"CQ GX"

[af2z]

"CQ GX" Contest [Revised: 07/feb/09]

Introduction:

"GX" stands for "groups": in this case, random character groups that are
exchanged between operators. The objective is to accurately exchange as many
groups as possible with other ops. These groups consist of five-character
alphanumerics, similar to those used by the FCC in their code tests years ago.
Since the groups are completely random and unpredictable some care must be taken
to transmit and copy them accurately.

The QSO requirement is minimal: each op transmits one group to the other op,
selecting a fresh group for each QSO. Scores are based upon the total number of
groups that are accurately exchanged.


Rules:

-- Random Groups. Prior to the start of the contest, obtain a list of randomly
generated groups, either via email from the contest coordinator or by generating
your own list by using the posted spreadsheet. This is your "transmit list".

-- QSO. Send one group from your transmit list, selecting a fresh group for each
QSO. Also copy one group from the other op. Log both groups as a pair.

-- Multiple Contacts. Two stations may contact each other multiple times on the
same band.

-- Monitoring. Points may be earned by monitoring and correctly copying groups
exchanged between other stations. This means that Code novices, SWLs and others
may also participate in the contest without operating a transmitter.

-- Entry Submissions. After the contest ends email your list of
transmitted/copied group pairs to the contest organizer. For monitored
transmissions submit a list of copied groups.

-- Score. Scoring software compares all the submitted groups, transmitted &
copied, attempting to reconcile them: any groups you have transmitted that are
accurately logged by other ops will earn points for you as well as for the ops
who copied them; likewise for the groups you have copied if they are matched
with other operators' transmitted groups. Effectively, both ops in a given QSO
receive credit for the correct exchange of each random group: 1-point for
copying a group and 2-points for transmitting a group. SWLs receive credit for
accurately copying groups exchanged between other stations, 1-point per group.

See FAQ for details.

-end-

[af2z]

"CQ GX" FAQ: [Revised: 07/feb/2009]

1. What do the random character groups look like?
2. Why use random groups? Isn't that a lot of trouble?
3. How do I obtain a list of groups for sending?
4. Why are multiple contacts permitted?
5. What if the other op makes a mistake in transmission or copy?
6. Can you compete by just monitoring other stations' transmissions?
7. Why does a transmitted group earn more points than a copied group?
8. What is to prevent two ops from repeatedly contacting each other?
9. Are there any multipliers or bonuses?
10. Why are RST, SKCC#, QTH, etc. not required in the exchange?
11. How are contestants ranked?
12. How do I log my contacts?
13. How do I submit my entry?
14. What happens if I don't submit my entry?
15. What if two ops happen to send the same random group?
16. Are random groups allowed to be sent in the amateur bands?


FAQ:

1. What do the random character groups look like?

They are randomly generated 5-character alphanumeric groups, such as: CB69Y,
6AAXI, 4K9JP, etc. Each op will have a list of groups from which to send. One
new group is sent by each op for each QSO.


2. Why use random groups? Isn't that a lot of trouble?

You only send one group and copy one group per QSO. Contacts will be less
monotonous than typical contest exchanges because each character group will be
unique. The random characters will constitute truly unknown information
and will require attention in sending, copying and verifying; in short,
communicating via CW.


3. How do I obtain a list of groups for sending?

Email the contest coordinator prior to start of contest. Or download the posted
spreadhseet and generate your own transmit list.


4. Why are multiple contacts permitted?

The advantage of multiple contacts is that if your station has only limited
coverage you can still remain active and competitive by exchanging additional
groups with previously contacted stations. Poor propagation will not limit your
activity. Low-power ops, DX stations and those who are antenna-compromised will
have more key-time than usual and can compete more readily with big-signal
stations due to the increased number of contacts available to everyone. This
will result in increased band activity. In short: more QSOs and less CQing.


5. What if the other op makes a mistake in transmission or copy?

The scoring software cannot tell which op made an error, only that the
transmitted group does not match the copied one. Both ops work as a team during
a given QSO to insure that each group is passed accurately. So both ops share
the points, or loss. If only one of the groups in a QSO is passed without error
both ops get credit for it but neither gets credit for the group having errors.


6. Can you compete by just monitoring other stations' transmissions?

You earn 2-points for each transmitted group but only 1-point for copied groups.
This means that even if you could copy every single group exchanged by a
particular station his point total for these groups would still be 50% higher
than yours. So, it pays to transmit.

Nevertheless, SWLs and others who cannot transmit may find it fun to monitor &
copy the groups exchanged by other stations. "SWL-only" entries can be
ranked separately from those who work 2-way.


7. Why does a transmitted group earn more points than a copied group?

Because copied groups are easier to get. They can be obtained simply by
monitoring exchanges between other stations. If both were worth the same,
contestants who do more monitoring would have a points advantage over those who
are on the air actively making contacts.

Even though a copied group is worth less than a transmitted group, both ops in a
QSO should take care to accurately copy the other; condsider that if neither op
copies the other accurately, neither op will receive any points for the QSO.

Contest results will show the number of copied groups as well as transmitted
groups (if any) for each contestant.


8. What is to prevent two ops from repeatedly contacting each other?

Nothing. If this seem like fun to you then go for it by all means. Anyone who
does this for the entire contest should probably get some kind of award.
However, doing so will result in overall fewer number of stations contacted.
This will be evident in the contest results.


9. Are there any multipliers or bonuses?

No, the score is based only upon accurate exchange of information, not on
particular stations, their locations or status, or even the number of different
stations contacted. These factors are typically a matter of chance or equipment
sophisitication. Also, they are largely redundant because stations with the most
contacts necessarilly get the most multipliers and bonuses. Multipliers and
bonuses are basically an exercise in score calculation, and have relatively little
to do with operating per se.

In "CQ GX", antenna sophistication, transmitter power and propagation conditions
are all less relevant than in typical contact-based contests. "CQ GX" focuses on
information transmission: "Get the message through!"


10. Why are RST, SKCC#, QTH, etc. not required in the exchange?

These are irrelevant to the contest scoring. They would also be very redundant
in multiple exchanges between the same ops. But they can certainly be exchanged
if desired. If someone sends you a RST or SKCC# you should reply in kind. Or if
you think you might need the other guy's SKCC nr, etc. (for awards, WAS, etc.)
make sure to send yours to him by way of request. He is bound to respond with
same.


11. How are contestants ranked?

Ranking is by points for total number of groups accurately transmitted (2-pts
each) and received (1-pt each). Additional stats could include: Op Status (2-
way, SWL, mixed); Nr. of Stations Worked; Nr. of stations who monitored your
transmissions (SWL).


12. How do I log my contacts?

For each QSO you should record the group you copy next to the one you transmit.
Callsigns, freq, time, etc. are not required for this contest. It is important
to keep your transmitted groups and copied groups paired, one pair per each QSO.
You may also copy groups exchanged between other stations; in this case you
would just record them as a list of single groups, not pairs.

If you use a paper log, it would be handy to pre-print your transmit list
(received from contest organizer or generated by spreadhseet); then pencil in
your copied groups as you get them. You will of course have to transcribe your
list of groups to a text file or email for final submission to the scoring
center.

For computer users you can copy & paste your transmit group list into a text
file or spreadsheet and just type in your corresponding copied groups as you
receive them.


13. How do I submit my entry?

Submit only your callsign (or other ID for SWLs, etc.) and the code groups you
have exchanged and/or copied. For 2-way contacts submit one pair of groups per
line (i.e., one transmitted and the corresponding copied group), like this:

QOGUE KGQO8
IRVLT EMSCJ
MG19C U2W47
6JNVP UTXJH
WDRUB 4K9JP
...etc.

For any groups that you monitored being exchanged between other stations, submit
them as one group per line:

VG298
KGQO8
EMSCJ
U2W47
...etc.


14. What happens if I don't submit my entry.

None of the people you contacted will receive any points for their contacts with
you.


15. What if two ops happen to send the same coded group?

Not likely if everyone's transmit lists are randomly generated. A 5-character
alphanumeric group has over 60 million possible combinations. However, in the
unlikely event that your score does suffer from a randomly generated duplicate
group you may sue the contest organizer for emotional distress.


16. Are random character groups permitted to be sent in the Amateur bands?

The groups involve no coding (other than Morse) or ciphering. They are contest
exchange tokens and carry no encrypted information.

-end-

[KI6BHB]

Drew---

I like this idea quite a bit. I especially like that stations are actively involved and scoring even while monitoring and waiting to break the pileup (hi) and that SWLs can get involved. Here is some feedback on the GX.

First, a quick comment on "code groups." I suggest use of the term "random groups" instead which will avoid the connotation of an encoding of some kind. The groups are not codes or an encoding of information which the term implies. While I really like the idea of using random groups, I still wonder if using them may make the powers that be uncomfortable as they would make a natural cover for actually encoded information. Perhaps a strict requirement that all valid random groups come from the contest coordinator would alleviate any concerns. This would also allow the coordinator to ensure there are no duplicate groups.

I understand from the FAQ logging description that each random group is to be exchanged only once. It would be helpful to mention this in the contest description.

I'd like to see the scoring rules fleshed out in the main contest description. I've had to piece them together from the FAQ, and I'm still not sure I have it right. My understanding is that stations participating in an exchange get two points if their transmitted random group is correctly reported by the recieving station and one point for correctly reporting the received random group for a total of three possible points per exchange. Incorrect copy does not invalidate the entire exchange only the points associated with the incorrect group. Monitoring stations also get a point for each random group copied for a possible two points per monitored exchange.

I think the imbalance in the scoring of a given group between sender and receiver results in a disincentive to careful copy. If I as a receiver get the transmitted group wrong, I'm down one point, but the sender is down two points. If it is desired to retain this imbalance, then I suggest the entire exchange be invalidated if either group is in error. However, I suggest that the imbalance be removed and have the qso partners have an equal investment in each group.

The two points for transmission I think was introducted as an incentive to transmit. I propose the following scoring scheme which balances the investment in each group and provides an alternate incentive to transmit.

As in the current scoring proposed, each correctly copied group earns the receiving station a point. In place of two points for a transmitted group, a transmitting station is awarded one point for each correctly reported reciept of the group. In the case where only the exchanging stations are involved, each station will get two points out of an exchange. If there are moitoring stations, each transmitter will get additional points for each monitoring station reporting the transmitted group correctly. All stations will be equally invested in each group exchange, so all have the same incentive for careful copy. Since monitoring stations do not have an opportunity to request fills and transmitting stations are now invested in their participation, there is an additional incentive to additional care in transmitting.

With respect to rules regarding repeated exchanges with the same station, I would look to either loosen the rule a little or provide incentives for working other stations. As to loosening the rule, perhaps just require that at the end of an exchange the calling station must wait a calling cycle or two before calling again. That would allow isolated stations to continue the contest without undue artificial rule following. As to use of incentives, a bonus or multiplier for each station worked would discourage two station ping pong. Also, issuing a limited number of random groups which are used cyclically along with a rule that a receiver is disallowed points for copying a previously copied group would serve the purpose.

Thanks for a great idea and I look forward to getting to play.

Regards,

Dave

[af2z]

Drew---

I like this idea quite a bit. I especially like that stations are actively involved and scoring even while monitoring and waiting to break the pileup (hi) and that SWLs can get involved. Here is some feedback on the GX.

Dave-

Thanks for your comments! I wll post a reply here in the next day or so...

73,
Drew

[af2z]

I like this idea quite a bit. I especially like that stations are actively involved and scoring even while monitoring and waiting to break the pileup (hi) and that SWLs can get involved. Here is some feedback on the GX.

Dave, thanks for all your comments...

First, a quick comment on "code groups." I suggest use of the term "random groups" instead which will avoid the connotation of an encoding of some kind. The groups are not codes or an encoding of information which the term implies. While I really like the idea of using random groups, I still wonder if using them may make the powers that be uncomfortable as they would make a natural cover for actually encoded information. Perhaps a strict requirement that all valid random groups come from the contest coordinator would alleviate any concerns. This would also allow the coordinator to ensure there are no duplicate groups.

Good idea, I have changed all the references to "random groups".

As for who generates the random groups... I would like to make it as easy as possible for people to jump into the contest. Probably a lot of people only decide at the last minute (or later!) that they can spend some time in a contest; so it would be good if they have the option of immediately generating their own list of groups on the fly. The chances of duplicate groups will be extremely small using a random generator spreadsheet. And it is instantaneous.

In any case, the contest coordinator has no control over what people actually transmit, whether he supplies the groups or not, so not sure that is a concern for the security minded...

I understand from the FAQ logging description that each random group is to be exchanged only once. It would be helpful to mention this in the contest description.

Done.

I'd like to see the scoring rules fleshed out in the main contest description. I've had to piece them together from the FAQ, and I'm still not sure I have it right. My understanding is that stations participating in an exchange get two points if their transmitted random group is correctly reported by the recieving station and one point for correctly reporting the received random group for a total of three possible points per exchange. Incorrect copy does not invalidate the entire exchange only the points associated with the incorrect group. Monitoring stations also get a point for each random group copied for a possible two points per monitored exchange.

That is all correct. But I have edited the description a little. See what you think of it. I still have to tighen up the description so that it mostly stands on its own without the FAQ...

I think the imbalance in the scoring of a given group between sender and receiver results in a disincentive to careful copy. If I as a receiver get the transmitted group wrong, I'm down one point, but the sender is down two points. If it is desired to retain this imbalance, then I suggest the entire exchange be invalidated if either group is in error. However, I suggest that the imbalance be removed and have the qso partners have an equal investment in each group.

The two points for transmission I think was introducted as an incentive to transmit. I propose the following scoring scheme which balances the investment in each group and provides an alternate incentive to transmit.

As in the current scoring proposed, each correctly copied group earns the receiving station a point. In place of two points for a transmitted group, a transmitting station is awarded one point for each correctly reported reciept of the group. In the case where only the exchanging stations are involved, each station will get two points out of an exchange. If there are moitoring stations, each transmitter will get additional points for each monitoring station reporting the transmitted group correctly. All stations will be equally invested in each group exchange, so all have the same incentive for careful copy. Since monitoring stations do not have an opportunity to request fills and transmitting stations are now invested in their participation, there is an additional incentive to additional care in transmitting.

Yes, I see your concerns. However, there are a couple of things to consider:

First, the scoring options are somewhat limited by the "software". Actually, I am not a programmer and the scoring software is only a spreadsheet. For the given rules it is a relatively simple spreadsheet, though potentially huge. There probably aren't many scoring changes I can make without making the spreadsheet formulas a lot more complex. Believe it or not, invalidating the entire exchange when only one group is in error, as you have suggested, would be pretty complicated. I am a little hesitant to get into that, not knowing if the contest will ever even get off the ground. But perhaps later on...

As for the "disincentive to careful copy" that you mentioned, I have added a new FAQ question for this-- Nr.7. Please see what you think of it...

If copied and transmitted groups are given the same point values I think those who monitor a lot of transmissions will have a points advantage over those who spend most of their time working other stations. Consider, you would only have to park your VFO(s) on a strong station(s) and just copy away! You would not have the "wasted" overhead of having to roam the dial in search of contacts, perhaps also exchanging greetings or SKCC nrs, etc.

As for your suggestion "each transmitter will get additional points for each monitoring station reporting the transmitted group correctly...", that is something I will look into. Again, this involves the coding of the spreadsheet formulas. It is certainly an interesting idea... but consider, the "big signal" stations will have the advange due to having more people monitoring them. I would like to minimize the equipment/antenna disparity among stations as far as possible.

With respect to rules regarding repeated exchanges with the same station, I would look to either loosen the rule a little or provide incentives for working other stations. As to loosening the rule, perhaps just require that at the end of an exchange the calling station must wait a calling cycle or two before calling again. That would allow isolated stations to continue the contest without undue artificial rule following. As to use of incentives, a bonus or multiplier for each station worked would discourage two station ping pong. Also, issuing a limited number of random groups which are used cyclically along with a rule that a receiver is disallowed points for copying a previously copied group would serve the purpose.

The "software" can determine the number of different stations worked by a particular station. How that figure would be incorporated into the scoring, I haven't thought about too much yet. As a simplest case "nr of different stations worked" could be a separate ranking category. It's pretty hard to meaningfully fine tune a single scoring formula. In general, bonuses and multipliers are too blunt; they tend to randomize the ranking of contestants over a short range of standings. But I will look into it.

Thanks for a great idea and I look forward to getting to play.

Thanks for the great comments, Dave! I have probably missed something you said. I will look it over again and respond later if needed...

73,
Drew

[af2z]

Drew---

With respect to rules regarding repeated exchanges with the same station, I would look to either loosen the rule a little or provide incentives for working other stations. As to loosening the rule, perhaps just require that at the end of an exchange the calling station must wait a calling cycle or two before calling again. That would allow isolated stations to continue the contest without undue artificial rule following.

I think some explicit rule is definitely needed... What if the requirement was
changed to: "Each op may transmit only one group per QSO." [I have changed the
rules and FAQ to reflect this.]

As to use of incentives, a bonus or multiplier for each station worked would discourage two station ping pong.

It's really difficult to make meaningful multipliers or bonuses when station
coverage capability can vary by a factor of ten. Instead of trying to fine-tune
a scoring formula it might be better to just rank stations in two categories:
"Nr of Groups Exchanged" and "Nr of Different Stations Worked". That at least is
easy for anyone to interpret as they will, compared to dealing with the
uncertain effect of multipliers and bonuses...

Also, issuing a limited number of random groups which are used cyclically along with a rule that a receiver is disallowed points for copying a previously copied group would serve the purpose.

I think that would be too complex.

Thanks for a great idea and I look forward to getting to play.

Well, truth is, there is little likelihood that it will get off the ground (see
the poll at top of this thread). But if there were a handful of people
interested in it we could run a short trial or two to see what the dynamic is
like. I think we'd proably need at least 25 participants. Though a smaller
number of participants over a longer period of time, say 1 week, might serve as
well. That would be less of a sprint and more of a "quest". If that proved to be
fun it might draw further interest in a more sprint-like contest.

Any other ideas? Please post them...

73,
Drew

[KI6BHB]

I think some explicit rule is definitely needed... What if the requirement was
changed to: "Each op may transmit only one group per QSO." [I have changed the
rules and FAQ to reflect this.]

I agree, there needs to be an explicit rule. I tried to figure out what it was about the original rule, "each station must QSO with another station before a subsequent QSO," and it mostly comes down to the fact that the individual stations don't have enough information to know if the rule has been met. In other contests, stations know who they've contacted previously.

In any case, I think the original rule is probably the only one that will promote diversity of contacts and perhaps diversity on bands.

Well, truth is, there is little likelihood that it will get off the ground (see
the poll at top of this thread). But if there were a handful of people
interested in it we could run a short trial or two to see what the dynamic is
like. I think we'd proably need at least 25 participants. Though a smaller
number of participants over a longer period of time, say 1 week, might serve as
well. That would be less of a sprint and more of a "quest". If that proved to be
fun it might draw further interest in a more sprint-like contest.

Actually, I was viewing the poll results with a more positive spin. About half of those responding are likely to play. The polling sample is bound to include many non-contesters (much fewer than half of SKCC partake in SKS) so, its likely that more than half of contesters may be likely to play GX. Perhaps after settling on the rules you want, post for some volunteers on the Yahoo groups and do the test run you mentioned. Then maybe Dick would allow a GX theme with bonus points for one of the sprints.

[af2z]

I agree, there needs to be an explicit rule. I tried to figure out what it was about the original rule, "each station must QSO with another station before a subsequent QSO," and it mostly comes down to the fact that the individual stations don't have enough information to know if the rule has been met. In other contests, stations know who they've contacted previously.

That never occurred to me. I assumed people would just remember who they
contacted last, even if they are not required to log any callsigns or freqs as a
contest requirement. For myself, I would probably log these things as usual,
plus the sent/rcvd code groups required by the contest.

In any case, I think the original rule is probably the only one that will promote diversity of contacts and perhaps diversity on bands.

So, is "you can't work the same station twice in a row" a good rule? Or is it
too "Mickey Mouse"... I think I will leave it out for awhile. As it stands I
think there is probably enough motivation to work as many different stations as
one can in the contest. (The Rules & FAQ have both been updated on this matter.)

Actually, I was viewing the poll results with a more positive spin. About half of those responding are likely to play. The polling sample is bound to include many non-contesters (much fewer than half of SKCC partake in SKS) so, its likely that more than half of contesters may be likely to play GX. Perhaps after settling on the rules you want, post for some volunteers on the Yahoo groups and do the test run you mentioned. Then maybe Dick would allow a GX theme with bonus points for one of the sprints.

I would like to run a trial of this contest, but not sure the established SKCC
contest periods would be the right place to do so. The contest is a little too
different and a lot of folks would not be happy with such a change. If we can
get enough people interested I'd like to hold a trial on some other night.

I will let it cook here for awhile, and maybe come up with a few more comments
or ideas. When it is firmed up a bit I'll try to recruit a group of ops to try
it out.

73,
Drew

[AC2C]

Drew, I like the concept.

What kind of time frame are you thinking of - Maybe 24-hours ?

Count me in (assuminig my rig will be repaired by then) !

73,
Ron, AC2C

[af2z]

--- In skcc@yahoogroups.com, WILLIS COOKE <wrcooke@...> wrote:
>
>Frank and Drew, who is going to score this contest? It will
require that logs be submitted and if a log is not submitted, then
the contacts made by that station cannot be scored, so they will
either need to all be allowed or all disallowed. Who is volunteering
to do the scoring? Frank? Drew? Not me! This is getting far away
from the KISS principle! The big contests are scored by robots and
contacts with stations that do not submit logs are allowed. Someone
has to write the robot. Who is volunteering for that awesome task?
>

Cookie,

The "log" consists only of a list of groups exchanged: one transmitted & one
copied per each QSO (no times, freqs, callsigns, rst). For "SWL" entries, the
log is simply a list of groups copied.

I think it is very simple for the contestants: you transmit one group and copy
one group in each QSO. (Or for SWLs: copy groups exchanged by other stations.)
Then you submit your list of groups to be scored. That's all there is to it. No
bonus calculations, no counting of QSOs or SPC multipliers or special bonuses
for particular stations on different bands while watching out four dupes, etc.
etc. It is very simple.

I will score the contest with a spreadhseet. This will be (initially) a semi-
manual process but probably not a big deal with less than 50 entries and
several thousand contacts.

If a character group that is exchanged on the air is not submitted for scoring
by both the transmitting op and the receiving op, it dose not earn any points
for either. If an op does not submit any of the groups he exchanged on the air
then none of the contacts he made will earn any points for him or for the
stations who worked him. This is a necessary consequence of the basic goal of the
contest: i.e., accurate exchange of information. This is automatic. There is no
process of allowing or disallowing. Submitted groups are matched and scored.
Groups not submitted or matched will earn no points.

The hardest part will be for the op who uses a paper log and makes a lot of
contacts. He will have to transpose his handwritten list of copied groups to a
text file or email. Better to type your copied groups on the PC as you receive
them if you think you will have a lot of contacts.

73,
Drew

[af2z]

>Hey Drew, I may be totally off base here (I usually am) but would FCC Part
>97 apply here? I'm thinking Part 97.113 titled MESSAGES IN CODES OR CIPHERS
>may apply. I admit that I haven't read 97.113 as yet but perhaps we should
>and get an interpretation prior to moving forward. Just a thought.
>
>73, Gary -K8KFJ-
>SKCC #2444

Gary,

There are no codes or ciphers involved (other than Morse Code) and no intention
to obscure the meaning of any messages. The random character groups are just
contest exchange tokens and carry no encryped information.

73,
Drew

[af2z]

Drew, I like the concept.

What kind of time frame are you thinking of - Maybe 24-hours ?

Count me in (assuminig my rig will be repaired by then) !

73,
Ron, AC2C

Ron,

I will add you to the list of those who are interested. All depends on getting
enough people together to set up a trial. Have to think about wich would be
better initially: a short sprint (1 or 2 hours), or a more leisurely extened
trial... I think the longer period might be better...

73,
Drew

[AC2C]

[quote="af2z"]... I think the longer period might be better...

[/quote]

Agreed, Drew. A longer period would be better for a trial run, especially with a limited number of participants.

73, Ron

[af2z]

"CQ GX" Contest [Revised: 07/feb/09]

Posted three years ago. Got 21 votes total in the attached poll to this date.

[k8jd]

An activity with pages of rules and endless postings asking for clarifications is just not attractive to a casual operator like me !
Count me out.
Sorry.
73, John....JD