Author Topic: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)  (Read 49968 times)

0 Members and 3 Guests are viewing this topic.

Online nctnico

  • Super Contributor
  • ***
  • Posts: 26755
  • Country: nl
    • NCT Developments
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #50 on: October 30, 2019, 01:25:38 am »
It only decodes what is in the screen, not the full sample memory
It doesn't work with "pixels" like the DS1054Z, it actually does a memory decode of the area shown on screen.
Which still sucks. Try to zoom in on parts of a long message to look at the waveform of the signal (which is the primary use of decoding on an oscilloscope). It doesn't work on oscilloscopes which only decode what is on screen. I don't understand why oscilloscope manufacturers still dare to put scopes on the market which work this way in 2019.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline IDEngineer

  • Super Contributor
  • ***
  • Posts: 1924
  • Country: us
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #51 on: October 30, 2019, 01:32:43 am »
I don't understand why oscilloscope manufacturers still dare to put scopes on the market which work this way in 2019.
Because firmware is cheap, and dedicated hardware is expensive.
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 26755
  • Country: nl
    • NCT Developments
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #52 on: October 30, 2019, 01:43:16 am »
I don't understand why oscilloscope manufacturers still dare to put scopes on the market which work this way in 2019.
Because firmware is cheap, and dedicated hardware is expensive.
You don't need dedicated hardware. Just an extra path inside the FPGA. The signal can be downsampled a lot before processing so decoding isn't a very demanding task either. MicSig can do it in their 500 euro tablet scopes...
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16561
  • Country: 00
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #53 on: October 30, 2019, 07:57:39 am »
Which still sucks. Try to zoom in on parts of a long message to look at the waveform of the signal (which is the primary use of decoding on an oscilloscope).

You don't need to capture a long message to see the waveform.

 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16561
  • Country: 00
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #54 on: October 30, 2019, 08:07:17 am »
You don't need dedicated hardware. Just an extra path inside the FPGA.

Says who?

You realize that what you're saying is like saying "my car has four wheels and an engine, why can't it go 200mph? There's no reason it won't except the people who made it are soooo stupid!"

Me? I'd say the people who programmed the DS1054Z are quite heroic for getting such a limited piece of hardware to do any decoding at all.

MicSig can do it in their 500 euro tablet scopes...

Feel free to buy one of those.  :-//

 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 26755
  • Country: nl
    • NCT Developments
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #55 on: October 30, 2019, 10:49:34 am »
Which still sucks. Try to zoom in on parts of a long message to look at the waveform of the signal (which is the primary use of decoding on an oscilloscope).
You don't need to capture a long message to see the waveform.
But you need to see the decoding to know where you are at inside the message.

You don't need dedicated hardware. Just an extra path inside the FPGA.
Says who?
Someone with FPGA programming and building/designing data acquisition systems experience.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline bd139

  • Super Contributor
  • ***
  • Posts: 23018
  • Country: gb
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #56 on: October 30, 2019, 11:17:36 am »
You don't need dedicated hardware. Just an extra path inside the FPGA.

Says who?

You realize that what you're saying is like saying "my car has four wheels and an engine, why can't it go 200mph? There's no reason it won't except the people who made it are soooo stupid!"

Me? I'd say the people who programmed the DS1054Z are quite heroic for getting such a limited piece of hardware to do any decoding at all.

I'm going to bitch and moan here.

No.

The DS1054Z has a 454MHz ARM9 is two orders of magnitude slower than a  HP 54600 with a 10MHz 68000 and 16k of RAM made 26 years ago...

They can do better.

Most of the problems with the DS1054Z are not in the acquisition side of things but obviously in the rest of it.
 

Online tv84

  • Super Contributor
  • ***
  • Posts: 3212
  • Country: pt
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #57 on: October 30, 2019, 12:01:06 pm »
a 454MHz ARM9 is two orders of magnitude slower than a  HP 54600 with a 10MHz 68000

In due fairness this is not correct.

You all know that we're talking about realtime equipments with limited resources. Of course we can fine tune indefinitely but there's a moment when we have to stop or when the proc scheduling can't accept any other task!

These are not academic projects. These are commercial products.

This theme can go on indefinitely...

We usually see discussions here about BW upgrading (all types of venues) and always discuss the theme from the analog part of the scope but, for sure, many times the problem is simply the processing power to crunch the same functionalities/options at a higher speed.

But, in the end, it's all about $$$
 

Offline bd139

  • Super Contributor
  • ***
  • Posts: 23018
  • Country: gb
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #58 on: October 30, 2019, 12:07:49 pm »
It's not real time. If it was real time, the UI would respond in a timely fashion. I know this from using VxWorks on Xilinx / PPC about two decades ago. Basic operations when it is not doing any heavy crunching (1 trace, low sample rate) is SLOOOOOOOOOOW as fuck and there is no noticeable difference in how slow as fuck it goes when it is crunching.

The scope is two things glued together:

1. Acquisition and preprocessing (mostly in the FPGA by the looks)
2. UI and postprocessing (mostly in the CPU side of things by the looks)

The latter is definitely not real time. It runs Linux, which is not real time. And it sucks.

It is all about $$$ in the end you're right. The second half of the scope is where they saved the money and it was in the software not the hardware.
« Last Edit: October 30, 2019, 12:11:25 pm by bd139 »
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 26755
  • Country: nl
    • NCT Developments
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #59 on: October 30, 2019, 12:20:37 pm »
It is all about $$$ in the end you're right. The second half of the scope is where they saved the money and it was in the software not the hardware.
I agree. I'm quite sure that the hardware of the DS1054Z is capable of delivering much better performance with better software. And it doesn't take going for some exotic realtime OS. Linux will do just fine. With better programming so tasks which take long can be interrupted the UI response will be much faster. With decoding inside the FPGA (this doesn't need to take a whole lot of logic as well) the end product could easely be 10 times better. But all this takes more time to program and better software engineers. Not Rigol's strong suit at all.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline bd139

  • Super Contributor
  • ***
  • Posts: 23018
  • Country: gb
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #60 on: October 30, 2019, 12:33:48 pm »
What's even more depressing is I played with an MSO5000 at a hamfest and it was just as bad. Seriously if you have that as an architectural problem and it's pretty obvious you need to go back and find out why and fix it. If that means you did something stupid and have to tear the thing apart and start again, then so be it.
 

Online tv84

  • Super Contributor
  • ***
  • Posts: 3212
  • Country: pt
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #61 on: October 30, 2019, 12:40:22 pm »
If the "whole system" wasnt realtime, it wouldn't work. And, you can have specific parts without realtime OSes as long as the "big picture" performs within the deadlines. That doesnt mean the system isn't realtime.

We both understand each other points.
 

Online tv84

  • Super Contributor
  • ***
  • Posts: 3212
  • Country: pt
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #62 on: October 30, 2019, 12:44:29 pm »
What's even more depressing is I played with an MSO5000 at a hamfest and it was just as bad.

Regarding this part: I've done quite a bit of reversing in this scope  ::) and, in this one, I totally agree with you . There is plenty of surplus code. That leads me to believe that the thing wasn't done with so much care as a DS1054Z.

That corollary that when you have plenty of HW power the SW guys can be sloppier...
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16561
  • Country: 00
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #63 on: October 30, 2019, 01:05:25 pm »
You don't need dedicated hardware. Just an extra path inside the FPGA.
Says who?
Someone with FPGA programming and building/designing data acquisition systems experience.

I'm not disputing the "path" part, it's the "just an extra" part that bothers me.

A serial decoder for multiple protocols won't be a small/trivial thing to implement in an FPGA. Maybe there simply isn't any space left in this one.
 

Offline bd139

  • Super Contributor
  • ***
  • Posts: 23018
  • Country: gb
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #64 on: October 30, 2019, 01:12:42 pm »
Yes agree there. It's mostly triggering concerns I suspect in the FPGA. Anything after that i.e. actually parsing the protocol should be a software decode function from the sample memory (not the bloody screen!)
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 26755
  • Country: nl
    • NCT Developments
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #65 on: October 30, 2019, 01:14:44 pm »
You don't need dedicated hardware. Just an extra path inside the FPGA.
Says who?
Someone with FPGA programming and building/designing data acquisition systems experience.

I'm not disputing the "path" part, it's the "just an extra" part that bothers me.

A serial decoder for multiple protocols won't be a small/trivial thing to implement in an FPGA. Maybe there simply isn't any space left in this one.
Actually it is very trivial. UART, I2C, etc peripherals are really simple blocks of logic. Protocol error detection is already standard on these blocks so the only thing needed is to timestamp what comes out and store that data. It doesn't need to be more fancy than that.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline bd139

  • Super Contributor
  • ***
  • Posts: 23018
  • Country: gb
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #66 on: October 30, 2019, 01:17:48 pm »
Actually it is very trivial. UART, I2C, etc peripherals are really simple blocks of logic. Protocol error detection is already standard on these blocks so the only thing needed is to timestamp what comes out and store that data. It doesn't need to be more fancy than that.

I disagree there. Those tend to only trigger/work when the protocol is not broken, which is one reason I had to break the scope out for a problem recently. The DS1054Z seems to trigger when the protocol is bust pretty well. There's a big difference between a functional trigger and a working protocol.
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 26755
  • Country: nl
    • NCT Developments
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #67 on: October 30, 2019, 01:44:56 pm »
Actually it is very trivial. UART, I2C, etc peripherals are really simple blocks of logic. Protocol error detection is already standard on these blocks so the only thing needed is to timestamp what comes out and store that data. It doesn't need to be more fancy than that.

I disagree there. Those tend to only trigger/work when the protocol is not broken, which is one reason I had to break the scope out for a problem recently. The DS1054Z seems to trigger when the protocol is bust pretty well. There's a big difference between a functional trigger and a working protocol.
Triggering on a protocol is entirely different compared to decoding. What you need for triggering on a protocol is pattern matching. This isn't difficult to implement in a statemachine or using partical FPGA reconfiguration to load the trigger engine.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16561
  • Country: 00
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #68 on: October 30, 2019, 01:54:00 pm »
Anything after that i.e. actually parsing the protocol should be a software decode function from the sample memory (not the bloody screen!)

I suspect the main DS1054Z CPU doesn't have direct access to sample memory, only to the downsampled version of the data you see displayed on screen.

 

Offline bd139

  • Super Contributor
  • ***
  • Posts: 23018
  • Country: gb
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #69 on: October 30, 2019, 02:11:41 pm »
I'm not so sure as you can pull the entire sample memory via network if I remember correctly. That'd have to go through the CPU.

Edit: http://int.rigol.com/File/TechDoc/20151218/MSO1000Z&DS1000Z_ProgrammingGuide_EN.pdf .. see :WAVEform: ....

« Last Edit: October 30, 2019, 02:30:39 pm by bd139 »
 

Offline rsjsouza

  • Super Contributor
  • ***
  • Posts: 5980
  • Country: us
  • Eternally curious
    • Vbe - vídeo blog eletrônico
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #70 on: October 30, 2019, 02:37:33 pm »
What's even more depressing is I played with an MSO5000 at a hamfest and it was just as bad.
If they are comparing themselves to Tektronix MDO3 or MDO4, then they should be just fine.  :popcorn:

Seriously if you have that as an architectural problem and it's pretty obvious you need to go back and find out why and fix it. If that means you did something stupid and have to tear the thing apart and start again, then so be it.
Siglent seems to have done this on their SDS2000 series - the software seems to have been quite overhauled at a certain point in time with their "V2" (reference 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...
 
The following users thanked this post: egonotto

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16561
  • Country: 00
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #71 on: October 30, 2019, 03:43:09 pm »
I'm not so sure as you can pull the entire sample memory via network if I remember correctly. That'd have to go through the CPU.

Sure, but it goes really sloooooow.

The FPGA obviously has a mechanism for copying chunks of sample memory to main memory (probably in 1200 byte blocks - twice the number of pixels on screen, it's a number that comes up a lot in the programming manual). It doesn't mean you can process the whole of sample memory quickly though.
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 37661
  • Country: au
    • EEVblog
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #72 on: October 31, 2019, 02:17:59 am »
I think more important question is, does Dave have it yet or on the way to him ?

Nope, just heard about this today, did see this thread.
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 26755
  • Country: nl
    • NCT Developments
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #73 on: October 31, 2019, 10:04:45 am »
I'm not so sure as you can pull the entire sample memory via network if I remember correctly. That'd have to go through the CPU.

Sure, but it goes really sloooooow.

The FPGA obviously has a mechanism for copying chunks of sample memory to main memory (probably in 1200 byte blocks - twice the number of pixels on screen, it's a number that comes up a lot in the programming manual). It doesn't mean you can process the whole of sample memory quickly though.
If data transfer is slow it is perfectly possible that data transfer from the acquisition memory (attached to the FPGA) to the CPU goes through SPI or I2C on the DS1054Z. On the Xilinx Zync platform based oscilloscopes the acquisition memory is connected to the main CPU directly.
« Last Edit: October 31, 2019, 10:06:37 am by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline ebastler

  • Super Contributor
  • ***
  • Posts: 6202
  • Country: de
Re: Rigol DS1202Z-E entry level scope (200MHz, 2 channel)
« Reply #74 on: November 07, 2019, 09:26:13 am »
Nothing to see here...
I agree. This is aimed at people who have "200Mhz, 2 channels" written on a shopping list and not much else.

Rigol's motivation for releasing this model seems pretty clear: They are consolidating their portfolio, replacing the DS2000E with a model based on the cheaper DS1000Z platform.

Makes good sense from their point of view, and I am sure Rigol checked that they had enough DS2000E customers to justify a replacement model. But I agree that the DS1202Z-E doesn't add anything new to their offering.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf