Wednesday, July 28, 2010

27 July 2010 - CT1BOH endorses RBN - " - "The King (Packet cluster network) is dead! Long live the King (Skimmer reverse beacon network)"

When one looks at the growth of QSOs during 48 hour contests (notably
maximum QSOs in the CQWW SOAB CW category in the past decades:
50's 1277 QSOs
60's 2623 QSOs
70's 4505 QSOs
80's 5970 QSO
90's 7555 QSOs
00's 7828 QSOs
one has to wonder where have all those extra QSOs have come from.

- The operators have better resources and share more information (more
knowledgeable with propagation and openings, more skillful with SO2R
resources, ...)
- The logging software and Computer Generator CW have added extra efficiency
when working stations
- The stations (split TX signals) and low band antennas (both RX and TX)
have added extra QSO potential
but If there is one aspect that has impacted Contesting the most, for
good or for bad, in my opinion, it is:
- The development of Packet Cluster network by AK1A in the late 80's and the
enhancement it has given to contest operators to be found while CQ'ing on a
band

There was one aspect in the Packet Cluster network that was not fair. Since
most users of the packet cluster network are DX'ers they tend to spot
stations more according to their DX rareness than to their actual operating
time.

Looking at the packet spot data for 2009 CQWW CW with its almost 51.000
packet spots, listing the stations with most spots shows great differences
between them. Almost all these stations were active during the 48 hours of
the contest but still great differences exist:

For stations with most spots worldwide
HC8GR-573, LX7I-271 , EA8URL-254, JA3YBK-249, B7P-219, VK9XW-216, EE2W-211,
K3LR-206, KC1XX-206, JA5FDJ-198

For stations with most spots in the USA
K3LR-206, KC1XX-206, W3LPL- 165, K1LZ-156, K1TTT-155, NQ4I-139, W2FU-115,
NR4M-103, NR5M-99, K0RF-63

For station in the SOAB Category worldwide
V47NT-190, ZS4TX-183, A25NW-182, 8P5A-174, VE2IM-160, P49Y-146, 6W1RW-145,
EF8M-144, VQ5M-132, CR3E-127

There is not fairness in the packet cluster network. The more DX you are,
the more known you are the more spots you get, even if you have a great
signal or are QRV all the time. The K3LR verus K1LZ or V47NT versus EF8M
are good examples.

Another interesting aspect also to note is that more and more people rely on
the packet cluster network to work stations in contests.
Looking at 2009 CW Claimed results, there were 783 stations submitting logs
in the SOAB category and 1020 stations submitting logs in the SOAB Assisted
category
The fact that there is a majority of stations operating the Assisted
category versus the Non Assisted category in the SOAB class also provides
extra QSOs to those CQing stations, because in order to maximize QSOs these
Assisted stations are not only going after DX multipliers but also after
stations that are not necessarily multipliers for new QSO points.

In 2008 when VE3NEA - Alex Shovkoplyas developed CW Skimmer I sent him the
following note on Mar 2008:

"(...) you realize this is the end of CW contesting as we have known it. It
will be a very different contest in the future. (...) Also, with a script
that puts all the calls from CW Skimmer into packet, no need for "old
version packet". It's like having a requirement in the rules for self
spotting.(...)"

What I did not realize was how quickly would Skimmer reverse beacon network
achieve such a degree of coverage and overcome Packet Cluster Network in a
devastating way.
The result can be tagged as "shock and awe" to use a military jargon.
In order for you to understand the degree of band coverage Skimmer reverse
beacon network is capable of I will compare the results of packet cluster
network performance against Skimmer reverse beacon network, using the last
IARU HF Championship contest as a reference and my CR3E operation.

I took the packet cluster network data from DX Summit Search option
http://www.dxsummit.fi/Search.aspx and Skimmer reverse beacon network from
http://www.reversebeacon.net/raw_data/

CR3E was spotted 40 times in the packet cluster network. An average of one
spot every 36 minutes
CR3E was spotted 500 time in the Skimmer Reverse Beacon network. An average
of one spot every 2.9 minutes

The difference is incredible.
What is the impact of this difference? In order for all to fully understand
I will use the packet window concept.
Most Contest logging programs have a packet spot lifetime option to display
new multipliers and QSO's in the band map. Using a common 30 minute default
option, l can calculate the packet coverage of both the packet and skimmer
reverse beacon network.

Imagine CR3E started the contest on 15 meter and was QRV for 120 minutes on
that band.
Imagine there was only one packet spot during that 120 minute QRV and that
it occurred in the minute 20
CR3E will be in Assisted stations band map (if not worked) during the packet
spot lifetime window from minute 20 until minute 49 (30 minutes), when the
spots lifetime ends.
CR3E was QRV 120 minutes and there was a coverage of 30 out of 120 minutes,
i.e. 25% packet coverage for the time of QRV

So what is the difference between Packet Cluster Network (PCN) and Skimmer
Reverse Beacon Network (SRBN) for CR3E IARU operation when it comes to
packet coverage?

Looking at the 1440 minute of the IARU contest I checked how many minutes
CR3E was QRV while CQing/Running. I removed minutes with SO2R QSOs that are
not spotted because I'm replying to stations:

On 28 Mhz CR3E was QRV 68 minutes, there was a 65% PCN and 78% SRBN coverage
On 21 Mhz CR3E was QRV 485 minutes, there was a 60% PCN and 96% SRBN
coverage
On 14 Mhz CR3E was QRV 503 minutes, there was a 53% PCN and 98% SRBN
coverage
On 7 Mhz CR3E was QRV 283 minutes, there was a 33% PCN and 99% SRBN coverage
On 3.5 Mhz CRE was QRV 28 minutes, there was a 86% PCN and 96% SRBN coverage
On 1.8 CR3E did not CQ, just S&Ped.

Skimmer Reverse Beacon Network (SRBN) achieved an incredible 96% window
coverage. while CR3E was CQ/running. i.e. CR3E was spotted ALL the time of
the contest, whenever he was CQing (TEST CR3E). To put it in another way the
minute I CQ the minute a Skimmer Robot will spot me.
There was even one time Skimmer reverse beacon network spotted me before I
even logged stations.

With Skimmer reverse beacon network, spotting will rely on robots. Everybody
will be spotted about the same for same conditions. It's democracy into
packet spots.

t's no wonder stations like K3LR have implemented their won skimmer network
http://www.k3lr.com/skimmer/ in order to have all the data of active
stations on the bands, in order to gain an extra competitiveness over their
competitors. Tim K3LR told me that last ARRL they even found a new opening
never heard of, thanks to skimmer.

Assisted stations, MS, M2, MM, DXers and alike don't need systems like K3LR
to gather all this data. They can feed into the telnet aggregated output of
all currently active skimmer reverse beacon receivers in real time at
telnet.reversebeacon.net port 7000. Check
http://reversebeacon.blogspot.com/ for details.

Relying in the old Packet cluster network is the best way to loose. Relying
into Skimmer reverse beacon network is the way for Assisted station in CW
Contests.

I guess it is proper to say The King (Packet cluster network) is dead! Long
live the King (Skimmer reverse beacon network).

I have never operated in the Assisted SOAB class in my 36 CQWW Contests as a
single operator that span over 21 years non stop since 1989.
http://www.qsl.net/ct1boh/operations.htm
I do recognize the great impact packet has had in my pile-ups and
operations. I do love packet pile-ups. "It's rush after rush". Packet
spotting is certainly here to stay, no matter what.

Will the contest be a different one? Information is tilting the nature of
the contest. It will not be worse nor better. It will be just different. And
I guess change is what people fear the most.

John F. Kennedy once said "Change is the law of life. And those who look
only to the past or present are certain to miss the future."
I don't really care if the good old days are better than today but it is an
interesting coincidence that for some, the less days one has left the better
the good old days are...


73
--
--
José Nunes
CONTEST CT1BOH - http://www.qsl.net/ct1boh
_______________________________________________
CQ-Contest mailing list
CQ-Contest@contesting.com
http://lists.contesting.com/mailman/listinfo/cq-contest


Friday, July 23, 2010

Spot Analysis Tool in Action

The Reverse Beacon Network's Spot Analysis Tool has been getting a good workout since the IARU contest. For those who haven't tried it yet, a small introduction.


This screen shows the basic features of the tool (click on the image for a magnified version). First, you choose a date (above the screen shown, on the left). Then select a continent, and the tool tells you which reverse beacons (or skimmers, if you prefer) were active on that day, and how many spots each made. Select a reverse beacon, and move to the bottom of the screen where a box invites you to "ADD" a callsign. Type one, click the ADD button, and any spots of the station will be plotted, band by band.

Now the best part. Add another, and the tool compares the signals of the two stations. Add more (up to 10, total), and their signal strengths are plotted on the screen.

If you want, you can zoom in on the graph and examine the individual spots more closely. You can also change reverse beacons, add or delete stations to compare, and zoom in on multiple graphs.

In the example above, I compared the signals of several WRTC stations in Europe. It looks to me that the organizers achieved their goal of equalizing signals pretty well.

73, Pete N4ZR


Monday, July 12, 2010

7-9 Spots a Minute During IARU

I'm happy to report that the Reverse Beacon Network made 312,867 spots
in the 34 hours of the IARU Contest. On average, each active Reverse
Beacon was spotting 7-9 stations per minute. We are working on
extracting numbers of spots of each WRTC station during the period, and
hope to have these later today.

73, Pete N4ZR

Friday, July 9, 2010

New RBN Feature Out for Beta Test During IARU Contest

The Reverse Beacon Network team is pleased to announce that a new and (we hope) exciting feature is being released for beta testing during and after the IARU Contest.

The Spot Analysis Tool, written by Nick Sinanis, F5VIH/SV3SJ, can now be found at http://reversebeacon.net/analysis . In a nutshell, it is intended to provide easily usable comparisons between multiple stations over an entire day at a glance. You can select a date and a Reverse Beacon station on any continent, then enter the callsign of a station, and in seconds all the spots of that station made by that Reverse Beacon are displayed graphically, with either Signal to Noise Ratio SNR) or frequency on the vertical axis, and the 24-hour time on the horizontal. Add another callsign, and spots of that station are superimposed on the first. Click the "i" icon at the upper right of the page for step-by-step instructions, if you need them.

If you select SNR, you will immediately be able to see which station opened the band to the area represented by the Reverse Beacon, which had the better signal during the middle of the opening, and which fared better in the waning hours. Select a time period and zoom in to get a close look at comparative signal strength. Selecting frequency to plot will show whether stations changed run frequencies often and which frequencies they were (or were not) able to hold, and when.

A few cautionary notes. This *is* an automated system. It only spots stations that it believes are CQing. The SNR for a given spot can be heavily influenced by QRM or QRN, so general trends are a lot more instructive than individual spots. If Slim decides to sign somebody else's call, and it's heard, it will be spotted, and the frequency calibration may vary (though we're working on that) depending on the Reverse Beacon's individual setup.

As we said, this is a beta test version. There are a few limitations right now - most important, it does not yet work with Internet Explorer, so you will need to use Firefox or Chrome to test it. While Nick is working on that problem, we'd like to get comments about the usability of the tool, reports on any bugs you encounter, and any other input you want to provide. While we're not averse to comments on CQ-Contest, you can avoid a glut there by sending your comments to us at skimmer@dxwatch.com.

We look forward to hearing from you!

73,

Nick, F5VIH/SV3SJ, Felipe PY1NB, and Pete, N4ZR

Thursday, July 1, 2010

Twenty Times as Many Spots!

A very interesting talk on new technology in contesting was given last week at the Friedrichshafen hamfest in Germany by Tobi, DH1TW. It featured the Reverse Beacon Network, and contrasted the number of spots generated by RBN during the recent WPX CW contest with those delivered by the conventional spotting network. You can see it (in English) at hhttp://www.dh1tw.de/how-new-technologies-affect-the-art-of-contesting.

It's pretty breathtaking - the RBN delivered about 837,000 spots from 41 reverse beacons in 21 countries, while the conventional spotting network delivered less than 40,000. The RBN also collected two and one half times as many callsigns, and about 35 percent more total prefixes during the 48-hour contest. Not bad.

The RBN is very close to the perfect information tool for WPX, but it offers the same benefits in every contest - huge volume, no subjective "to spot or not to spot" judgments, and spot quality as high or higher than conventional DX clusters. Now, thanks to K4TD, you can even apply full filtering to the spot stream from the RBN. The Telnet server at telnet.reversebeacon.net port 7000 uses the DX Spider cluster server, and you can set it up to deliver a very high level of specificity - filtering out all spots except those from your own area, for example, or spotting only stations from Europe (WAE is coming, after all).

Stay tuned for a major announcement, coming soon - the RBN Signal Analysis Tool, developed by Nick, F5VIH/SV3SJ. It's in test now, and once a few last bugs are smoothed out, you'll see why Felipe and I are so excited about it.

73, Pete N4ZR