Horus on low end laptop problems.

[Steve]

I've been trying to get Horus working on an old DELL windows 10 laptop -
I've tried both  horus.exe and webhorus on chrome - both only partly
work - maybe 25% of packets are decoded correctly.

Its a bit under-powered (2.4GHz P8600) 4GB ram, SSD - but for horus.exe
RAM is only showing 60% utilized and CPU is showing about 20%

Ive tried various soundcard settings, wifi on/off, checked the receivers
on other PCs & macs (all OK).

I'm thinking that its just not powerful enough - but the CPU & RAM
utilization are not really suggesting a problem.

Any clues/suggestions/things to try folks?

    Steve



--
This email has been checked for viruses by AVG antivirus software.
www.avg.com

[Mika Köching]

Decoding Horus also works on raspberry Pis so I think it should work on that hardware. I would just try with Linux (Debian).

[Mark Jessop]

I guess it's possible you could be getting sample drops somehow, but with that CPU utilisation I wouldn't really expect it.

I'm guessing you're running v0.6.1 of Horus-GUI, as I think 0.6.0 didn't support CPUs that old. We had to revert to an older version of Numpy to enable it to work on CPUs without AVX extensions (older than ~2009).

That you are decoding OK on other machines is a good sign - it suggests that at least the transmitter is behaving correctly.

Unfortunately at this point it gets pretty difficult to debug what's going on. 

To check out whether there's some weird audio bug, you could perhaps try recording a sample on that laptop and playing that back (through something like VBCable) on one of your other working machines?

73

Mark VK5QI

--
You received this message because you are subscribed to the Google Groups "UKHAS" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ukhas+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/ukhas/2d39517e-ba09-4886-ba15-3757545bf69fn%40googlegroups.com.

[Steve]

Yes v0.61.   Its a Dell Latitude E6400  - turns out the original release was 2008.  The E6400 based on the Intel Core 2 Duo P8600 @ 2.40 GHz and that doesn't support AVX extensions - so that will be the problem.

I didn't realize the laptop was that old.  

I did try 0.6.0 (as that supports Horus V3 and doesn't need AVX) - but that doesn't run.  In any event its no use running this on a PC that cant take a future upgrade. 

So time to shell out rather than muck about I think.

    Steve

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CABD2Lt2_2odO0J3%3DZ6ga-FbqNLK293i4n77kDogpu3iZ%3DLKQUQ%40mail.gmail.com.

Virus-free.www.avg.com

[Matt Taylor]

Hi Steve,

I was just about to put one of my laptops on eBay - a Dell XPS 15z (P12F). It's far from new, but would be a cheap but decent upgrade from what you have now :-) Let me know if you're interested

Cheers,

Matt M0XIN

To view this discussion visit https://groups.google.com/d/msgid/ukhas/2733b395-47ee-4801-bcf5-1dac2b2ad796%40randomaerospace.com.

[Mark Jessop]

"The E6400 based on the Intel Core 2 Duo P8600 @ 2.40 GHz and that doesn't support AVX extensions - so that will be the problem."

No - my point here was that we modified library dependencies in v0.6.1 so we could actually support these older machines. If the machine didn't support the software then it wouldn't start at all (like how v0.6.0 doesn't start at all).

I still think the issue is something relating to audio, and possible sample drops.

73

Mark VK5QI

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAKH8nALEBekKPjFB-MwJ1EuSGp%3DFsD6wGDncchW3s-VRC-YdOQ%40mail.gmail.com.

[Dave, G1OGY]

Agreed regarding the E6400's capabilities.
I have 3 - don't ask - both (Debian) linux boxes run MSHV for "JT" modes without breaking a sweat and the Win 10 box copes with WSJT-x without issue too.
Perhaps it is worth mentioning that the Mic socket is stereo; with FET mic bias power.  I wire my attenuator adapters as dual-channel from the mono transceiver source.
HTH

Dave, G1OGY

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CABD2Lt39cGXSyNrhkLJCtjpS7b0KtzhiJ-U3OOr5tC-3Jjnxhg%40mail.gmail.com.

[Steve]

I recorded a section of 10+ mins of audio on the E6400 using Audacity (44,100 WAV) - transferred it to my desktop and decoded with 0.61 Horus - similar results (only some packets were decoded) .

This points to the Mic/Line In socket or sound driver on the DELL laptop as being the problem.  

I'm driving the E6400 Mic/Line in from an FT817 DATA port (fixed level AF). That has a DC blocking cap on the output - but I'll add a 4.7K resistor to ground to give a DC path for the mic bias and see if that makes any difference.

    Steve

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAKOSXUFiMc7zyEsCxdnnXjSxR15bo3-2C3UfW0dyG05vxNFQMg%40mail.gmail.com.

[Steve]

The 4.7K resistors (one on each channel) didn't make any difference.   I listened to the recorded audio and there are clicks an occasional buzzing - not loud but I think that sort of points to the audio drivers.

I'm using the Generic Microsoft drivers - I don't think I'm going to find a manufacturer (IDT) suitable driver for Windows 10.

    Steve

To view this discussion visit https://groups.google.com/d/msgid/ukhas/f07d7f06-66ea-4bd7-b0e9-7a4d840c6c47%40randomaerospace.com.

[Oliver de Peyer]

Is there a mismatch in the sampling rates?
I am having a similar problem with an unrelated ham radio project and one of the suggestions I found online was that it could be e.g 44.1Khz outbound audio vs 48Khz inbound etc (or vice versa)

This is only a thing of course if both the devices are digital soundcards - but that is your example below, no? (e.g Audacity to Horus)

For comparison, my use case is a QMX transceiver, which presents itself as a USB soundcard at a 48Khz sampling rate, connected to a Linux laptop which was set up at 44.1Khz.
I am trying to change this to 48 but nothing has worked quite yet!
The advice I had was that eventually the two devices would be completely out of sync and one would be listening for several cycles whilst the other would not be generating any sound in sync, etc. This would of course come and go.

Hope this is helpful or at least makes sense

Oliver
M0LVR

[Ross G6GVI]

Steve, don't you get a choice of drivers on the pull-down menu in Horus GUI?

In the case of this (Dell i3 desktop), all of the "Microphone" options work, but on an old ThinkPad (i5?) laptop, I got all the way down the list to the very last option before I found one which worked properly!

I've used the AF output from the 6-pin mini-DIN on the back of my FT817, FT897 and now IC910 and that's always worked well into a PC mic socket.

The other thing to check is in the Sound Settings, Additional Device Properties menu that you're not over-driving it with "Boost" or have any unwanted "Enhancements" enabled:

Noise Suppression or Echo Cancellation would surely mess up a data signal?

73 de Ross

[Steve]

Thanks - I have tried all of the Horus MIC/Line choices and all result in missed packets - on further testing one of the drivers (Direct Sound) results in less missed packets - maybe decoding 50-60% of packets.  I'm watching all the packets get decoded correctly on another PC running WebHorus.     This lends weight to the device driver problem theory.

I get the Additional Properties menu (there is no microphone boost applied, gain is similar and horus thinks the levels are good).  There is no Enhancements tab - so I don't think any Effects are being applied in this driver.

    Steve

--
You received this message because you are subscribed to the Google Groups "UKHAS" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ukhas+un...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/ukhas/029615da-5c15-451f-8361-ec3a2b64d2ebn%40googlegroups.com.

Virus-free.www.avg.com

[Dave, G1OGY]

Steve

FWIW my `IDT High Definition Audio CODEC` is MSoft derived, Version 6.10.0.6274

Dave, G1OGY

To view this discussion visit https://groups.google.com/d/msgid/ukhas/1a4edf75-73aa-4d4e-8021-12ca132dc4b8%40randomaerospace.com.

[Steve]

Thanks Dave,  Yep 6.10.0.6274 (but no Enhancements Tab on its Advanced Properties!) - pics below.

And yours is a E6400 ? Have you tried horus (0.6.1) on it?

    Steve

no Enhancements tab: 

This one works best (but still only 50-60% of packets decoded):

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAKOSXUGA7i_pbNkWfwxLnmfCm9LNVe-q4orAXi9g4CfYYZnoAg%40mail.gmail.com.

[Dave, G1OGY]

I haven't run the software, Steve, as I have no way of generating Horus packets, AFAIK.

I'm happy to test it, mind you.  Perhaps there are some .WAV (?) files I can D/load?

Dave, G1OGY

To view this discussion visit https://groups.google.com/d/msgid/ukhas/652c5099-9abf-43ea-8174-cb1b8d602cdf%40randomaerospace.com.

[Steve]

Thanks - here are a couple of audio files:-

    https://drive.google.com/drive/folders/1jrAxbziI72106_SeeKYv-7NXr-taPxXi?usp=sharing

Good.wav is recorded on my desktop PC and when played back on that, works for almost all packets (the ones that don't work are a problem with the beta sonde software where it runs several transmissions together).

Bad.wav is recorded in the P6400 and exhibits the problem when played back on my desktop (I think the problem is audio clicks mid packet).

I've played Good.wav back on the P6400 and most packets are not decoded. 

    Steve

On 18/02/2026 11:51, Dave, G1OGY wrote:

I haven't run the software, Steve, as I have no way of generating Horus packets, AFAIK.

I'm happy to test it, mind you.  Perhaps there are some .WAV (?) files I can D/load?

Dave, G1OGY

On Wed, 18 Feb 2026 at 10:40, Steve <st...@randomaerospace.com> wrote:

Thanks Dave,  Yep 6.10.0.6274 (but no Enhancements Tab on its Advanced Properties!) - pics below.

And yours is a E6400 ? Have you tried horus (0.6.1) on it?

    Steve

[snip....]

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAKOSXUEdf2vA8Cha2B%2BkbiFgG5-NM%3Dv6oA1SJhNtahiG1vkKbA%40mail.gmail.com.

[Mark Jessop]

"I've played Good.wav back on the P6400 and most packets are not decoded. " 

How are you playing this back on the P6400?

Looking at the 'Bad' file, it looks like the clicking might actually be dropped samples somewhere in the audio chain. This would really screw up the timing estimator in the modem.

Cheers,

Mark

To view this discussion visit https://groups.google.com/d/msgid/ukhas/180cc2ae-bdbd-487f-8854-ed187c39178e%40randomaerospace.com.

[Steve]

I'm looping back the physical audio - line out to line in with a short cable.  I was running Media Player and Horus at the same time. 

I'm thinking probably dropped samples too.

There are a few posts and videos on line - one of which suggests that the click & pop problem occurred after windows 10 update 1803 (so I'm guessing Dave is running pre that?)

My E6400 is on 1905,  The problem being discussed is on audio output - the fix is to disable eSATA - I'll give that a try tomorrow.

    Steve

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CABD2Lt3da3mr4QzFjJc7%3DUOegEcTtrwJOuee_O7AMCLR5b1p3g%40mail.gmail.com.

[Dave, G1OGY]

Edition:  Win 10 Pro N

Version: 22H2

Build:     19045.6466

As updated as possible prior to the cut-off.

Regarding the recordings: I too used loopback - Headphone-Out cable-linked to Mic I/P

I think there is a fair amount of clipping going on.  The Spectra plot peaks up to at least -30dBFs at the start of each packet before dropping back to -60/-70 during packet duration.  Could be FT-817 AGC action.  What setting do you have for AGC?

Clipping appears on both `Good` and `Bad`.

`Bad` has far more (background) receiver noise - for the same audio adjustment the software often complained of LOW Input level between packets during `Good` playback. Background level drop out was not evident - and the Spectra display underlined the fact - on `Bad`.

Even so, I would expect the software to be able to cope with an inherently `noisy` channel as long as the tones are clean. 

I have not counted the packets in each recording but it seemed to me that `Bad` had packets closer together than `Good`.  Anyhow, the CSV decode logs are attached; I guess you'll be able to see what is missing - it does not look good for `Bad`!

I noticed an RS41 flight off the N Kent coast on SondeHub just after I ran up the software this afternoon but I was unable to get to the shack to set up a receiver.  Now I have the software running on the E6400/Win10 I'll keep an eye on what might be in range.. 

My e-SATA port is "Enabled" - its a default, apparently.

Dave, G1OGY

To view this discussion visit https://groups.google.com/d/msgid/ukhas/b887fc3f-7e52-479c-9f93-fd59c84629c8%40randomaerospace.com.

[Mark Jessop]

"I noticed an RS41 flight off the N Kent coast on SondeHub just after I ran up the software this afternoon but I was unable to get to the shack to set up a receiver.  "

If you're seeing this on sondehub.org or tracker.sondehub.org, then this is a 'stock' RS41 launch by a meteorological agency. The RS41 will be running its normal firmware, which is *not* decodable using Horus-GUI. It will also be transmitting on the 400-406 MHz radiosonde band, not the amateur or ISM bands. To decode these, you need to be running other software like radiosonde_auto_rx, rdz_ttgo_sonde, dxlAPRS, or SondeMonitor.

 

Horus-GUI will only decode "Horus Binary" modulation signals, which can be transmitted from a range of payloads, including *reflashed* RS41s.

73

Mark VK5QI

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAKOSXUEZrKemH4kX_ScjzYzPuLewEoTUES5bdYJte_oJRP2xBg%40mail.gmail.com.

[Dave, G1OGY]

Ah! OK, Mark.

Thanks for saving my time.

Dave, G1OGY

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CABD2Lt0ZF_ME0ruB3aFbnZT5AbPqMNp8cuPXPj6COpmKGeLE0Q%40mail.gmail.com.

[Steve]

FIXED IT.    

Disabling the eSATA ports in the BIOS fixed the problem.   There are several references online to a similar problem (Win10 E6400 clicks in audio output) - evidently became a problem in Windows 10 build 1803 and was never fixed before Win 10 end of support.   There is a different fix which involves installing old SATA drivers - but I don't think I'll bother as I have no plans for using the external eSATA port.

I'm running Windows 10 Enterprise on it (10.0.19045) - so that may be why my audio driver doesn't display an enhancements tab.

Thanks to everyone that helped.

    Steve G8KHW

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAKOSXUEZrKemH4kX_ScjzYzPuLewEoTUES5bdYJte_oJRP2xBg%40mail.gmail.com.

[Dave, G1OGY]

Good news.

Were the clicks and bumps made on the recording or on the playback?

My e-SATA is enabled but I don't need it, will I get better results from the WAVs if I turn it off?

Dave, G1OGY

To view this discussion visit https://groups.google.com/d/msgid/ukhas/f42e82bc-299c-4ab9-81f6-2aee11e2b989%40randomaerospace.com.

[Dave, G1OGY]

One other point of note is that `Good` is in stereo and `Bad` is Mono.

Dave, G1OGY

[Steve]

They were on recording (you can hear them on the "Bad.wav" recording:   https://drive.google.com/file/d/1Tt3BEa2wxyEkpZljCplfZcdxRbgr1p2U/view?usp=drive_link)   

The E6400 problem referred to online is about clicks in audio output - but it seems to apply to the input too.  I suspect the clicks are actually caused by missing samples (my guess would be due to SATA driver interrupts). 

    Steve G8KHW

To view this discussion visit https://groups.google.com/d/msgid/ukhas/CAKOSXUGv_%3DAyvYk7zF33QZCM%2Bjfau5UosMtXDuH00BXuQQa-iw%40mail.gmail.com.

[Steve]

For completeness I've added a recording from the P6400 - after the fix - demonstrating no clicks - Called Fixed.wav:

    https://drive.google.com/file/d/1ugVykePi1vLU2XUX6rVjd94-sMCyjQDK/view?usp=drive_link

    Steve

To view this discussion visit https://groups.google.com/d/msgid/ukhas/7d1d10af-0432-4513-833a-468025dc8b86%40randomaerospace.com.