Some interesting stats.... so far this month there have been 37,544,226 spots uploaded to wsprnet. 6,875,914 by 318 reporting stations using the built in wspr extension. About 18%. Impressive!!!!
That doesn't count the spots by people using kiwis and running AI6VN's kiwidaemon or wsprdaemon which would be another large number.
I'd be interested to see an estimate of the combined number. Have kiwis perhaps become THE major spotting device for WSPR? If so, it would seem that they may rival or exceed the impact that the U3 and other QRPLabs transmitters have had on transmitted WSPR, "spottees". Can this really be the case?
the WD ID in the wsprnet archive is MIA, so I can't get that count easily right now.... if I knew everyone running kiwidaemon and wsprdaemon, I could get it by their call
The SW id can only be loaded into wsprnet.org by executing the 'one-spot-per-http' curl POST upload method. Many kiwiwspr and wsprdaemon sites upload 10 to 30 spots at the end of each 2 minute cycle, and using POST to upload them increases the transmission and computation burden on the clients and more importantly on the wsprnet.org server(s). The frequent 60-90 second hangs (and occasional timeouts) during http uploads compounds the inefficiencies and errors introduced by POST uploads as used by WSJT-x. In contrast WD uploads all spots from all bands gathered during a 2 minute cycle in one MEPT http transaction and the resulting efficiency allows for retries of the occasional http transfer failure. I am not aware of any other wsprnet.org client which uploads using MEPT, so I think you can count any spots with *no* swversion field as coming from WD and I would be interested in getting the CALL of those reporters.
3,671,855 more spots were contributed by 10 kiwi stations known to be running Rob's daemon. So that equates in total to 10,547,769 spots or 28%. John, your kiwi has really impacted the WSPR world!!
WSPR mode? as many HF bands as possible, tiny bandwidth each RX, no waterfall, no SDR.hu remote connection or processor heavy tasks etc I think it is a great device, somewhat ahead of the minds of us users.
It's working but everyone is so engrossed in their SDR's (which are working without issue) they don't have time to post.
I got extra supplies but am trying to do some CAD drawings for metal cases recycling junk I have to hand, I didn't record the last one. Can find 3D Beaglebone models but am taking too long to make bits from the KiwiSDR PCB/Gerbers. I used to enjoy CAD but have abviously got rusty and only have free programs installed these days.
WSPR stuff still showing interesting trends thanks.
Conditions yesterday meant I had the best day so far for WSPR RX, at one point I was up there with the serious stations. I was looking forward to the day's total if only to see what a bit of increased solar wind can do. I retired to bed thinking "should be a good total, if nothing goes wrong".
Long story short, just after going to bed we had power cut of over 6 hours affecting me and 71 other homes in the local area. One thing hit me at about 1AM, since my adoption of fibre there are more items in-line to keep powered up in the event of a power failure, maybe I should keep a spare router, long network cable, DC-DC adapters and leisure battery for those events, I do love the chance to record a bit of spectrum when everyone else is forcibly powered down.
Mains power loss can be a wonderful time for evaluating one's station. In the past I've been able to compare household, local neighborhood and ~10km square area power loss to see the results by having a station that can survive it. At that time I got ~10 dB noise floor improvement for each of the three steps. It's very impressive to see HF like it used to be about 1960 in rural California! Since that time, I used the information to recapture at least 20 dB of that improvement through careful control of common mode current and near-field coupling, even when all the power was back on. It takes some effort but the rewards are large. The broadband sweep of the kiwi is very helpful in understanding these sources and mechanisms, which is a necessary precursor to squashing them.
The quest for noise reduction is about my main obsess... focus. Agreed the full spectrum is one of the most effective tools for noise identification, before the Kiwi I would identify a problem or background noise at one band but not see the general redistribution of energy depending on grounding, shielding or ferrites etc. Here noise can be severe but that is good in some ways as improvements achieved at the worst times are still there when they would be hard to spot. It's still a poor location from the fact I can't get the antenna far enough away the multiple sources but I'm learning along the way.
One thing I have reluctantly accepted is that it is easier to get a station that works well for digital than acceptably for human ears.
WSPR reporting has gone astray a bit. There's now ad hoc competitions that occur to see who can spot the most uniques. Some of the methods used to do that can obscure WSPR's original intention, propagation analysis. Multiple spots for the same station can be determined by post processing of the online wsprnet logs. But not uploading spots in a 2 minute period because they are not new uniques is lost data which cannot be used for propagation analysis. Whatever method you use be cognizant of wspr's primary mission.
Especially running wsprdaemon V2.5 it is important that all spots are uploaded. Those spots and the background noise level data are now being studied by academic researchers to better understand the effect of solar weather upon the ionosphere. So please contribute clean data to our effort!
Jim said "WSPR reporting has gone astray a bit. There's now ad hoc competitions that occur to see who can spot the most uniques. Some of the methods used to do that can obscure WSPR's original intention, propagation analysis. Multiple spots for the same station can be determined by post processing of the online wsprnet logs. But not uploading spots in a 2 minute period because they are not new uniques is lost data which cannot be used for propagation analysis. Whatever method you use be cognizant of wspr's primary mission."
How about if there were a parallel report, one that ranks on total spots rather than uniq's. I suppose it needs to be filtered to allow only one spotter-band-time to prevent someone from stacking up a bunch of decoders/receivers just to inflate their position. Would that provide a useful metric and also help to restore transparancy?
Comments
That doesn't count the spots by people using kiwis and running AI6VN's kiwidaemon or wsprdaemon which would be another large number.
If so, wow!
I am not aware of any other wsprnet.org client which uploads using MEPT, so I think you can count any spots with *no* swversion field as coming from WD and I would be interested in getting the CALL of those reporters.
I think it is a great device, somewhat ahead of the minds of us users.
I got extra supplies but am trying to do some CAD drawings for metal cases recycling junk I have to hand, I didn't record the last one.
Can find 3D Beaglebone models but am taking too long to make bits from the KiwiSDR PCB/Gerbers.
I used to enjoy CAD but have abviously got rusty and only have free programs installed these days.
WSPR stuff still showing interesting trends thanks.
I retired to bed thinking "should be a good total, if nothing goes wrong".
Long story short, just after going to bed we had power cut of over 6 hours affecting me and 71 other homes in the local area.
One thing hit me at about 1AM, since my adoption of fibre there are more items in-line to keep powered up in the event of a power failure, maybe I should keep a spare router, long network cable, DC-DC adapters and leisure battery for those events, I do love the chance to record a bit of spectrum when everyone else is forcibly powered down.
Since that time, I used the information to recapture at least 20 dB of that improvement through careful control of common mode current and near-field coupling, even when all the power was back on. It takes some effort but the rewards are large. The broadband sweep of the kiwi is very helpful in understanding these sources and mechanisms, which is a necessary precursor to squashing them.
Agreed the full spectrum is one of the most effective tools for noise identification, before the Kiwi I would identify a problem or background noise at one band but not see the general redistribution of energy depending on grounding, shielding or ferrites etc.
Here noise can be severe but that is good in some ways as improvements achieved at the worst times are still there when they would be hard to spot. It's still a poor location from the fact I can't get the antenna far enough away the multiple sources but I'm learning along the way.
One thing I have reluctantly accepted is that it is easier to get a station that works well for digital than acceptably for human ears.
Can you elaborate (by whom, will there be publications)?
How about if there were a parallel report, one that ranks on total spots rather than uniq's. I suppose it needs to be filtered to allow only one spotter-band-time to prevent someone from stacking up a bunch of decoders/receivers just to inflate their position.
Would that provide a useful metric and also help to restore transparancy?
Could it be that 6m spots disappear? I had one today but it doesn't show up on V/U.