Oscilloscope Music on DSOs- Post your Results

[Martin72]

Hi,

The topic seems to be of interest at the moment, I also think it's a nice gimmick and it's a shame that the results will "sink" into the respective oscilloscope threads.
Hence this thread for it.
Post your results, name the settings, the "music" material used...
I will post my results with the SDS824X HD here again this weekend.

Martin

[KungFuJosh]

Here's a copy of my test from the 2kX+ thread:

I bought a copy of OsciStudio to test this out, and eventually make a very stupid video.

In the meantime, it was useful to test both refresh rates and audio frequency effect on the XY results:




The first section is more obvious, that's scrolling through refresh rates (first 50 seconds). The last 30 seconds is screwing around with the frequency of the shape. Chris's tutorial shows him using A1 at 55Hz, which is (one of the reasons) why my results aren't as great with his music. A4 at 440Hz (and really most notes over ~100Hz) show much greater stability on my scope.



Here's my original video with the music:


10kpts for the memory to keep it faster/smoother. Going up higher either adds weird thickness, or slows it down. Timebase: 10bits, 100us/div, 10kpts, 10MSa/s; channels at  DC50, 300mV/div.

[tautech]

Member TopQuark's excellent work with SDS2504X HD

Inspired by KungFuJosh (https://www.eevblog.com/forum/testgear/siglent-sds2000x-plus-coming/msg5424095/#msg5424095)

I tried making oscilloscope music with the SDS2000X HD, the results are great, much better than I anticipated.

Scope settings:
Input settings (CH1, CH2) : 50-ohm termination, DC coupled, 20M BW limit
Timebase : 1 ms/div, 1 MSa/s, 10 kpts memory
Trigger : Normal, alternating edge, CH1, AC coupled, 0V offset
Acquire : XY mode (duh!)

Reference video produced with Oscilloscope! on a Macbook Pro 14
https://oscilloscopemusic.com/software/oscilloscope/

Audio DAC : Fiio KA3 @ 96.0 kHz, 24 bit, 3.5mm stereo output into coax and scope 50 ohm input

[Martin72]

This looks even better than the SDS800X HD.
What might be the reason for this, the sound file, its resolution?

[Anthocyanina]

oscillofun by Atom Delta on AD2, settings are visible on screen, auto trigger, 220mV/, 2ms/, 8k points at 400ksps



oscillofun on Owon VDS1022i, settings also visible, auto trigger, 500mV/, 1ms/, 5k points at 250ksps



potential is forever by Chris Allen on AD2, settings are not visible, no trigger, 400mV/, 2ms/, 8k points at 400ksps



not music but a chaotic attractor, which also makes cool XY plots, AD2, settings also visible, auto trigger, channel 1 200mV/, channel 2 1V/, 2ms/, 16k samples at 800ksps

[Anthocyanina]

This looks even better than the SDS800X HD.
What might be the reason for this, the sound file, its resolution?

the sound file you use matters. using the playback from youtube i've seen distorted shapes. i've had the best results using the .wav or .flac files they provide. it is also important to not have any equalization or other DSP done as that will also result in distorted shapes

[tautech]

This looks even better than the SDS800X HD.
What might be the reason for this, the sound file, its resolution?

Not obvious is applying the scope settings TQ used....fixed sampling rate.
Set Mem depth first (down near Keysight levels) then change Mem Management to Fixed Smpl Rate.

Example attached.

[tautech]

Another attractor vid from a while back in post #1432 in SDS2000X Plus thread.

[TopQuark]

From the SDS2000X HD thread:

Yo dawg, I heard you like oscilloscopes. So we put an oscilloscope in your oscilloscope, so you have waveforms in your waveforms.  

I think this demo video is even more impressive than the two videos I put up yesterday. The complexity of the graphics is impressive, and I think the SDS2000X HD rendered the it really well.



Then there's this one. This is my favorite, and I think I have finally nailed down the best settings to use. Not to mention the graphics of this music is the coolest of all.

[TopQuark]

In case anyone is wondering, here's the physical setup of the "measurement". Two pigtail coax soldered to 3.5mm jack, and connected to 50 ohm terminated scope input.

Using the high quality source music file (usually .wav) provided by the artist is very important. Play it back at the highest possible bitrate and resolution. The audio from the youtube videos won't get you clean results.

Audio DAC : Fiio KA3 @ 96.0 kHz, 24 bit, 3.5mm stereo output into coax and scope 50 ohms input.

Scope settings:
Input settings (CH1, CH2) : 50-ohm termination, DC coupled, 20M BW limit
Timebase : 1 ms/div, 1 MSa/s, 10 kpts memory
Trigger : Auto, alternating edge, CH1, AC coupled, 0V offset
Acquire : XY mode (duh!)
Display : Color grading on

Timebase is interesting, I find C.Allen's music to work best at 1 ms/div, 1 MSa/s, 10 kpts memory.
Whereas Jerobeam Fenderson's music seems to work best at 2 ms/div, 500 kSa/s, 10 kpts memory.

[tautech]

WOW just WOW !     

DSO's are shit at XY mode, yeah right ! 

[DimitriP]

WOW just WOW !     

DSO's are shit at XY mode, yeah right ! 

Gotta love it when digital finally is able to do what analog could do all along. Just sayin'

[tautech]

WOW just WOW !     

DSO's are shit at XY mode, yeah right ! 

Gotta love it when digital finally is able to do what analog could do all along. Just sayin'

Sure but this shows the days of analog scopes are done.
Why even would anyone still own one when a DSO has so much more to offer. 

Nearly 15 years back I could see the writing on the wall for CRO's even with the old lunchbox Tek DSO's and way before I'd even heard of Siglent.
My, haven't things come a long way in just a few years.....

[Grandchuck]

Here is a video that marks the progress

https://youtu.be/1tTeXPmbxW0

[TopQuark]

I found one trick that made graphics rendering more consistent is to feed a fast waveform (e.g. 1MHz sine wave) to a spare input as the trigger source, to force as fast an update rate as possible on the scope. This ensures as little dead time as possible between triggers, and that the waveform update rate is consistent.

Also, this avoids situations where the graphics is just a dot, or a vertical/horizontal line, the waveform does not update unless in auto trigger mode.

The improvement in graphics quality from this arrangement kind of makes sense, as XY mode on analog scopes is just a completely continuous vector plot without triggers. So by actively forcing the waveform acquisition to trigger unconditionally with as little dead time as possible, the graphics rendering would be the closest to what an analog scope could produce.

By feeding a 1MHz signal from the scope built-in AWG to the external trigger input, I could achieve 46.9 wfm/s , when the timebase is set to 10k sample memory, 500kSa/s, 2ms/div. The shroom on my screen looks really nice.

Side note 1: Listening to the original quality audio tracks with earphones is a trippy experience, the data modulation audio pulses hits your ears like no other music. I imagine this is what the life of a RF demodulator sounds like.

Side note 2: A discussion on analog graphics rendering capabilities of digital oscilloscopes, what a time to be alive. 

[Anthocyanina]

I found one trick that made graphics rendering more consistent is to feed a fast waveform (e.g. 1MHz sine wave) to a spare input as the trigger source, to force as fast an update rate as possible on the scope. This ensures as little dead time as possible between triggers, and that the waveform update rate is consistent.

Also, this avoids situations where the graphics is just a dot, or a vertical/horizontal line, the waveform does not update unless in auto trigger mode.

The improvement in graphics quality from this arrangement kind of makes sense, as XY mode on analog scopes is just a completely continuous vector plot without triggers. So by actively forcing the waveform acquisition to trigger unconditionally with as little dead time as possible, the graphics rendering would be the closest to what an analog scope could produce.

By feeding a 1MHz signal from the scope built-in AWG to the external trigger input, I could achieve 46.9 wfm/s , when the timebase is set to 10k sample memory, 500kSa/s, 2ms/div. The shroom on my screen looks really nice.

Side note 1: Listening to the original quality audio tracks with earphones is a trippy experience, the data modulation audio pulses hits your ears like no other music. I imagine this is what the life of a RF demodulator sounds like.

Side note 2: A discussion on analog graphics rendering capabilities of digital oscilloscopes, what a time to be alive. 

This makes a lot of sense, i found i get better results with trigger set to "none" on my AD2, and that's what i used for potential is forever! have you tried with trigger set to auto but placed at a higher voltage than any of the two channels will ever reach?

[joeqsmith]

Thanks for starting this thread.  Some of these are very impressive.

I found one trick that made graphics rendering more consistent is to feed a fast waveform (e.g. 1MHz sine wave) to a spare input as the trigger source, to force as fast an update rate as possible on the scope.

Maybe there is a better way to detect the update rate (of the screen).   Manufactures seem to post the number of sweeps or waveforms they can store, but not how fast they can display the data. 

[KungFuJosh]

It was suggested that triggering from a 3rd channel might help with performance. I setup CH3 with a 1MHz sine wave, and set the edge trigger to that channel. I activated and deactivated my AWG and saw no change whatsoever with the performance. Triggering in general hasn't shown to have an effect here IME. At least that's how it is with my scope. 🤷

[jjoonathan]

RTO1024 YouScope

[KungFuJosh]

I did another quick test showing object frequency from OsciStudio. I went from A1 - A4, in sequency that's 55Hz to 110Hz to 220Hz to 440Hz:

[TopQuark]

Maybe there is a better way to detect the update rate (of the screen).   Manufactures seem to post the number of sweeps or waveforms they can store, but not how fast they can display the data.

Not a completely scientific measurement, bu you can check the "Siglent SDS2504X HD - Dancer" video which was shot at 60 FPS. If you pause the video and advance it frame by frame by pressing , and . on your keyboard, you'll see the screen does update at 60Hz, with new content on the screen every video frame.

Edit:

Maybe there is a better way to detect the update rate (of the screen).   Manufactures seem to post the number of sweeps or waveforms they can store, but not how fast they can display the data.

Another question that could also be asked is how well the multiple acquisitions made between every screen update are presented to the user on the screen. Scopes with good "digital phosphor" tech will stack the multiple acquisitions made during each screen update, and use brightness intensity to convey how often the multiple waveforms overlap a single point on the screen.

I don't think LCD screens on most scopes with 8-bit color depth and uniform backlight brightness can match the display dynamic range on a phosphor CRT, but I guess one could use antialiasing techniques and good gamma curves to convey more of the frequency-of-occurrence information within multiple stacked acquisitions. Color intensity grading can also boost the effective dynamic range of the display by using all the color channels I guess.

In an ideal world, my scope would have the retina, 1000000:1 contrast ratio, mini-led local backlight, 1600 nits, 144 Hz display on my Macbook, plus the 4.5 million wfm/s update rate on the R&S MXO4, plus a really good digital phosphor implementation, to show me every bit of detail in all the waveform samples. 

I wonder how one could quantitatively measure the performance of an oscilloscope in this respect. But I digress, this thread is about how well scopes can rock and roll.

[Martin72]

Hi,



(1080p avaible)

Excerpt from "Shrooms", 2 minutes, SDS1104X-E, then SDS804X HD.
Nearly the same settings, the difference is huge in my opnion.

[KungFuJosh]

Nearly the same settings, the difference is huge in my opnion.

Definitely an improvement, but I want to see the same excerpt on your SDS3000XHD! (in the next 2 weeks to 6 months, depending on how long it takes to get it 😉😉)

[Martin72]

Not funny....
By the way, the settings:
1ms/Div., 200mV/div each channel, Focusrite 2i2 192khz, wav file (downloaded the complete album) fed in via 50ohm terminators, trigger ch1/ac/each, 10kpts (804)/7kpts (1104X-E).
External trigger like TopQuark suggested was not possible.

[Anthocyanina]

Hi,



(1080p avaible)

Excerpt from "Shrooms", 2 minutes, SDS1104X-E, then SDS804X HD.
Nearly the same settings, the difference is huge in my opnion.

does the 1104 not have high res mode in XY?

thank you!