Matrix Software
Moderator: hbartel
- Ed Kettler
- Posts: 3437
- Joined: Sat Dec 15, 2001 6:05 pm
- Location: USA
- Contact:
Matrix Software
There are two software packages that are on the Workshop page: MatrixGen and Matrix Maker. MatrixGen seems to have some challenges on getting a valid matrix generated, and Matrix Maker does not have some features that would be useful for CDs and scorekeepers.
Randy Hodges has identified a computing person who could create an updated version of Matrix Maker that incorporates additional features, if we can identify and prioritize them. The purposes of this thread is to provide feedback to Randy to pass to the programmer.
My list of desired features include
1. Allowance for additional criteria to establish 'soft conflicts' to spread folks up. Example: State or AMA district. The hard conflicts on frequency would be resolved, then a second pass would spread pilots from TX across the heats. Field is optional.
2. Print preview
3. Output to Excel spreadsheet with formatted summary scoresheet to assist the scorekeeper
4. Matrix analysis - ability to calculate summary of pilot vs pilot alignment to make sure we have a balanced contest.
Please put your comments here.
Thank you
Randy Hodges has identified a computing person who could create an updated version of Matrix Maker that incorporates additional features, if we can identify and prioritize them. The purposes of this thread is to provide feedback to Randy to pass to the programmer.
My list of desired features include
1. Allowance for additional criteria to establish 'soft conflicts' to spread folks up. Example: State or AMA district. The hard conflicts on frequency would be resolved, then a second pass would spread pilots from TX across the heats. Field is optional.
2. Print preview
3. Output to Excel spreadsheet with formatted summary scoresheet to assist the scorekeeper
4. Matrix analysis - ability to calculate summary of pilot vs pilot alignment to make sure we have a balanced contest.
Please put your comments here.
Thank you
A nice item would be to print the name of the contest and the class the heat card is for. I've seen guys use the wrong heat card a few times!
I like the spreading out by area so you get to fly against somebody besides the guys you always fly with. It's also nice if you don't fly with the same guys all contest long.
Also set it up so I never have to fly against Lee, Eric, Kirk, Brian etc. Put me in with all newbies! Maybe I'll have a shot!
I like the spreading out by area so you get to fly against somebody besides the guys you always fly with. It's also nice if you don't fly with the same guys all contest long.
Also set it up so I never have to fly against Lee, Eric, Kirk, Brian etc. Put me in with all newbies! Maybe I'll have a shot!
- Ed Kettler
- Posts: 3437
- Joined: Sat Dec 15, 2001 6:05 pm
- Location: USA
- Contact:
-
- Posts: 3330
- Joined: Thu Jun 20, 2002 10:30 pm
- Location: USA
Here's the latest screen shot the the program:
Some of the items discussed have already been incorporated.
There is still a lot of work to do. If anyone has any saved matrix files, you can send them to me so that I can run them through the new Matrix V2 program for debugging. I already have the LSN matrix files btw.
We need to be careful about adding too many constraints to the Matrix Generation. Obviously we can't do anything about Freq Conflicts, this is the nature of the beast. The Soft Constraints we've called them "Zones." Adding pilots to different zones will, when completed, prevent them from flying against each other. If this results in a matrix that has every round being very similar, then we may need to use a "weighting system" :: allow some pilots from the same zones to fly against each other but limit the "some." If that made any sense what so ever. I think a weighting system will yield the most diverse heats.
So right now we have (1) Freq Conflicts (2) Zones. Adding too many more constraints will generate a matrix where ever round is identical.
Because, or at least to the best of my knowledge, we generally like lots of pilots per heat but yields to having far fewer heats than we have pilots per round. This in itself leads to even fewer possible combinations.
It will be a lot easier to decide how the constraints effects the matrix once I have a working program.
Until next time
Chunk
Some of the items discussed have already been incorporated.
There is still a lot of work to do. If anyone has any saved matrix files, you can send them to me so that I can run them through the new Matrix V2 program for debugging. I already have the LSN matrix files btw.
We need to be careful about adding too many constraints to the Matrix Generation. Obviously we can't do anything about Freq Conflicts, this is the nature of the beast. The Soft Constraints we've called them "Zones." Adding pilots to different zones will, when completed, prevent them from flying against each other. If this results in a matrix that has every round being very similar, then we may need to use a "weighting system" :: allow some pilots from the same zones to fly against each other but limit the "some." If that made any sense what so ever. I think a weighting system will yield the most diverse heats.
So right now we have (1) Freq Conflicts (2) Zones. Adding too many more constraints will generate a matrix where ever round is identical.
Because, or at least to the best of my knowledge, we generally like lots of pilots per heat but yields to having far fewer heats than we have pilots per round. This in itself leads to even fewer possible combinations.
It will be a lot easier to decide how the constraints effects the matrix once I have a working program.
Until next time
Chunk
One more thing. Ed suggested a way to Export the Matrix to Excel. I think this is a good idea (read as I have already made programs to do just that.) If someone will make a "template" Excel sheet on how they wont it to look, then I can just stick the data into the appropriate wholes in the spread sheet.
Chunk
Chunk
- Ed Kettler
- Posts: 3437
- Joined: Sat Dec 15, 2001 6:05 pm
- Location: USA
- Contact:
I sent a template to Randy to pass to you. You will need to plug in the Contest name and date, the event (e.g. SSC), pilots and the number of rounds flown. This template is set up to calculate the pilot's average score in addition to total points. It would probably be easier to just create one workbook per event, rather than try to combine the contest into a single workbook.
For the sake of discussion, 'hard conflicts' are on frequencies, and 'soft conflicts' will be on the 'Zone' which is optional.
On the Frequency list, we need an entry 'SS' for spread spectrum. This will not conflict with anybody. We may need to consider 'SSnn' where 'nn' is a number 01-10 to allow father/son teams to be supported. Could also be S0 - S9 if we want to keep to 2 letters. We also need to have the Ham channels.
Looking good!
Thanks!
Ed
For the sake of discussion, 'hard conflicts' are on frequencies, and 'soft conflicts' will be on the 'Zone' which is optional.
On the Frequency list, we need an entry 'SS' for spread spectrum. This will not conflict with anybody. We may need to consider 'SSnn' where 'nn' is a number 01-10 to allow father/son teams to be supported. Could also be S0 - S9 if we want to keep to 2 letters. We also need to have the Ham channels.
Looking good!
Thanks!
Ed
Ed you must be reading my mind. I was thinking about this very thing this morning. This will mean a programing change. [:(!]
What I'll do is make the frequency just text instead of a number.
Can will still keep it to only 2 digits? At any rate, the length will have to stay the same. For example for 3 fixed digits, Ch 11 would be [011] or [ 11] or similar. If not, it's going to mean a big rewrite.
If you want to see what we've got to this point, take a look a the latest complied version here:
http://spadfest.rcspads.com/displayimag ... m=79&pos=5
Be certain to read the Readme file before installing or it will most likely crash when you try and run it.
It's no where near complete and I haven't tested it past 2 rounds. So I have no idea what it will do. Guess what I'm going to try now? lol
FYI, V2 will read Version 1 Matrix files but not the other way around. (ie backwards compatible not fowards compatable.)
cc
What I'll do is make the frequency just text instead of a number.
Can will still keep it to only 2 digits? At any rate, the length will have to stay the same. For example for 3 fixed digits, Ch 11 would be [011] or [ 11] or similar. If not, it's going to mean a big rewrite.
If you want to see what we've got to this point, take a look a the latest complied version here:
http://spadfest.rcspads.com/displayimag ... m=79&pos=5
Be certain to read the Readme file before installing or it will most likely crash when you try and run it.
It's no where near complete and I haven't tested it past 2 rounds. So I have no idea what it will do. Guess what I'm going to try now? lol
FYI, V2 will read Version 1 Matrix files but not the other way around. (ie backwards compatible not fowards compatable.)
cc
- Ed Kettler
- Posts: 3437
- Joined: Sat Dec 15, 2001 6:05 pm
- Location: USA
- Contact:
- Ed Kettler
- Posts: 3437
- Joined: Sat Dec 15, 2001 6:05 pm
- Location: USA
- Contact:
http://rcsource.hobbypeople.net/faqs/freqlist.htm
This contains a list of the channels for 50Mhz - HAM, and 72 Mhz aircraft. Do not use 27 or 75 Mhz lists. So you should end up with:
H00 - H09 for the HAM channels
C11 - C60 for 72 Mhz
S01 - S10 for spread spectrum. I think 10 entries will be more than sufficient, as the most I've seen for channel conflicts is 4, and really the only need for this is to split some teams up (e.g. father/son pairs), not deconflict frequencies.
Thanks for taking on this challenge!
This contains a list of the channels for 50Mhz - HAM, and 72 Mhz aircraft. Do not use 27 or 75 Mhz lists. So you should end up with:
H00 - H09 for the HAM channels
C11 - C60 for 72 Mhz
S01 - S10 for spread spectrum. I think 10 entries will be more than sufficient, as the most I've seen for channel conflicts is 4, and really the only need for this is to split some teams up (e.g. father/son pairs), not deconflict frequencies.
Thanks for taking on this challenge!
- Ed Kettler
- Posts: 3437
- Joined: Sat Dec 15, 2001 6:05 pm
- Location: USA
- Contact:
There is a good write-up in the current issue of <i>Model Aviation.</i> It operates on 2.4 Ghz (I think) and can isolate on two channels within the band, even if other SS sets are in operation. There is no interference with 72 Mhz or 50 Mhz bands, or within the SS. You 'bind' the receiver to the transmitter to lock out interference.