Another Pluto with TX stutter on V3.0.10 (but OK with SDR Angel) #pluto
|
HBottiki
Hi,
I am just reading through the various threads here on this group regarding the stuttering interruptions on TX with ADALM Pluto. Having the same symptoms here that some others commented as well.
I already tried all sort of tricks: Different PC, using USB 2.0 vs. 3.0 ports on same machine, Powered USB Hub, second USB power cable, etc. etc. (also tried additionally FW update to Pluto, frequency hack and second core activation hack)
Running the Pluto on minimum RX BW (550kHz) and also tried increased TX buffer settings to 250ms with no major improvement.
Tried machines are HP Desktop PCs with i5 3470 and 4570 processors and CPU load is quite low while SDR Console runs (< 10% CPU load most of the times) Maybe worthwhile to mention that they both all still running on Win7, 64 bit.
Now, to my astonishment, I tried SDR Angel (Windows version) tonight and with SDR Angel the transmission is perfect…no TX stuttering at all, even at higher RX BW (2MHz) which should cause much higher load on the USB port. I also did some fullduplex tests via QO100 and had no issues at all with any TX stutter. All smooth with this other software.
Question to Simon:
73,
Oscar DJ0MY |
|
|
|
And the screenshots are where? Please attach screenshots si I can see your Pluto in action and I must see the Pluto settings.
Note: I do not steal code from other people.
Simon Brown, G4ELI
From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> On Behalf Of HBottiki … Question to Simon: |
|
|
|
Siegfried Jackstien
i can use pluto full duplex on atom tablet with 4 gig ram and
1.8gig clock... rx and tx buffers set both to 20ms only!!! (simon
what do you say?!?!? ... underpowered pc???) only thing i did is i increased the tap numbers in rx and tx
filters (so i still have a bit of latency but not added in the rx
tx buffers but the filters that also have a buffer area .... simon
sure knows what i did ... ) works with 3.0.10 and 3.0.11 (20th so newest) but i still need to use that "workaround" with another app on taskbar to let sdr console run smooth and react fast on the ptt (tx) button and have no breaks in rx (expecially in the rx to tx transtion it often hangs without that workaround) BUT WITH IT .... it works as fast as on a much faster machine .... (seems as i still have to fooool windows a bit ... but it works ... so what?!?) 2nd cpu in pluto activated .... pluto on external psu ... 3m usb cable (but good quality) to that tablet ..... is my all day long setup here ... sometimes i let console
run a few days not switching off ... no problem ... (only switch
of psu of the poweramp) ... i thoink that tooo long bufers do not help in the stuttering .... longer buffer also means MORE DATA .... try also shorter combinations i had good results with shorter settings ... good luck . (maybe an antivirus software slowing all things down?!?) dg9bfc sigi Am 23.07.2019 um 19:27 schrieb
HBottiki:
|
|
|
|
jdow
The trick is to open settings->system->power & Sleep.
toggle quoted message
Show quoted text
"Additional power settings" is on the right 1/2 of the way to the bottom (feedback) entry. Open "Show additional plans". Select High performance. You can then "Change plan settings" and on the new panel "Change advanced power settings". Then make sure "Processor power management" entries Minimum processor state and Maximum processor state are set to 100%. And if you are writing to disk you might improve performance slightly with the "Turn off hard disk after" setting. Set it to 0, which leaves it running all the time. Obviously this is for when you are on external power. {^_^} On 20190723 14:02:40, Siegfried Jackstien wrote:
i can use pluto full duplex on atom tablet with 4 gig ram and 1.8gig clock... rx and tx buffers set both to 20ms only!!! (simon what do you say?!?!? ... underpowered pc???) |
|
|
|
HBottiki
Hi Simon,
which exact Pluto settings you want in a screenshot? Just the main GUI showing the RX/TX parameters? Or any of the options window? Or all of them? And I didn mean stealing code but just looking under the hood to get some inspirations for a totally independed modification :-) 73, Oscar DJ0MY |
|
|
|
HBottiki
HI,
here we go with the requested Pluto Screenshots of the relevant settings where something was activated....(mostly default settings, except the buffers that I increased) Also the sliders of the TX DSP are adjusted during TX as to get sufficient output to drive the AMP. Also tested half-duplex instead of full-duplex, but sounds the same. Tested on another frequency range, e.g. 70cm band in FM and SSB and received with analog TRX - same stuttering on TX occurs. The stuttering freqency increases with higher buffer setting (as someone commented here before me) sounds almost like 20ms buffer = 20ms stutter and 250ms buffer = 250ms stutter (just judged by ear not by oscilloscope, hi) Energy options is set to high performance (100% CPU) and no energy savings plan. 73, Oscar DJ0MY |
|
|
|
Hi Oscar,
The Radio window – at the bottom left of the GUI screenshot. Just make sure I see it all. There shouldn’t be any stutter – your CPU load is low, you have a very good PC!
Simon Brown, G4ELI
From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> On Behalf Of HBottiki
Hi Simon, |
|
|
|
HBottiki
Hi Simon,
here you go - bottom left of the GUI scrolled down...most features are OFF and Pluto temperature just at room temperature (but also happens when it warms up during longer TX periods and gets around 35-40 degrees.... 73, Oscar |
|
|
|
jdow
You may get better dynamic range with the manual RF gain set lower. Mid 30s come to mind.
toggle quoted message
Show quoted text
{^_^} On 20190724 07:36:20, HBottiki wrote:
Hi Simon, |
|
|
|
Oscar,
PLEASE do not cut off part of a screenshot, always show the whole image.
Anyway, while I was out shopping this afternoon I remembered that someone else had this problem a few months ago and it was solved with the test team members, but I don’t remember what the solution was. I assume you tried other USB ports?
What I would like to see is the logfile after you have experienced the stutter problem – which is more a gap at the end of each buffer I think.
Simon Brown, G4ELI
From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> On Behalf Of HBottiki
Hi Simon, |
|
|
|
jdow
Increase buffer sizes?
toggle quoted message
Show quoted text
{^_^} On 20190724 08:49:11, Simon Brown wrote:
Oscar, |
|
|
|
HBottiki
Sorry JDOW.
my LNB input is highly attenuated before entering the Pluto , so this kind of gain is really needed here in my setup. If I remove LNB attenuation of course I reduce the LNA gain.... The radio is VERY FAR from being overdriven. 73, Oscar |
|
|
|
HBottiki
Sorry SImon,
I thought the rest you already checked in my other screenshots, thats why this time I took only the area you specifically requested. Would you still like to see the whole thing? Would you like a Video recording of the stutter? (can post a short video clip via Twitter) I will get you the log file. Just taking the log after a stuttering TX sequence should be OK? The stutter is on every transmission continuesly - we do not talk about a sporadically ocurring problem. Therefore your assumption of a gap after each buffer is also my guess, becasue that explains why at short buffer settings the stuttering frequency is fast and with largest buffer setting the stuttering frequency is somewhat slower. Would it help if I connect an audio scope software and analyze if it exactly correlates with the set buffer time? 73, Oscar |
|
|
|
Oscar,
Please slow down. As I wrote, we had this problem and it was fixed, I just don’t remember the scenario. I’ll invite you to the test team soon, there we can look at it in peace. It’s a bit hot now and I must concentrate for the next few days on some coding.
Simon Brown, G4ELI
From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> On Behalf Of HBottiki
Sorry SImon, |
|
|
|
jdow
It should not be. Adding gain in the Pluto is not needed to keep a decent system noise figure. When you say "highly attenuated" I ask what IS the gain between the input between the first amplifying stage the antenna feeds and the output of the coax feeding your Pluto. If it is not a positive gain making this up in Pluto is impossible. Your noise figure is already set by that point in a good system.
toggle quoted message
Show quoted text
So, why do you have LNB attenuation in there? Why do you feel you must make it up? Have you reduced Pluto's gain and seen signal drop relative to the noise level? (If they drop exactly the same you can reduce Pluto RF gain again. Repeat until the signal to noise ratio drops. That is the first attenuation step that actually hurt anything.) {^_^} On 20190724 12:05:58, HBottiki wrote:
Sorry JDOW. |
|
|
|
HBottiki
I will not discuss RX chain further as this is not related to the TX problem at all - there is plenty of remaining dynamic range as on this 10GHz QO-100 satellite downlink the strongest SNR will be around 30dB with this setup so if you pay attention to the existing noise baseline level in my screnshot the Pluto is VERY FAR away from being anywhere overdriven and due to the typical SNR levels will NEVER be overdriven for this specific use case.
73, Oscar |
|
|
|
HBottiki
Hi Simon,
as requested here is the log file. 73, Oscar |
|
|
|
Hi,
I’ve invited you into the test team – please post there, saves cluttering up this group and somewhere in the test team messages we resolved this problem.
I didn’t see any error messages…
Simon Brown, G4ELI
From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> On Behalf Of HBottiki
Hi Simon, |
|
|
|
andr3wz@...
Hello I have the same problem. Is it possible to know the issue?
Thanks Andrea IU6DEV |
|
|
|
Which version are you using?
From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> on behalf of andr3wz via groups.io <andr3wz@...>
Sent: 29 January 2023 16:08 To: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> Subject: Re: [SDR-Radio] Another Pluto with TX stutter on V3.0.10 (but OK with SDR Angel) #pluto Hello I have the same problem. Is it possible to know the issue?
Thanks Andrea IU6DEV -- - + - + -
Please use https://forum.sdr-radio.com:4499/ when posting questions or problems. |
|
|