www.allpcb.com

Author Topic: Help with Oscilloscope purchase decision.  (Read 2857 times)

0 Members and 1 Guest are viewing this topic.

Online nctnico

  • Super Contributor
  • ***
  • Posts: 11513
  • Country: nl
    • NCT Developments
Re: Help with Oscilloscope purchase decision.
« Reply #50 on: April 21, 2017, 03:47:17 AM »
This secondary buffer is all the main CPU has direct access to, not the original sample data. This is the reason most calculations are done "on screen".

It's not impossible to process the whole of sample memory but it would either have to be done in batches on the main CPU or directly inside the FPGA/ASIC.
Utter nonsense!
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 5550
Re: Help with Oscilloscope purchase decision.
« Reply #51 on: April 21, 2017, 03:59:21 AM »
This secondary buffer is all the main CPU has direct access to, not the original sample data. This is the reason most calculations are done "on screen".

It's not impossible to process the whole of sample memory but it would either have to be done in batches on the main CPU or directly inside the FPGA/ASIC.
Utter nonsense!

It's true! I read it on the Internet.

eg. http://debugmo.de/2013/03/whats-inside-tektronix-dpo5034/



The "display memory" in that block diagram is what the CPU has access to, not the "memory".
« Last Edit: April 21, 2017, 04:01:21 AM by Fungus »
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 11513
  • Country: nl
    • NCT Developments
Re: Help with Oscilloscope purchase decision.
« Reply #52 on: April 21, 2017, 04:27:00 AM »
And you believe that is the only way to make a digital oscilloscope?  :palm:
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 5550
Re: Help with Oscilloscope purchase decision.
« Reply #53 on: April 21, 2017, 04:39:44 AM »
And you believe that is the only way to make a digital oscilloscope?  :palm:

What do you mean "only"? A few seconds ago it was "Utter nonsense!"  :popcorn:

Of course it's not the only way, but it's a very sensible way, especially on a budget device where you have to keep chip costs down.

I guess you can unify the two memories into one but that'll drive up the cost of the RAM chips and memory controllers.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 2414
  • Country: us
Re: Help with Oscilloscope purchase decision.
« Reply #54 on: April 21, 2017, 04:44:48 AM »
There are multiple different architectures that can and have been used in digital scopes, they all have advantages and disadvantages, some are geared toward a specific use case while others are more general purpose. Which is "better" is a subjective matter.
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 5550
Re: Help with Oscilloscope purchase decision.
« Reply #55 on: April 21, 2017, 04:49:27 AM »
There are multiple different architectures that can and have been used in digital scopes, they all have advantages and disadvantages, some are geared toward a specific use case while others are more general purpose. Which is "better" is a subjective matter.

I'm sure.

The only point I'm trying to make is that the decoding CPU doesn't necessarily have access to the sample data, that doing a full-memory serial decode on a device that isn't designed for it could be a programming nightmare.

 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 2414
  • Country: us
Re: Help with Oscilloscope purchase decision.
« Reply #56 on: April 21, 2017, 04:52:33 AM »
I would say that it's a feature that a sensible modern scope ought to be designed for though. If the capture memory is separate from the CPU RAM it should be possible to transfer the entire capture RAM into CPU RAM for additional processing. IMO if it's going to bother to do protocol decoding, it ought to do so on more than what is displayed on the screen at once.
 

Online Electro Fan

  • Super Contributor
  • ***
  • Posts: 1614
Re: Help with Oscilloscope purchase decision.
« Reply #57 on: April 21, 2017, 05:22:21 AM »
Unfortunately that doesn't always work. Sometimes it is a specific message which can cause a device to misbehave every now and then. At some point I had to deal with a problem where a message went wrong once every 30 minutes to an hour.
Was the problem in the analog domain?
Yes. A logic analyser would not have caught this problem:

What happens here is that an I2C device is stretching the clock to make the I2C master wait but the I2C master doesn't care.

Sorry if I missed it, but what model Agilent scope made the decoder view in the image you posted?  Thx
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 5550
Re: Help with Oscilloscope purchase decision.
« Reply #58 on: April 21, 2017, 05:56:19 AM »
IMO if it's going to bother to do protocol decoding, it ought to do so on more than what is displayed on the screen at once.

Of course!

All we need is to persuade them to do it.


« Last Edit: April 21, 2017, 05:58:24 AM by Fungus »
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 11513
  • Country: nl
    • NCT Developments
Re: Help with Oscilloscope purchase decision.
« Reply #59 on: April 21, 2017, 06:20:14 AM »
Unfortunately that doesn't always work. Sometimes it is a specific message which can cause a device to misbehave every now and then. At some point I had to deal with a problem where a message went wrong once every 30 minutes to an hour.
Was the problem in the analog domain?
Yes. A logic analyser would not have caught this problem:

What happens here is that an I2C device is stretching the clock to make the I2C master wait but the I2C master doesn't care.
Sorry if I missed it, but what model Agilent scope made the decoder view in the image you posted?  Thx
A DSO7104A (but it has moved on to a new owner).
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline GeorgeOfTheJungle

  • Frequent Contributor
  • **
  • Posts: 360
  • Country: es
Re: Help with Oscilloscope purchase decision.
« Reply #60 on: April 21, 2017, 06:40:46 AM »
A DSO7104A (but it has moved on to a new owner).

And what have you put in its place? If you don't mind me asking?
 

Offline kcbrown

  • Frequent Contributor
  • **
  • Posts: 334
  • Country: us
Re: Help with Oscilloscope purchase decision.
« Reply #61 on: April 21, 2017, 06:48:57 AM »
OTOH oscilloscopes aren't really designed internally for this sort of programming pattern and I can see why they don't do it.

It depends entirely on whether or not the acquisition memory is available at scope stop.   Frankly, it almost has to be, because you can zoom and pan on that same data

Nope. It has to be done in hardware because of the enormous amount of data.

The acquisition memory will be on one bus. Data is fed to it via DMA direct from the ADC.

Also connected to that RAM ss some sort of FPGA/ASIC which takes data from there and copies it to another buffer, scaled to fit the number of pixels visible on screen (depending on your timebase). It will also do the sin(x)/x interpolation, etc. There's no way this scaling/interpolation could be done in software on a cheap CPU, data can be coming in a 1GS/s even on low-end oscilloscopes.

This secondary buffer is all the main CPU has direct access to, not the original sample data. This is the reason most calculations are done "on screen".

It's not impossible to process the whole of sample memory but it would either have to be done in batches on the main CPU or directly inside the FPGA/ASIC.

I certainly don't disagree with this.  But see below.


Quote
No idea, but since we're talking about software decoding, there isn't anything in principal or in practice (that I'm aware of) that would prevent it.

Apart from (eg.) trying to access the 24MB of sample memory in 1200 byte chunks (on a DS1054Z).

That isn't something that prevents it.  And remember, the situation here is when the scope is stopped.  And especially if the implementation is done in the way I described, you'd be accessing chunks backwards starting with the one immediately prior to that of the display, and accessing only those chunks sufficient to arrive at the packet start location, then moving forwards from there, and saving the decoded values (along with their locations) so that you can easily display them should the user decide to pan/zoom.


Quote
In principal it could be done - they did something like it for the DS1054Z's improved FFT.

That's exactly my point.  There isn't anything that prevents them from implementing correct decoding.  And I'd argue that it's not necessarily uneconomical to do so either since it's just software -- it takes some additional time, but it only has to be done once.


Quote
In practice it doesn't seem like anybody is. Not for serial decoding.

I'm thinking it would be too slow.

Why in the world would it be too slow if you only do it when the scope is stopped??

It's a different matter altogether if the scope is running.  In that case, you can just do on-screen decoding and be done with it.   It might not be correct, but at that point there are real tradeoffs that you can't get away from.


Quote
Yes, the decode could be done in the FPGA/ASIC but that assumes there's enough gates left over and they bothered to do it.

Yeah, and it would also require additional memory that the FPGA/ASIC would have to write data and position information to.   I was thinking a strictly software-based approach.  Having hardware support opens up a lot of possibilities.
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 5550
Re: Help with Oscilloscope purchase decision.
« Reply #62 on: April 21, 2017, 06:58:17 AM »
it only has to be done once.

...assuming there's enough spare RAM to store all the decoded data along with its position on the timeline, etc.

How much serial data+position could 24Mb of samples actually decode to? What's the maximum?

Data is one byte, position would need 3 bytes. I'm thinking it could be a megabyte or more of data in total if you set the timebase to worst case. :popcorn:
 

Offline kcbrown

  • Frequent Contributor
  • **
  • Posts: 334
  • Country: us
Re: Help with Oscilloscope purchase decision.
« Reply #63 on: April 21, 2017, 08:20:55 AM »
it only has to be done once.

...assuming there's enough spare RAM to store all the decoded data along with its position on the timeline, etc.

Sorry, I meant that the programming only has to be done once.  It was a comment about the engineering resources needed to pull it off (and if you architect it properly, then you can make it easily adaptable to future architectures).


Quote
How much serial data+position could 24Mb of samples actually decode to? What's the maximum?

Data is one byte, position would need 3 bytes. I'm thinking it could be a megabyte or more of data in total if you set the timebase to worst case. :popcorn:

Well, the scope is stopped.  If the timebase is such that you wouldn't be able to display the decoded data, then you might not bother to do the decoding at all (until, of course, the user changes the timebase such that decoded data would be possible to show).  That tends to pretty severely limit the circumstances in which a decoding pass would be necessary.

If, on the other hand, you have a scrollable list that you're overlaying on the display, then you might have to deal with that problem.

For modern hardware, a megabyte of memory is nothing.   The memory for this doesn't have to be SRAM or anything special like that -- it's just general-purpose RAM for the CPU.  Modern ARM-based SBCs have gigabytes of RAM these days.   Even the DS1054Z might have enough in total for that (looks like the processor has a 32M RAM chip hanging off of it, the SK Hynix H5PS5162GFR: https://www.skhynix.com/product/filedata/fileDownload.do?seq=1909).


« Last Edit: April 21, 2017, 11:24:33 AM by kcbrown »
 

Offline snoopy

  • Frequent Contributor
  • **
  • Posts: 337
  • Country: au
    • Analog Precision
Re: Help with Oscilloscope purchase decision.
« Reply #64 on: April 21, 2017, 10:57:15 AM »
There are multiple different architectures that can and have been used in digital scopes, they all have advantages and disadvantages, some are geared toward a specific use case while others are more general purpose. Which is "better" is a subjective matter.

I'm sure.

The only point I'm trying to make is that the decoding CPU doesn't necessarily have access to the sample data, that doing a full-memory serial decode on a device that isn't designed for it could be a programming nightmare.

I believe it is changed in the Tek mdo3000 scopes where decodes and search occur across the whole sample memory and not just what's on the screen, that is why the Tek will take sometime to complete decodes after a new trigger ;)

cheers
 

Offline MrW0lf

  • Regular Contributor
  • *
  • Posts: 226
  • Country: ee
Re: Help with Oscilloscope purchase decision.
« Reply #65 on: April 21, 2017, 09:10:00 PM »
I believe it is changed in the Tek mdo3000 scopes where decodes and search occur across the whole sample memory and not just what's on the screen, that is why the Tek will take sometime to complete decodes after a new trigger ;)

Surely you can gate to screen, zoomed area or cursors? In case if want check surroundings only if spotting something interesting around (trigger point) can stop and change gate to full record.
Dunno why opinion that full-mem decoding scopes have only option to decode all (which is inevitably bit slow)  :-//
Another option - not decode at all on-the-fly, and then enable full record (or gated) decoding on stop.
Get luxurious event table over hundreds of MB with analog voltage stats for each packet etc...
IMHO this whole battle DSO vs DMO (data mining oscilloscope) is weird. Like comparing DSO and multimeter. Of course simple instrument is better/faster for simple task. But cannot do complex task at all...
MrW0lf=MrWolf. Personal messages please to MrW0lf only.
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 5550
Re: Help with Oscilloscope purchase decision.
« Reply #66 on: April 21, 2017, 11:13:54 PM »
IMHO this whole battle DSO vs DMO (data mining oscilloscope) is weird.

I get the feeling that if the 'scope manufacturers give people a full memory decode they'll just start saying things like, "Wahhh! It's such a pain to have to scroll through the data by twisting a knob in zoom mode, why can't we have a vertical list view? CSV dump? Automatic search for bad signals?"

It'll never end.  :popcorn:

 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 11513
  • Country: nl
    • NCT Developments
Re: Help with Oscilloscope purchase decision.
« Reply #67 on: April 21, 2017, 11:19:45 PM »
IMHO this whole battle DSO vs DMO (data mining oscilloscope) is weird.

I get the feeling that if the 'scope manufacturers give people a full memory decode they'll just start saying things like, "Wahhh! It's such a pain to have to scroll through the data by twisting a knob in zoom mode, why can't we have a vertical list view? CSV dump? Automatic search for bad signals?"
My GW Instek GDS-2204E can do all that so I'm not complaining and I'm sure it is not the only oscilloscope with such features.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline MrW0lf

  • Regular Contributor
  • *
  • Posts: 226
  • Country: ee
Re: Help with Oscilloscope purchase decision.
« Reply #68 on: April 21, 2017, 11:22:40 PM »
I get the feeling that if the 'scope manufacturers give people a full memory decode they'll just start saying things like,

If?! 128MB of mixed analog/digital with database-like interface. Including search/sort by packet voltages:



Seems I have to continue my Pico 2000 review sometime. Strangely limited perception of modern budget scope powers is...
MrW0lf=MrWolf. Personal messages please to MrW0lf only.
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 11513
  • Country: nl
    • NCT Developments
Re: Help with Oscilloscope purchase decision.
« Reply #69 on: April 21, 2017, 11:41:14 PM »
IMHO Pico should change their UI so it can work with a touch-screen and create an Android version so you can connect your scope to a tablet. I have used a Picoscope a couple of times many years ago and it didn't inspire me to buy one. Needing to mess around with the mouse to adjust something made using it way too tedious.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 5550
Re: Help with Oscilloscope purchase decision.
« Reply #70 on: April 21, 2017, 11:47:12 PM »
IMHO Pico should change their UI so it can work with a touch-screen and create an Android version so you can connect your scope to a tablet. I have used a Picoscope a couple of times many years ago and it didn't inspire me to buy one. Needing to mess around with the mouse to adjust something made using it way too tedious.

Yep, I get that their software is powerful but I just don't think having to use it via a PC for the basic stuff is good.

Maybe they could make another USB box with knobs and buttons on it.  :popcorn:

« Last Edit: April 21, 2017, 11:48:53 PM by Fungus »
 

Offline MrW0lf

  • Regular Contributor
  • *
  • Posts: 226
  • Country: ee
Re: Help with Oscilloscope purchase decision.
« Reply #71 on: April 22, 2017, 12:02:27 AM »
IMHO Pico should change their UI so it can work with a touch-screen and create an Android version so you can connect your scope to a tablet.

I agree that going touch is inevitable. They are trying to go in this direction with new "big buttons" in taskbar etc. However for me as oldschool PC user this whole touch business does not compute. Microscopic movement of wrist with "tuned" laser mouse vs waving hands around... what next - throwing stones at cyborgs trying to "communicate" with them?  :scared: Due to my line of work I'm chained to PC 8+ hours a day... cannot image how tiresome would be poke (and clean) screen all the time. Plus touch is extremely inaccurate. In "old days" playing Quake you got so good that could basically hit individual pixels with rail gun, on the run... So dunno, think its very personal / cultural thing and depends on persons background and task at hand. I'm not saying that DMO with classical PC interface is for everyone and every situation. Would not climb comms pole with it... Maybe only fix tiny USB scope at the top of pole, or maybe robot-DUT and enjoy the show from safe place over LAN/WiFi extender.
« Last Edit: April 22, 2017, 12:10:25 AM by MrW0lf »
MrW0lf=MrWolf. Personal messages please to MrW0lf only.
 
The following users thanked this post: james_s

Online rsjsouza

  • Super Contributor
  • ***
  • Posts: 1387
  • Country: us
    • Vbe - vídeo blog eletrônico
Re: Help with Oscilloscope purchase decision.
« Reply #72 on: April 22, 2017, 06:05:17 AM »
Very interesting discussion about the protocol decoding.

Out of the whole discussion I gather that, in practice, to have access to a well-round decoding feature the low entry bench oscilloscopes such as the DS1054Z or others are not the way to go. Only the more expensive models such as the GDS2204E and the DSO7104 have a complete decoding in memory. Not a surprise given the low-entry have more processing constraints.

Pico seems to have better options in this regard - no wonder, as the embedded hardware does not have to spare its time to process the User Interface. That and the fact they have "the world" to do whatever they can with the incoming data.

Just like having a capacitance meter in a regular DMM - limited when compared to a real LCR but good for the occasional quick check - the limited decoding has its place in low entry bench scopes. We only have to be aware of that.

Similarly to other facts of life, you must make a choice between cost and how well a feature is implemented... No surprises here.
Vbe - vídeo blog eletrônico http://videos.vbeletronico.com

Oh, the "whys" of the datasheets... The information is there not to be an axiomatic truth, but instead each speck of data must be slowly inhaled while carefully performing a deep search inside oneself to find the true metaphysical sense...
 

Offline kcbrown

  • Frequent Contributor
  • **
  • Posts: 334
  • Country: us
Re: Help with Oscilloscope purchase decision.
« Reply #73 on: April 22, 2017, 06:16:56 AM »
IMHO this whole battle DSO vs DMO (data mining oscilloscope) is weird.

I get the feeling that if the 'scope manufacturers give people a full memory decode they'll just start saying things like, "Wahhh! It's such a pain to have to scroll through the data by twisting a knob in zoom mode, why can't we have a vertical list view? CSV dump? Automatic search for bad signals?"
My GW Instek GDS-2204E can do all that so I'm not complaining and I'm sure it is not the only oscilloscope with such features.

It can automatically search for bad signals?   That's quite a nice capability if that's the case.

What's the WFM/s rate difference between when you have full memory decoding turned on and when you have it turned off?
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf