this task only a few weeks 
ago, and I must say he's handling things quite well :-)

In fact, we're all somewhat late getting started on this -- it is not 
Microsoft's fault, it is just that it took some time to decide for sure 
that we were doing this, and to get everything in place.

Having said all that, I would suggest that we go forward with a plan 
something like:

Sunday - Weds July 1-4, work together to get teams playing versus teams 
-- let's avoid penalties for code not integrating, but focus on getting 
it to work.  Once we get there we'll be able to assess how good the 
teams are and so forth.

Thurs-Friday, July 5-6, Competitions of some sort -- could be just skill 
demonstrations, could be games.

Sat-Sun, July 7-8, Demonstrations at CRC (e.g. we'll pick "best of" 
examples and run them).

Afterwards -- lets create an after-action report to summarize advice to 
Microsoft.

Tucker


Nikos Vlassis wrote:
> Hi Kyle,
> 
> In my view the primary feature of a simulator is not the visualization
> but the cheap collection of data for experimentation and testing. (In
> my previous emails I've asked questions towards this: how to switch
> off visualization, how to programmatically set/get the position of
> entities in the field, etc.) I believe that if the MSRS soccer
> framework wants to be a standard RoboCup framework, it will be
> critical to have a fully programmatic development environment that
> doesn't rely at all on visualization, and where testing will be cheap
> and easy. I know this is not possible to do in a week time, and in
> fact it's useful for us to experiment with low-level MSRS issues, but
> I'm not so sure how sophisticated soccer games you should expect from
> the teams when there are still low-level technical issues that are
> hard to master.
> 
> Perhaps the above applies only to our team :-) In any case, let me
> take the freedom to raise the question: How far are the other teams
> with the development of their code? Are there teams that can already
> play a decent game of soccer? (In our team we are far from that.) From
> the discussion in the mailing list I have noticed that we are all
> still struggling with low-level details like configuration files, etc.
> How many teams are finally participating? Would it make more sense to
> have some kind of demonstration of basic soccer skills instead of
> expecting us to develop a complete soccer team by next week, while we
> are still waiting software patches to arrive this week?
> 
> Please take the above as constructive criticism that only tries to help things.
> 
> cheers,
> Nikos
> 
> 
> On 6/25/07, Kyle Johns <kylej at microsoft.com> wrote:
>> Hi Nikos (and others working on this issue),
>>
>> Many of the robudog services are started programmatically in Robosoft's code rather than from the manifest.  I don't see a way to easily disable these services from starting up without modifying Robosoft's code.  The entities are also created and placed by Robosoft's services independently of the manifest.
>>
>> It is fairly easy to delete the robot entities manually.  As I mentioned in a previous post, you just go into the simulation editor (F5), select the entity you want to delete in the upper-left pane, and press Delete.  I realize that this is painful to do every time you start up the environment but it is much simpler than the programmatic way of doing it.
>>
>> Is it a problem for you that the services for the other players are loaded?  If you delete the entities, the services won't be doing anything in the simulation environment that will affect the remaining player.  It will be as if they are not running at all.
>>
>> Is it critical for you to have a programmatic way to delete the entities?
>>
>> -Kyle
>>
>> -----Original Message-----
>> From: robocup-msrs-bounces at cc.gatech.edu [mailto:robocup-msrs-bounces at cc.gatech.edu] On Behalf Of Nikos Vlassis
>> Sent: Monday, June 25, 2007 10:39 AM
>> To: Andrew Williams
>> Cc: Whitney O'Banner; robocup-msrs at cc.gatech.edu
>> Subject: Re: [robocup-msrs] CLARIFICATION: changing the number of Robudog players inthesimulator
>>
>> To summarize the discussion: The issue of how to start up only one
>> robot (so that all other services and simulation entities don't start
>> up) is still unresolved. Commenting out code in the manifests doesn't
>> seem to suffice. Some help here would be welcome.
>>
>> thanks
>> Nikos
>> _______________________________________________
>> robocup-msrs mailing list
>> robocup-msrs at cc.gatech.edu
>> https://lists.cc.gatech.edu/mailman/listinfo/robocup-msrs
>>
> _______________________________________________
> robocup-msrs mailing list
> robocup-msrs at cc.gatech.edu
> https://lists.cc.gatech.edu/mailman/listinfo/robocup-msrs

-- 
Tucker Balch, Associate Professor, Interactive Computing
Director, Institute for Personal Robots in Education
Chair, RoboCup 2007 Atlanta
Georgia Institute of Technology

phone: +1 404 385-2861 phax: +1 404 601-3185
http://www.cc.gatech.edu/~tucker