Re: LF: Re: WSPR-15

[Markus Vester <markusvester@aol.com>]

The new script seems to do the job as it should. I'm currently running it to feed WSJT-X v2.0, side-by-side with old WSPR-X v0.8 r3058 set to -15. IW4DXW is decoding well on both screens (hence the double uploads). But a few minutes ago IZ7SLZ who is weaker here appeared only on the accelerated WSJT feed.

Domenico's solution (new wsprd.exe under old WSPR-X) seems much more comfortable though, I hope that I can also make it work here.

Tonight I'll occasionally send WSPR-15 on 137622, a couple of Hz above Stefan's DFCW.

All the best,
Markus


-----Ursprüngliche Mitteilung-----
Von: Markus Vester <markusvester@aol.com>
An: rsgb_lf_group <rsgb_lf_group@blacksheep.org>
Verschickt: Do, 15. Nov. 2018 18:04
Betreff: Re: LF: Re: WSPR-15

First bug: Sndinput decimation factor needs to be 1024 not 512. 

73, Markus


-----Ursprüngliche Mitteilung-----
Von: Markus Vester <markusvester@aol.com>
An: rsgb_lf_group <rsgb_lf_group@blacksheep.org>
Verschickt: Do, 15. Nov. 2018 17:58
Betreff: LF: Re: WSPR-15

Hi Stefan,

here is an attempt for a wspr-15 to wspr-2 receive converter which should also work under 64-bit Windows versions. Operation is similar to the old wspr-8 and -32 utilities
http://df6nm.bplaced.net/wspr/slowWSPR.zip
which also contains a description worth reading. The "waitraster" program has been replaced by a DOS loop based on the %time% variable.

To run the attached script, you will need to rename it from .txt to .bat. Then download DL4YHF's sound utilities from
http://www.qsl.net/dl4yhf/soundutl/sndutl_e.zip
and the little sleep utility
http://df6nm.bplaced.net/VLF/fec_tests/sleep.zip

The speeded-up audio should be routed from sndoutpt to WSJT-X using a virtual audio cable, e.g. VAC or VB-Cable from Vincent Burel
https://www.vb-audio.com/Cable/
This needs to be selected as the default audio output device, as well as the WSJT-X input device.

Note that
- reported frequencies will be translated from 137600-137625 to 137400-137600 Hz,
- reported timestamps will be 15 or 16 minutes late,
- reported SNR will be 9 dB too good.

I have to admit that the script hasn't been tested yet, so there is no guarantee that it will work at all. If I was to do this again, I'd probably not go down the speeded-up audio route, but rather make a script which calls wsprd.exe directly.

Good luck,
Markus (DF6NM)



-----Ursprüngliche Mitteilung-----
Von: DK7FC <selberdenken@posteo.de>
An: rsgb_lf_group <rsgb_lf_group@blacksheep.org>
Verschickt: Do, 15. Nov. 2018 11:26
Betreff: Re: LF: Re: WSPR-15

Hi Markus,

Congrats to last nights successes!

Well, since WSPR-15 is certainly the better mode for LF DX, and the WSPR team is not interested in LF, will you find the motivation to release an updated version of your utilities that enable 64-bit users to receive WSPR-15 using the latest WSPR-2 decoder?

73, Stefan



Am 15.11.2018 00:49, schrieb Markus Vester:

> maybe they don't work on 64 bit machines?

yes that is the problem. These "tools" are just a simple DOS script. All the real work is done by the audio utilities from Wolfgang DL4YHF, which are running fine as ever. But my little "waitraster.exe" utility, which did nothing but hold the script until the beginning of a desired minute, will no longer work under Win-64.

Best 73,
Markus

-----Ursprüngliche Mitteilung-----
Von: N1BUG <paul@n1bug.com>
An: rsgb_lf_group <rsgb_lf_group@blacksheep.org>
Verschickt: Mi, 14. Nov. 2018 23:03
Betreff: Re: LF: Re: WSPR-15

I am interested in this too, but if I recall there was a reason I
didn't install those tools. I can't remember now... maybe they don't
work on 64 bit machines? If that was it, I am out of luck.

My WSPR2/WSPR15 transmitting tests last season showed that WSPR15
was much better than WSPR2... even though most WSPR2 listeners were
using the new improved decoder and WSPR15 listeners were using the
old WSPR-X.

73,
Paul

On 11/14/18 4:09 PM, jrusgrove@comcast.net wrote:
> Stefan
>
> Vague memories ... the tools, if they still exist, would be on an old computer that's not so easy to
> resurrect.
>
> Yes, it seems the current WSPR2 decoder is much improved. Think your time might be better spent on
> that mode since there are already many listeners. It would be interesting to see how far west into
> the US you could be decoded on a good night. Just my opinion ;~) .
>
> Jay W1VD
>
>
> ----- Original Message -----
> From: "DK7FC" <selberdenken@posteo.de>
> To: <rsgb_lf_group@blacksheep.org>
> Sent: Wednesday, November 14, 2018 1:51 PM
> Subject: LF: WSPR-15
>
>
>> Jay, LF,
>>
>> I already have the next idea, maybe for tomorrow? ;-)
>>
>> You remember the early days before K1JT released WSPR-15.
>> Markus developed the slow-WSPR tools and we (you, as the DX RX station ) tried WSPR-8 and even
>> WSPR-30 successfully.
>>
>> Using Markus tools, the slowly transmitted WSPR (with accurate FSK spacing) was accelerated
>> replayed in the next time slot and then fed to the normal WSPR-2 decoder, beeing connected to the
>> database.
>>
>> Meanwhile the WSPR-2 decoder performance seems to have much improved. So, it may be interesting to
>> try the 'old' tool of Markus together with the latest WSPR-2 decoder.
>>
>> What do you think? Do you still have the config files and so on?
>>
>> Worth a try?
>>
>> 73, Stefan
>>
>> PS: 137.620 kHz is right in the WSPR-15 defined slot. I think this is why my coil is resonant
>> there ;-)