Author Topic: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown  (Read 23698 times)

0 Members and 1 Guest are viewing this topic.

Offline analogNewbie

  • Contributor
  • Posts: 46
  • Country: cn
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #25 on: April 09, 2016, 02:39:33 am »
I am wondering if this family share the same hardware. :)
AFAIK they surely do, however the 2 channel versions are without some front panel HW and the main PCB will be only populated with enough componentry to support those 2 channels.

Good. I should take a look at the firmware to see what I can do before spend the money. ;D
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3069
  • Country: fi
  • Starting with DLL21
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #26 on: April 09, 2016, 09:06:22 am »
I am wondering if this family share the same hardware. :)

If look SDS2072X to SDS2304X there is same HW (exept possible small component value differencies in analog front end)
As tautech tell  in 2 channel models they can leave some areas without components.  It is much more expensive to produce different boards to different versions.

But then you show your location is in China.

There situation is bit more "complex"

There you have more models in SDS2k series.
You have there full set of these all models
SDS2000  28M(?) memory, 8bit LA (perhaps this 28M is obsolete info and also this have now 70M memory for both 2 channel group also in china version?)
SDS2000X-E  70M  memory, no LA and less memory (memory is 70M for both two channel groups. So 4 channel have 70+70M)
SDS2000X  140M  memory, 16bit LA (memory is 140M for both channels group, so 4 channel model have 140+140M)
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 15977
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #27 on: April 11, 2016, 09:16:17 am »
New FW for the SDS2000X series:
http://www.siglentamerica.com/USA_website_2014/Firmware&Software/firmware/SDS2000X-P33R1.rar

From the website changelog:
1. Improved the user experience on the universal knob
2. Added virtual numeric keypad function in cases inputting large number is possible (push the universal knob to call it)
3. Optimized the persistence display in pass/fail mode
4. Added ASCII decoding
5. Fixed some bugs
a) Pushing the trigger level knob in AC coupled trigger mode does not bring the level back to zero
b) Value jumps up and down when setting baud rate in trigger setting
c) Arrow in decoding list displays abnormally
d) Digital channel display problem
e) All frames are not mapped to the display in sequence mode with frames quantity > 1024

Avid Rabid Hobbyist
 

Offline F5D

  • Newbie
  • Posts: 4
  • Country: fi
    • F5D @ SoundCloud
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #28 on: April 11, 2016, 06:22:46 pm »
I am also eagerly waiting for reports how the 2000X is doing with the new firmware. Does the ERES work? I read from the v2 firmware thread that it does not do anything in the case of the sds2000. If the ERES works and if the FFT mode was a little bit better implemented, I would be all over this at this point, especially due to the discount campaign, but now waiting a little bit before deciding which way to go. Also, would be nice to have png or jpeg screencap instead of bmp.

Actually, what are the differences between the sds2000 and sds2000x regarding software and functionality? Regarding hardware, I noticed they moved the trigger level knob further away from the intensity adjust knob, good move.
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3069
  • Country: fi
  • Starting with DLL21
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #29 on: April 12, 2016, 10:28:51 am »
I am also eagerly waiting for reports how the 2000X is doing with the new firmware. Does the ERES work? I read from the v2 firmware thread that it does not do anything in the case of the sds2000. If the ERES works and if the FFT mode was a little bit better implemented, I would be all over this at this point, especially due to the discount campaign, but now waiting a little bit before deciding which way to go. Also, would be nice to have png or jpeg screencap instead of bmp.

Actually, what are the differences between the sds2000 and sds2000x regarding software and functionality? Regarding hardware, I noticed they moved the trigger level knob further away from the intensity adjust knob, good move.

SDS2000 have 70M memory for 2 channel and 2 x 70M for 4 channel.
SDS2000X have 140M memory for 2 channel and 2 x 140M for 4 channel.

SDS2000 have 8 channel LA
SDS2000X have 16 channel LA

Front panel/UI useability is better in X model. (There is several changes)

I have not tested 2000X but what I have discussed with Siglent there is also some improvement in analog front end.
-------------

ERES works  in SDS2000. 
I will later show some images and some data after I have enough time for do this.

I hope they never add jpg saving. Imho, png is much better. (jpg is not good for tekchical images where is details, it is well ok for usual photographs.)  At this time,  irfan view and batch conversion if more than just pair of images.
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3069
  • Country: fi
  • Starting with DLL21
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #30 on: April 14, 2016, 06:59:33 am »
Does the ERES work? I read from the v2 firmware thread that it does not do anything in the case of the sds2000. If the ERES works...

There are rumors and beliefs - whether intentionally or unintentionally. Sometimes, it may also be the case whether the person does not know how to use the device, or that do not understand what a function is. As in the past, for example with regard to the SDS1000X Dave went completely confused when he introduced the sequence mode in ranting video aka "rewiev".  (or was it lesson: "do not abuse scope")

These tests are from older FW 1.02.01.01.28R1
(new FW is 1.2.1.1.3301)

NOTE: because this topic is about SDS2000X need emphasize that this my test is made using SDS2000.
I can not guarantee SDS2000X is exactly same if do exactly these same tests


In  images a to i signal is same.  In image k signal externally  combined noise is removed and frequency changed but sine level to scope is still 600mVp-p

a
image a is just normal mode with full 2GSa/s samplerate. 1 channel in use.
Signal is 16kHz sinewave where is also added gaussian noise ~0 to ~80MHz (two signals, sine and noise, combined before scope input)
Wfm/s speed is around 970 wfm/s. No detectable difference if use dots or lines and/or sinc on or off or persistence.
(in this image: lines, sinc on, persitence off)

b
same normal mode as a  but samplerate dropped to 100MSa/s  (14k memory. In Siglent, whole memory in use is always mapped to display, not overlapp)
Wfm/s speed is now ~2900. No detectable difference if use dots or lines and/or sinc on or off or persistence.
(in this image: lines, sinc on, persitence off)

c
same as b but mode is PeakDetect
Wfm/s speed is now ~2900 as also in b. No detectable difference if use dots or lines and/or sinc on or off or persistence.
(in this image: lines, sinc on, persitence off)


d, e, f, g, h, i 
same as b exept now mode is ERES 0.5/1.0/1.5/2.0/2.5/3.0
Wfm/s speed is now ~ 7* No detectable difference if use dots or lines and/or sinc on or off
(in this image: lines, sinc on, persitence not selectable)
* this is for 10us/div ERES.


k
Signal 700kHz pure sinewave 600mVp-p  where can see -3dB drop.
Ref level 0dB  as 16kHz sinewave.

(if drop samplerate to 10MSa/s and keep ERES3.0  then (just math) -3dB points also drops to 70kHz
If 100MSa/s and change ERES 3.0 to ERES 2.5 then -3dB points move to 1400kHz  ... and so on...)


« Last Edit: April 14, 2016, 07:11:05 am by rf-loop »
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 17630
  • Country: nl
    • NCT Developments
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #31 on: April 14, 2016, 08:37:39 am »
Two questions:
- Can you zoom in vertically on the Eres signal so you can use the extra resolution to see more detail?
- Is the Eres still limited to only 14kpts of memory depth (aka a checkbox feature with no practical use)?
« Last Edit: April 14, 2016, 08:45:56 am by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3069
  • Country: fi
  • Starting with DLL21
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #32 on: April 15, 2016, 07:34:54 am »
I noticed they moved the trigger level knob further away from the intensity adjust knob, good move.

Intensity adjust knob/multifunction knob. Nearly all parameters need set with this knob. It was previously frusttrating to turn this knob minutes until reach wanted value. If need turn from 0 to 10000 it takes time. Even when there is some acceleration. In some cases also this affect so that fast turning and oops, then after long time you jump over wanted, then back and agen over... exept if do it carefully. Problem with SDS2000 is that it is too close trigger level knob. If y turn multifunction knob and just finger accidentally touch trigger level you jump out from this parameter setting what you was doing.  It was mandatory change what they do for SDS2000X. Also in control panel there is some other good changes.

But now, with new FW multifunction knob different, there is new feature. With some function where need set some value using this multifunction knob user can push it and get small virtual numeric keyboard on the screen. Using this user can fast select exact value..  Of course if there is hardware numeric keyboard it is super good but also this virtual is better than nothing.
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 

Offline Hydrawerk

  • Super Contributor
  • ***
  • Posts: 2380
  • Country: 00
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #33 on: April 15, 2016, 06:30:11 pm »
I am looking forward to Dave's review of SDS2000X.
Amazing machines. https://www.youtube.com/user/denha (It is not me...)
 

Offline radar_macgyver

  • Frequent Contributor
  • **
  • Posts: 380
  • Country: us
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #34 on: May 02, 2016, 06:39:06 pm »
Is there a dedicated thread for the SDS2000X? I recently bought one, and would like to list out a couple of issues I came across in the hope that Siglent would address them in future firmware updates.

Quick example: the waveform update rate is dramatically reduced when signal measurements are enabled. You can see the trigger status indicator flashing between ARM and TRIG'd when measurements are enabled.
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 17630
  • Country: nl
    • NCT Developments
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #35 on: May 02, 2016, 06:44:56 pm »
Is there a dedicated thread for the SDS2000X? I recently bought one, and would like to list out a couple of issues I came across in the hope that Siglent would address them in future firmware updates.
I already went down that road  |O and ended up losing my money on a useless scope. Seriously: just send it back if it doesn't do what you need because hoping Siglent will solve any issue (quick or at all) is just as useful as hoping Christmas and Easter will happen on the same day.
« Last Edit: May 02, 2016, 06:46:44 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 15977
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #36 on: May 02, 2016, 08:06:15 pm »
Is there a dedicated thread for the SDS2000X?
Not as yet.

Quote
I recently bought one, and would like to list out a couple of issues I came across in the hope that Siglent would address them in future firmware updates.
Here's as good as anywhere.

Quote
Quick example: the waveform update rate is dramatically reduced when signal measurements are enabled. You can see the trigger status indicator flashing between ARM and TRIG'd when measurements are enabled.
At what timebase and other settings?
Avid Rabid Hobbyist
 

Offline radar_macgyver

  • Frequent Contributor
  • **
  • Posts: 380
  • Country: us
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #37 on: May 02, 2016, 08:46:59 pm »
At what timebase and other settings?

I started at 100us/div, and 14k memory depth, CH1 enabled, all others off. Memory depth is set to 14k. I hooked up the TRIG OUT to my other scope to try to make sense of what's going on. It seems like its not the update rate, but the dead time that changes. And it doesn't change in a way that makes sense. Under the scope setup listed above, I could see that the dead time without measurements enabled is about 3 ms. Changing the input waveform caused the display to update virtually instantly.


With measurements enabled, dead time jumps to over 40 ms, and the waveform on screen appears to lag the changes in input. I mistook the lag to mean lower update rate. Still, it's very disconcerting to see.


Now, the strangest thing is that when I enable channel 2 (which shares the ADC with channel 1), the dead time once again drops to ~3 ms.


Fiddling with other settings such as memory depth caused some other weirdness to show up in the dead time. For example, at 140k memory depth, the scope does two bursts of acquisitions, then waits for nearly 120 ms (!) before doing another pair. Note that the horizontal timebase (on the 54602B) was doubled for this image.


It almost seems like the scope is trying to sync together several different internal processes to generate the display, and that the process that handles the measurement computations is holding up everything else, including acquisition. The fix would be to decouple these, at the risk that you may have measurements that don't match what the scope is currently displaying.

edit: added inline images.
« Last Edit: May 02, 2016, 08:50:57 pm by radar_macgyver »
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 15977
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #38 on: May 02, 2016, 09:02:29 pm »
The FW version that you're using may have something to do with this.  :-\
Latest:
http://www.siglentamerica.com/USA_website_2014/Firmware&Software/firmware/SDS2000X-P33R1.rar

rf-loop discusses the changes in latest FW and trig-out waveforms:
https://www.eevblog.com/forum/testgear/siglent-sds1000x-series-oscilloscopes/msg920943/#msg920943
Sure it's not the same scope but the FW is very similar.

More reading:
https://www.eevblog.com/forum/testgear/siglent-sds2000-new-v2-firmware/

Members Performa01 or rf-loop might have further explanation.
Avid Rabid Hobbyist
 

Offline radar_macgyver

  • Frequent Contributor
  • **
  • Posts: 380
  • Country: us
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #39 on: May 02, 2016, 09:17:35 pm »
Ah yes, should have mentioned - I did grab the newest firmware available from Siglent, put the .ads file on a USB stick and went through the upgrade procedure. Not sure if it was needed, but there's no easy way to tell since the firmware version page indicates the version numbers of all the individual components inside the scope, but not the overall firmware update file name.

@Siglent: this would be a nice, easy thing to fix, either (1) name the update files to match the numbers shown in the scope, or (2) have a line shown on the scope indicating the overall firmware revision, in addition to the FPGA versions, etc.
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 15977
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #40 on: May 02, 2016, 09:38:52 pm »
Further explanation of your observations are made in this post;
https://www.eevblog.com/forum/testgear/siglent-sds1000x-series-oscilloscopes/msg926963/#msg926963
The acquisition process is outlined and use of dots is recommended to maximise update rates.

Ah yes, should have mentioned - I did grab the newest firmware available from Siglent, put the .ads file on a USB stick and went through the upgrade procedure. Not sure if it was needed, but there's no easy way to tell since the firmware version page indicates the version numbers of all the individual components inside the scope, but not the overall firmware update file name.

@Siglent: this would be a nice, easy thing to fix, either (1) name the update files to match the numbers shown in the scope, or (2) have a line shown on the scope indicating the overall firmware revision, in addition to the FPGA versions, etc.
Could you elaborate more on this ^
Pics?
Avid Rabid Hobbyist
 

Offline radar_macgyver

  • Frequent Contributor
  • **
  • Posts: 380
  • Country: us
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #41 on: May 02, 2016, 10:30:52 pm »
Could you elaborate more on this ^
Pics?

When I go to Utility->System Status, I can see the following:

Software version: 1.2.133.1
FPGA version: 16.1.29
Hardware version: 6-4
Serial numbers, etc.

When I go to Siglent's website, I download a file called SDS2000X-P33R1.rar. From these two bits of information, how can I tell that I have the latest firmware installed? Either Siglent should name the file something like SDS2000X-1.2.133.1-16.1.29.rar, or have an additional line in Utility->System Status that says Firmware Revision: P33R1.

Unless, of course, it's already listed and I missed it.

As for the update rate maximization, yes, I had read that thread. I don't think that what I'm seeing is desired behavior, though. It seems highly unlikely that a ~600 MHz Blackfin processor would take 40-3 = 33 milliseconds to compute the peak-to-peak voltage and time period for 14k samples, this seems like it was implemented incorrectly.
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 15977
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #42 on: May 03, 2016, 04:15:08 am »
Could you elaborate more on this ^
Pics?

When I go to Utility->System Status, I can see the following:

Software version: 1.2.133.1
FPGA version: 16.1.29
Hardware version: 6-4
Serial numbers, etc.

When I go to Siglent's website, I download a file called SDS2000X-P33R1.rar. From these two bits of information, how can I tell that I have the latest firmware installed? Either Siglent should name the file something like SDS2000X-1.2.133.1-16.1.29.rar, or have an additional line in Utility->System Status that says Firmware Revision: P33R1.

Unless, of course, it's already listed and I missed it.
You did sort of, see bold highlights ^^^

Quote
As for the update rate maximization, yes, I had read that thread. I don't think that what I'm seeing is desired behavior, though. It seems highly unlikely that a ~600 MHz Blackfin processor would take 40-3 = 33 milliseconds to compute the peak-to-peak voltage and time period for 14k samples, this seems like it was implemented incorrectly.
I do wonder if you have other features enabled that has slowed the update rates to a point where it is noticeable?  :-\
Try a "Default" to disable all settings not necessary and see if your noticed behaviour remains.

There's further recent discussion here:
https://www.eevblog.com/forum/testgear/siglent-sds2000-new-v2-firmware/msg923035/#msg923035

Anyways, Siglent tech support have been linked to your posts.
Avid Rabid Hobbyist
 

Offline radar_macgyver

  • Frequent Contributor
  • **
  • Posts: 380
  • Country: us
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #43 on: May 03, 2016, 05:01:05 am »
You did sort of, see bold highlights ^^^
Lol, not the most obvious, but yes, it is there once you know what to look for.

I do wonder if you have other features enabled that has slowed the update rates to a point where it is noticeable?  :-\
Try a "Default" to disable all settings not necessary and see if your noticed behaviour remains.

There's further recent discussion here:
https://www.eevblog.com/forum/testgear/siglent-sds2000-new-v2-firmware/msg923035/#msg923035
The discussion on that thread seems to focus on the sequence mode, which I'm not using. I repeated the test after pressing "Default", the only settings changed after that are the trigger level, vertical and horizontal scale, and the acquisition settings as described in my previous post.

Anyways, Siglent tech support have been linked to your posts.
Thank you much! Hopefully they'll figure out what's the cause.
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3069
  • Country: fi
  • Starting with DLL21
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #44 on: May 03, 2016, 08:49:36 am »
Here how it looks with SDS2000 without X.
Measurements and tracking cursors and if measurements are Gated or not,  do not affect in this test.
If shut off measurements result in trig out is exactly same or least I can not detect difference.

But, I can believe X model may have some bug there.
Just noted this also in some caases with SDS1kX with version P06


SDS2304  all meas and tracking curs on and measurement mode gated and statistics also on.



Here what it looks from SDS2304 Trigger Output using SDS1202X for monitoring.
SDS1202X-S    Signal from SDS2304 Trig Out (previous image situation)


Result: No any slow down nor does anything strange noticeable in acquisition cycle when use cursors and measurements things in SDS2304. Just rock solid.
« Last Edit: May 03, 2016, 09:22:10 am by rf-loop »
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 15977
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #45 on: May 03, 2016, 09:59:21 am »
Here how it looks with SDS2000 without X.
Measurements and tracking cursors and if measurements are Gated or not,  do not affect in this test.
If shut off measurements result in trig out is exactly same or least I can not detect difference.

But, I can believe X model may have some bug there.
Just noted this also in some caases with SDS1kX with version P06



Result: No any slow down nor does anything strange noticeable in acquisition cycle when use cursors and measurements things in SDS2304. Just rock solid.
Tech support confirms your X series suspicions and radar_macgyver's findings and it's been passed to R&D.
We'll keep you all informed.
Avid Rabid Hobbyist
 
The following users thanked this post: rf-loop

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3069
  • Country: fi
  • Starting with DLL21
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #46 on: May 03, 2016, 10:33:03 am »
Here how it looks with SDS2000 without X.
Measurements and tracking cursors and if measurements are Gated or not,  do not affect in this test.
If shut off measurements result in trig out is exactly same or least I can not detect difference.

But, I can believe X model may have some bug there.
Just noted this also in some caases with SDS1kX with version P06



Result: No any slow down nor does anything strange noticeable in acquisition cycle when use cursors and measurements things in SDS2304. Just rock solid.
Tech support confirms your X series suspicions and radar_macgyver's findings and it's been passed to R&D.
We'll keep you all informed.

TNX. I heve been just thinking that I will made further tests with SDS1kX and report to Siglent but, due to lack of time and other work with SDS1000X  Wavegen this issue have been waiting. Good to know issue is delivered  to Siglent. (I was thinking to give bit more detailed information to Siglent but lets hope theu find themselves enough - lets hope there is some who is enough calm and circumspect for do careful tests so that it is fully analyzed first all what conflict there and only after then take FW on the table for stop oops...oops...oops... rounds.
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 

Offline radar_macgyver

  • Frequent Contributor
  • **
  • Posts: 380
  • Country: us
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #47 on: May 04, 2016, 04:16:44 am »
Thanks, tautech! Please let me know if there's any additional info I can provide.
 

Offline radar_macgyver

  • Frequent Contributor
  • **
  • Posts: 380
  • Country: us
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #48 on: May 04, 2016, 05:21:54 am »
Another bug to report: the display persistence doesn't really behave properly. I expected that if the persistence is set for, say, 5 seconds, and a random pulse shows up on the display, it will stay on the display for 5 seconds before disappearing. Instead, the firmware effectively implements infinite persistence, and clears the persistence once every 5 seconds. This is not expected behavior and makes the timed persistence function mostly useless.

Additionally, when the persistence menu is set to 1s, it never clears the display (same as infinite).

Also, the way the persistence works is a bit odd. The historic data is shown at 50% intensity, and after the timeout it disappears immediately. IMO it is preferable to fade out the historic data gradually, like the Tek DPO series does. The same hardware in the display processor FPGA that handles intensity grading can accomplish this easily.
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3069
  • Country: fi
  • Starting with DLL21
Re: EEVblog #864 - Siglent SDS2000X Series Oscilloscope Teardown
« Reply #49 on: May 05, 2016, 08:05:57 am »
Another bug to report: the display persistence doesn't really behave properly. I expected that if the persistence is set for, say, 5 seconds, and a random pulse shows up on the display, it will stay on the display for 5 seconds before disappearing. Instead, the firmware effectively implements infinite persistence, and clears the persistence once every 5 seconds. This is not expected behavior and makes the timed persistence function mostly useless.

Additionally, when the persistence menu is set to 1s, it never clears the display (same as infinite).

Strange. But agen, I do not have SDS2kX, I have SDS2k and SDS1kX.
Just tested with SDS1000X and Persistence works. When scope running (using gaussian noise as signal) whole time new samples are hold on screen until time limit and older shutted off from image, continuously just like fifo works.



Quote
Also, the way the persistence works is a bit odd. The historic data is shown at 50% intensity, and after the timeout it disappears immediately. IMO it is preferable to fade out the historic data gradually, like the Tek DPO series does. The same hardware in the display processor FPGA that handles intensity grading can accomplish this easily.

What additional information this slow fade out give to user? Or do you mean it is cosmetically more "nice".
Fast acquisition and image daata processing is bit time critical process.  I do not accept if example persistence on slows wfm/s speed. If we need keep track for every acquisition  age  it is bit hard.  If there is example 100000 wfm/s there is roughly 4000 acquisitions in each TFT image what are renewed every 40ms with new set of 4000 acquisitions. This is still quite easy case but when we work with random undefined signals scope can not design just for one case, sontinuous repetitive signal. Before this fade out gradattion is good we need time stamp every single acquisition and after time start decay intensity individually for every acquisition. If we now want 5 second persistence and after then say example 5s fade out. Total time for every acquisition is 10s and in "worst" case there is now 1000000 acquisition in ageing queye.
Easy case if there is 100wfm/s acquisition speed. Bit harder if 1000 times more or even more.

If do not want this perfect system what may be useful in some cases if also user can adjust these keep and fadeoff parameters for his signal inder test, in some cases I can see it may give some kind of advantage over "only nice image".
More easy it is if we want only just cosmetically "nice image" it can do more easy. But I can not see how much it really give added value.  Oscilloscope, as caar and as camera is always compromise, add something and perhaps loose something like car tires, get more longevity - get more slipery in wet. 

Get more nice image and it (may) need more processing force what may need take from some other place just when there is critical need for all possible resources if do not add more brute force in HW.

Let us wish that the development is developing.
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf