Author Topic: EEVblog #310 - Cheap Siglent and Agilent scopes  (Read 37532 times)

0 Members and 1 Guest are viewing this topic.

Offline darrenb

  • Contributor
  • Posts: 34
  • Country: au
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #25 on: July 14, 2012, 09:18:10 am »
Hey this is just China dragging themselves out of second world or third world status.  It won't go on forever.  They are not overtaking anyone, they are just catching up.

Many years ago everything used to be labelled "Made in Japan" or "Made in Taiwan", now it's all "Made in China" as China is the current cheap goods manufacturer of the world.  As the prices and wages go up in China people will realise that there are other even cheaper places to get stuff made (I hope it will be somewhere in Africa).

When I was last in China five years ago you could hire someone for US2 dollars per day.  That was and is their advantage but it won't last forever.  Wages in China are already going up significantly.  Soon the cheap goods will get made somewhere else and China will emerge like another Taiwan or Japan, not overtaking or dominating but simply catching up.

Sorry for the rant, off topic and all.



 

Offline hans

  • Super Contributor
  • ***
  • Posts: 1038
  • Country: nl
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #26 on: July 14, 2012, 10:07:43 am »
@DaveXRT: I must be mis informed then. 450 euro's for a 100MHz scope with 10MPts memory sounds very interesting.

I believe the only 2 major flaws in the scope was firmware bugs (but every scope has that I suppose) and some instability (rubber feet not grippy enough). I'm considering it , though, as it seems a nice basic entry level.

I can't find the video you mentoined, do you got a link?

1)
Big picture
2)
http://www.mediafire.com/?62z6s1ap9md3am8

Nice table, but not exactly what I meant to be asking. What I meant was how many waveforms/second it can do. I clearly see the Hantek is rubbish at that, because a 100kHz signal should be sufficient for 100k updates/second. The Owon is set to 2-ch and 1Mpts memory depth, which makes it a lot slower.

I looked at some rigols; DS1000E is unknown, DS1000B is 400/s, DS1000CA is 2000/s, DS2000 series is up to 50k/s. I got the user manual, and it says the display needs to be at 20ns (so that's like 2ns/div - lowest setting), auto record length (probably lowest, which is 14k), 1-channel and at dots display. So, basically it's quite a wank figure to look at, but the update speed shown by the Hantek is just horrible (i.e.: I worked with USB oscilloscopes that were faster than that).

Ah well, I guess you can't complain for 450 euro's. I just looked at Altium's video of their latest cloud services. A few shots around their new China office. What do you know: Owon scopes everywhere! Oh, and Aoyue soldering equipment too. Budget stuff :)
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 29658
  • Country: au
    • EEVblog
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #27 on: July 14, 2012, 12:02:23 pm »
Ah well, I guess you can't complain for 450 euro's. I just looked at Altium's video of their latest cloud services. A few shots around their new China office. What do you know: Owon scopes everywhere! Oh, and Aoyue soldering equipment too. Budget stuff :)

Watching the budget is very un-Altium like!  ;D

Dave.
 

Online Psi

  • Super Contributor
  • ***
  • Posts: 7280
  • Country: nz
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #28 on: July 14, 2012, 12:08:47 pm »
Ah well, I guess you can't complain for 450 euro's. I just looked at Altium's video of their latest cloud services. A few shots around their new China office. What do you know: Owon scopes everywhere! Oh, and Aoyue soldering equipment too. Budget stuff :)

i wouldn't call Aoyue budget, it's quite popular in china manufacturing.
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline T4P

  • Super Contributor
  • ***
  • Posts: 3706
  • Country: sg
    • T4P
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #29 on: July 14, 2012, 12:21:29 pm »
Ah well, I guess you can't complain for 450 euro's. I just looked at Altium's video of their latest cloud services. A few shots around their new China office. What do you know: Owon scopes everywhere! Oh, and Aoyue soldering equipment too. Budget stuff :)

i wouldn't call Aoyue budget, it's quite popular in china manufacturing.

It IS budget in our eyes, they can easily setup a few benches with the prices they are charging for their cad stuff
 

Offline T4P

  • Super Contributor
  • ***
  • Posts: 3706
  • Country: sg
    • T4P
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #30 on: July 14, 2012, 12:24:52 pm »
@DaveXRT: I must be mis informed then. 450 euro's for a 100MHz scope with 10MPts memory sounds very interesting.

I believe the only 2 major flaws in the scope was firmware bugs (but every scope has that I suppose) and some instability (rubber feet not grippy enough). I'm considering it , though, as it seems a nice basic entry level.

I can't find the video you mentoined, do you got a link?

1)
Big picture
2)
http://www.mediafire.com/?62z6s1ap9md3am8

Nice table, but not exactly what I meant to be asking. What I meant was how many waveforms/second it can do. I clearly see the Hantek is rubbish at that, because a 100kHz signal should be sufficient for 100k updates/second. The Owon is set to 2-ch and 1Mpts memory depth, which makes it a lot slower.

I looked at some rigols; DS1000E is unknown, DS1000B is 400/s, DS1000CA is 2000/s, DS2000 series is up to 50k/s. I got the user manual, and it says the display needs to be at 20ns (so that's like 2ns/div - lowest setting), auto record length (probably lowest, which is 14k), 1-channel and at dots display. So, basically it's quite a wank figure to look at, but the update speed shown by the Hantek is just horrible (i.e.: I worked with USB oscilloscopes that were faster than that).

Ah well, I guess you can't complain for 450 euro's. I just looked at Altium's video of their latest cloud services. A few shots around their new China office. What do you know: Owon scopes everywhere! Oh, and Aoyue soldering equipment too. Budget stuff :)

I forgot i posted owon only

You won't be able to know the exact updat speed, but sample rates is the foremost important thing when choosing owon vs rigol
You get more at 1M sample length and that it doesn't use overclocked ADC's
 

Offline hans

  • Super Contributor
  • ***
  • Posts: 1038
  • Country: nl
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #31 on: July 14, 2012, 01:51:09 pm »
Seeing that chart, the Owon is indeed far superior.
Well I guess sampling speed is quite important, so you can make use of the high memory at high sample rates. I would just favour a scope that retriggers fast.

Ah well, I guess you can't complain for 450 euro's. I just looked at Altium's video of their latest cloud services. A few shots around their new China office. What do you know: Owon scopes everywhere! Oh, and Aoyue soldering equipment too. Budget stuff :)

Watching the budget is very un-Altium like!  ;D

Dave.

Oh yeah, but I consider Altium still good value for money, if you're a (big) electronics development company :)
Honestly, I was quite surprised. Did Altium have a special 'measurement lab' in Australia, back when you worked there?
Ah well, I guess you can't complain for 450 euro's. I just looked at Altium's video of their latest cloud services. A few shots around their new China office. What do you know: Owon scopes everywhere! Oh, and Aoyue soldering equipment too. Budget stuff :)

i wouldn't call Aoyue budget, it's quite popular in china manufacturing.

Still way cheaper than what I paid for a Weller WSD80 station. Second hand for 130 euro's (4 year old unit). It works fine, but it's exterior is a bit damaged and I got a 150W iron stand with it (so a 80W iron slides and falls off). New price was 300 - 400 euro or there about.
 You can get an Aoyue 70W iron for like 60-70 euro's, and hot air for 80. Rework: 150 euro's.

That's still way way cheaper than Weller. I tried a Weller WR3000M rework station at college once and works very well, although I only removed a couple of SMT components.. Price at Farnell: just 2200 euro's excl VAT. I have worked with Aoyue hot air stations, they do the job but are less luxurious to use (easier to use, less buttons to push).

That still makes Aoyue budget stuff, despite it works :)
« Last Edit: July 14, 2012, 01:53:43 pm by hans »
 

Offline tinhead

  • Super Contributor
  • ***
  • Posts: 1903
  • Country: 00
    • If you like my hacks, send me a donation
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #32 on: July 14, 2012, 02:21:30 pm »

I forgot i posted owon only


when you have no idea what you doing - doin't do it ^^

You posted overview of Tekway/Hantek/Voltcraft memory vs. timebase vs. sampling rate - however from a hardware revision (hw1005)
which was produced for 2 months a yeah ago and has been replaced by hw1007.
Since Nov 2011 the sampling rate on hw1007 has been improved to meet the 1GSs/500MSs/500MSs/250MSs specs - they now like on Rigol E
and not anymore 1GSs/500MSs/400MSs/200MSs.

I don't want to be human! I want to see gamma rays, I want to hear X-rays, and I want to smell dark matter ...
I want to reach out with something other than these prehensile paws and feel the solar wind of a supernova flowing over me.
 

Offline T4P

  • Super Contributor
  • ***
  • Posts: 3706
  • Country: sg
    • T4P
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #33 on: July 14, 2012, 02:26:28 pm »

I forgot i posted owon only


when you have no idea what you doing - doin't do it ^^

You posted overview of Tekway/Hantek/Voltcraft memory vs. timebase vs. sampling rate - however from a hardware revision (hw1005)
which was produced for 2 months a yeah ago and has been replaced by hw1007.
Since Nov 2011 the sampling rate on hw1007 has been improved to meet the 1GSs/500MSs/500MSs/250MSs specs - they now like on Rigol E
and not anymore 1GSs/500MSs/400MSs/200MSs.



No idea? Hahaha... still doesn't change the fact that owon speeds are stock and that old hantek's speed is before a upgrade



--
Good for you, my 858D crapped itself again  :)
Though luckily i only paid 20USD
 

Offline tinhead

  • Super Contributor
  • ***
  • Posts: 1903
  • Country: 00
    • If you like my hacks, send me a donation
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #34 on: July 14, 2012, 02:29:01 pm »
@DaveXRT: I must be mis informed then. 450 euro's for a 100MHz scope with 10MPts memory sounds very interesting.

I believe the only 2 major flaws in the scope was firmware bugs (but every scope has that I suppose) and some instability (rubber feet not grippy enough). I'm considering it , though, as it seems a nice basic entry level.

I can't find the video you mentoined, do you got a link?

10Mpoint is not everything, Owon SDS waveform update rate is slow as hell (max 30wfms/s). Even slow Rigol E is doing 800wfms/s.

There is new decent UNI-T, i'm honestly not a big fan of that company, however their new model UTD2102CM looks nice (on paper)
- up to 16Mpoint and up to 150000wfms/s - in 6kpoint mode:

https://www.eevblog.com/forum/product-reviews-photos-and-discussion/has-anyone-seen-the-utd2102cm/

I hope i get one for teardown/some measurments soon.
I don't want to be human! I want to see gamma rays, I want to hear X-rays, and I want to smell dark matter ...
I want to reach out with something other than these prehensile paws and feel the solar wind of a supernova flowing over me.
 

Offline tinhead

  • Super Contributor
  • ***
  • Posts: 1903
  • Country: 00
    • If you like my hacks, send me a donation
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #35 on: July 14, 2012, 03:36:54 pm »
No idea? Hahaha... still doesn't change the fact that owon speeds ..

yes no idea.

Of course 10Mpoint is "better" than 1Mpoint, but honestly you can capture 10 times on 1Mpoint scope and you will get the
complete length (when you really need this, which i doubt) or be bit more inteligent and set trigger to capture on what
you (assume to) looking for.

Owon does not have any markers, no search, primitive zoom, primitive trigger, extremely low wfms/s rate
and only some basic functions in the firmware - but hey VGA out, 800x600 screen, 10Mpoint and battery powered.

Even the biggest SDS model - SDS9302 is not better (actually even worse - 10Mpoints only with up to 800MSa/s;
3.2GSa/s only with 10kpoints) and this is already 1500USD device.

Knowing what hardware they used, the resulting product it's a shame.

... it doesn't use overclocked ADC's

and?

did you ever measured the influence of overclocked ADC (AD9288-40 running as -100 and/or AD9288 clocked with 125MHz)
on the signal integrity? I did, and to be very honest there is nothing bad about (unless you screw up power supply), there is
even still room for more (up to 159MHz). It seems the person who recognized it at first (i assume someone from Instek or Rigol)
did great job!

The ADC used in Owon SDS is a cheap semi-clone of ADC08D500 - command compatible but not pinout. When you look what RuiFeng (the manufacturer) is producing you will find out that most datasheets are just 1:1 copy from the original AD parts.

So what do you think, what is better - known but overclocked ADC where you can compare data or unknown product
from unknown manufacturer having no real product datasheet?
Even now, based on RuiFeng datasheet from MXT2002 the ENOB, THD, SNR, SINAD are better on AD9288 (even overclocked)
as on RuiFengs MXT2002. Honestly, i don't want to know the real data, but it seems that even such "crap" company as UNI-T
recognized few months ago that RuiFeng ADCs are too bad, they switched to Analog Deviced ADCs - and that's someting
like 20USD difference per DSO which is in this device class A LOT of money, you will not pay this without a reason.

I don't want to be human! I want to see gamma rays, I want to hear X-rays, and I want to smell dark matter ...
I want to reach out with something other than these prehensile paws and feel the solar wind of a supernova flowing over me.
 

Offline hans

  • Super Contributor
  • ***
  • Posts: 1038
  • Country: nl
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #36 on: July 14, 2012, 05:13:45 pm »
I don't entirely agree with the advantage of the memory length. Yes , good triggering is essential. It' s not only about the acquire speed and rate, if you can't trigger a signal you have nothing. But with 10 times bigger memory you can afford to sample 10x faster on the same piece of time. However, that's only valid when you are working on large timer scales and can view it. If the interface design is poor and primitive, it's next to useless.

The long memory depth is also quite nice to capture a long chunk of data with high speed. So high memory is only useful if you can use it at the claimed maximum speed. Like... filling a 32MBit memory at 100ksps is something a PIC32 or ARM chip can do..

And for Owon's top SDS model.. with 1500USD you can get a lot more scope. There are some interesting models from Rigol (DS2000), Hameg and Agilent. I think Tektronix is still lacking behind, and Lecroy doesn't seem to be delivering much good neither (firmware)

I guess the thing is with test&measurement; you can always spend more. First you look at a scope of 400 euro's, and then see another one that's faster and only costs 80 euro more. Quickly you drift into scopes of 1000+ euro's because they contain more and more features, and only spending 50 - 200 euro's more sounds like a bargain. Each step you think hope to get more scope for your money.

I don't know what to say about the Uni-T. Seeing how much ASIC Agilent put into their 2000  & 3000 series DSO, I am not sure. Well, probably quite a lot of it is down to triggering and advanced math features..  I'd be very interested in how much of it is true. But, 600$ (I saw it priced at 700 euro's around here) is quite a chunk of money already.
« Last Edit: July 14, 2012, 05:15:54 pm by hans »
 

Offline T4P

  • Super Contributor
  • ***
  • Posts: 3706
  • Country: sg
    • T4P
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #37 on: July 14, 2012, 05:44:42 pm »
You people don't agree with me, but people have agreed with me on this forum about this OC/NO-OC ADC thing.
https://www.eevblog.com/forum/general-chat/buying-low-cost-dso-in-erope/msg107600/#msg107600

@DaveXRT: I must be mis informed then. 450 euro's for a 100MHz scope with 10MPts memory sounds very interesting.

I believe the only 2 major flaws in the scope was firmware bugs (but every scope has that I suppose) and some instability (rubber feet not grippy enough). I'm considering it , though, as it seems a nice basic entry level.

I can't find the video you mentoined, do you got a link?

10Mpoint is not everything, Owon SDS waveform update rate is slow as hell (max 30wfms/s). Even slow Rigol E is doing 800wfms/s.

There is new decent UNI-T, i'm honestly not a big fan of that company, however their new model UTD2q02CM looks nice (on paper)
- up to 16Mpoint and up to 150000wfms/s - in 6kpoint mode:

https://www.eevblog.com/forum/product-reviews-photos-and-discussion/has-anyone-seen-the-utd2102cm/

I hope i get one for teardown/some measurments soon.
UNi-T's scope which i talked about before is not actual but rather it's fake so please notice i was the one who started the topic  :)
Even though you replied in it before

« Last Edit: July 14, 2012, 05:48:48 pm by DaveXRT »
 

Offline tinhead

  • Super Contributor
  • ***
  • Posts: 1903
  • Country: 00
    • If you like my hacks, send me a donation
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #38 on: July 14, 2012, 06:40:41 pm »
UNi-T's scope which i talked about before is not actual but rather it's fake so please notice i was the one who started the topic  :)

no, this is real new UNI-T model, some dealers got them already, e.g.:

http://www.pinsonne-elektronik.de/pi1/pd125.html


I don't want to be human! I want to see gamma rays, I want to hear X-rays, and I want to smell dark matter ...
I want to reach out with something other than these prehensile paws and feel the solar wind of a supernova flowing over me.
 

Offline T4P

  • Super Contributor
  • ***
  • Posts: 3706
  • Country: sg
    • T4P
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #39 on: July 14, 2012, 06:58:17 pm »
ANDDDDDDDDD would you believe 15kwfrms from someone like UNi-T
Nope. Not going to. Not with their bench equipment
 

Offline tinhead

  • Super Contributor
  • ***
  • Posts: 1903
  • Country: 00
    • If you like my hacks, send me a donation
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #40 on: July 14, 2012, 07:55:21 pm »
ANDDDDDDDDD would you believe

wrong question :)

15kwfrms from someone like UNi-T
Nope. Not going to. Not with their bench equipment

I will give you an example :

A company "One Hung Low" think about producing DSOs. The market is huuge (1 billion USD in China!)
but there are lot of competitors producing good DSOs:
- Hantek/Tekway with 2500wfms/s
- Rigol with best seller DS1000E
- Owon with "huge" 10Mpoint
- ATTEN with low price
and so on.

Imagine "One Hung Low" have no real idea about DSOs but can "google" and have a simple math calculator :)
Now what they need is:
- cheap product (so let's use single RuiFeng ADC like Owon or multiple cheap from AD like others)
- cheap product (so let's use DPF displays - as 800x480 costs the same as 480x234 let's use these with better resolution)
- cheap product (so let's re-use other ppl design - frontend, DSP, what so ever necessary)
- something new (so let's replace the 1-2M SRAM with 16M DDR)
- something new (so let's design a mixed firmware - bit from here, bit from there ...)
- something new (so let's design really fast hardware)

16Mpoint DDR instead of 1-2Mpoint SRAM give them some money saving, 800x480 display give them at least same resolution
as on Hantek/Tekway (netto on Owon is not really better), everythign else can be copy/pasted from other product.

What still missing is wfms/s. This is probably the hardest thing ... but "One Hung Low" have no idea what wfms/s means.
But they can google - the first good result is R&S pdf http://www2.rohde-schwarz.com/file/1ER02_1e.pdf

So what we can see there? A lot of blah blah blah, complex things, but wait .. there is something :

There are various ways to evaluate the actual waveform acquisition rate ...
A... possibility is to monitor the trigger out of the oscilloscope. Every rising edge represents a new acquisition.


Bingo! That's it. So "One Hung Low" designed FPGA design having 6k dual port RAM, having trigger out from
FPGA (before any postprocessing - remember we have dual port RAM), and running at 250MHz internally.

Now we need cheapo calculator ... 150k wfms/s thats "trigger out event" or "6k of sample data stored in dual port RAM" every 6us.
"One Hung Low" have no idea what their blind time is ... why? well, they have no postprocessing in FPGA, so no blind time.
But they know that in R&S pdf "e.g." blind time is about 1000 the acquisition time.

"One Hung Low" have skilled engineers, so they can for sure do it bit faster, let say 500x acq time. So we have now 306us total,
that's 3.3kHz x 80bit data (5x AD9288) x 6000 point / 8bit bus = 200MHz ... so we clock FPGA with 100MHz,
sample buffer on each edge in ... output / 32bit, so we need only ARM with 50MHz recieving the data.
Simple eh? No doubt, cheapo calculator make it possible.

So now "One Hung Low" designed finally a 16Mpoint DSO, having 150000 wfms/s (who cares about post-processing, R&S said "trigger out events" so let's measure you western idiot our DSO!!!), a big 800x480 display and funny UI

Do we have same situation with UNI-T ? I don't know, i don't even try o think about it too much. In principle "One Hung Low" idea
from my example was ok, "count of trigger out event" can be defined as wfms/s. So if UNI-T decided to use this way then they for
sure managed to get such high value, on the other side nobody ever checked (yet) what inside ... so i would say
let's wait until we know more about hardware and about where the trigger out is connected too before we judge about UNI-T.
« Last Edit: July 14, 2012, 08:14:44 pm by tinhead »
I don't want to be human! I want to see gamma rays, I want to hear X-rays, and I want to smell dark matter ...
I want to reach out with something other than these prehensile paws and feel the solar wind of a supernova flowing over me.
 

Offline hans

  • Super Contributor
  • ***
  • Posts: 1038
  • Country: nl
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #41 on: July 14, 2012, 08:52:04 pm »
I looked into it a bit, and I must say; the long memory depth acknowledges the waveform/s.

Putting in large amounts of memory is ofcourse at an expense. The Uni-T has 16Mpts. Interesting.. is it just to beat Owon (they put 10M in so it's more than Rigol's and Hantek's), or?

First make it clear; waveforms/second is fully pronounced as the display of waveforms/second. They can acquire the data and throw it away. But that does every scope I think. How can you otherwise get 50% pre and post trigger data? You can't predict a trigger, so it must continuously be sampling data. So basically, to count as a wf/s, it needs to be on the screen.

To display all waveforms on the screen, you either need a 50kHz, 150kHz or 1MHz refresh rate screen.. don't think they exist. And otherwise I'm sorry mr. scope
Other way: you need to 'remember' all the waveforms and process them onto 1 screen at once.

If you read along in the same application note, chapter 4, they talk about the memory technology involved.
If you refresh the screen at 50Hz, you need to store 3000 waveforms (150k/50) into memory. With 6000 samples, thats 18Mpts of memory required.
Obviously 50Hz is not right, if you take 60Hz, you need 15Mpts. So that sounds doable, because the scope must contain 16Mpts to get to it's specs anyway.
At one time you'd need to downsample all that data to 1 screen. R&S say that's where up to 90% of the scope is busy with (at 10ns/div). So apparently the Uni-T is quick in doing that.
Obviously, you can't use on-chip memory of a FPGA for this (maybe only to continuously sample and once triggered & sampled, sent it to the 'main' RAM banks).

If you look at the Rigol DS2000; they can acquire 50k waveforms/second. The Rigol does this with a record length of 14kpts. If they refresh the screen at 50Hz, you need to store 1000 waveforms. 1000*14kpts = 14MPts. What a coincidence, the maximum memory depth of the scope is 14MPts.
There is an upgrade to 56Mpts, but the scope can't process the data fast enough to make use of that. Otherwise you can use 56MPts to  get 200k waveform/s (in theory).
If you take the DS4000: it can do 110k waveform/s and has 140Mpts of memory.
Once again, it can do 14kpts minimum. So you can fit 10000 waveforms in memory, which means you only have to refresh the screen 11x a second. I think the processing speed is limiting there; I would have thought you must be able to get 500k wf/s on that memory length.

I have no idea what Agilent is doing with their 2000X and 3000X series. The 2000X series has 100kpts of memory. What? How can you 'store' 1k waveforms for 50k wf/s? At a length of 1k, times 1k, it's still 1Mpts.
The 3000X has 2Mpts. It does 1M wf/s, so to make that work with a memory depth of 1k you need 1M*1k/50 = 20Mpts. Obviously I guessed the record length of only 1k here.
 I can't find the 2000X or 3000X series minimum memory depth.. These calculations suggests it's only 100 samples per trigger. That's very very small, but probably sufficient to fill 1 screen wide. It's not single shot so you don't need the zooming capability.
100 samples * 1Mwf/s = 100MS/s. The scope can do 1Gs/s+, so that's fine.
So either Agilent must be doing something else (fast screen? they drive it in an unique manner) or just polished up some figures.. (yeah we can capture 1 million things a second, you can't see anything on it, but whatever)

R&S RTO series: 20Mpts maximum, 1M wf/s. That's  what was required to make the 3000X series work with 1ks per trigger. So that sounds right.

First I was sceptic, but the R&S would actually confirm the Uni-T performance a bit.
Still though, if it's too good be true, it probably is.
They could be lying here and say they can pulse the trigger output 150k wf/s and can't figure out/too expensive how to make an ASIC/FPGA deal with the intense memory operations (R&S talked about 20 memory lanes each running at 500Mbps to store 8x10Gbps of ADC data). I would be very impressed if it did indeed work out. Keep us updated if you can get your grabs on a Uni-T + a screwdriver :)
« Last Edit: July 14, 2012, 09:42:43 pm by hans »
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 29658
  • Country: au
    • EEVblog
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #42 on: July 14, 2012, 09:54:10 pm »
I have no idea what Agilent is doing with their 2000X and 3000X series. The 2000X series has 100kpts of memory. What? How can you 'store' 1k waveforms for 50k wf/s? At a length of 1k, times 1k, it's still 1Mpts.
The 3000X has 2Mpts. It does 1M wf/s, so to make that work with a memory depth of 1k you need 1M*1k/50 = 20Mpts. Obviously I guessed the record length of only 1k here.
 I can't find the 2000X or 3000X series minimum memory depth.. These calculations suggests it's only 100 samples per trigger. That's very very small, but probably sufficient to fill 1 screen wide. It's not single shot so you don't need the zooming capability.
100 samples * 1Mwf/s = 100MS/s. The scope can do 1Gs/s+, so that's fine.
So either Agilent must be doing something else (fast screen? they drive it in an unique manner) or just polished up some figures.. (yeah we can capture 1 million things a second, you can't see anything on it, but whatever)

Both the 2000 and 3000 Agilent use the same ASIC with the same internal memory (4Msamples). The 2000 is software limited to 100K.
Both transfer data directly from the ASIC to the display memory, that's how they get the massive update rate.
You can see this on my video where the 3000 failed, and the corrupted waveform data inside the box.
It is also why the Agilent cannot get rid of the menu to give you more display area. The GUI processor and waveform display processor are separate.



Dave.
 

Offline tinhead

  • Super Contributor
  • ***
  • Posts: 1903
  • Country: 00
    • If you like my hacks, send me a donation
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #43 on: July 14, 2012, 10:41:53 pm »
hans,

don't mix things, wfms/s have nothing to do with display refresh rate.

If you build a DSO in a traditional way, there will be no trigger event before the last sampled data has been not
displayed (this was standard years ago, Owon used it on SDS models too - no idea why, probably because they lazy)
The wfrms/s is then typical maximal the screen refresh rate.

When you use threading you can already improve the wfrm/s, this is because each process runs independant
- acquisition
- UI

The data will be then simply skipped/accumulated/what so ever you wish - and displayed.
The resulting wfms/s rate will be higher, especially the peak value, but you will not get over 500 wfms/s.

If this is still not enough you can use something like "digital phosphor".
The sampled data - the waveforms (for example 5000 per second) will be stored into fast DPO buffer,
accumulated there to build so called "digital phosphor".

Every chinese DSO manufacturer know how store 1k-24k of data sampled at 1-2GSs, all they need
to speed-up their DSOs is to design something like the "digital phosphor" technology.

The problem is always the data throughput, with standard parts (low cost FPGAs - high end FPGA are to expensive)
you will get such example values:

- Rigol CA - 2000wfms/s with 10k buffer
- Tekway/Hantek - 2500 wfms with 4k buffer
- Hameg HMO3xxx - 2500 wfms/s with 24k*

If you need more you have to chose ASICs. Tektronix, with their (older) DPO2xxx and DPO3xxx series
can only do (even with ASICs) high peak values, DPO2xxx for example 55k wfms/s, but like Agilent
said "while in 10ns/DIV with 10k buffer only 2600wfms/s"

With better ASICs (with lot of fast integrated memory), like Agilent DSOX2/3 is using, much higher data throughput is possible allowing more wfms/s with deeper buffer.

And UNI-T ? No idea, for sure no ASIC inside, but that's all we can assume from the price.

* - i assume Hameg is switching to 24samples depth when you chose "optimal update rate", unfortunately there
is no single information about available to publis (maybe someone can test it?)
« Last Edit: July 14, 2012, 10:44:00 pm by tinhead »
I don't want to be human! I want to see gamma rays, I want to hear X-rays, and I want to smell dark matter ...
I want to reach out with something other than these prehensile paws and feel the solar wind of a supernova flowing over me.
 

Offline hans

  • Super Contributor
  • ***
  • Posts: 1038
  • Country: nl
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #44 on: July 14, 2012, 11:29:58 pm »
You're talking about CPU driven screens. Yes, you can run a thread in the background of the main proccesor drawing graphs together and dumping it on the screen. The refresh rate is irrelevant then; you can do 10 waveforms @ 50fps, or 100 waveforms 5fps.

The thing is, you need to store each waveform. And with figures of 50k/s the required memory is just so high. The data throughput is not possible on a CPU. 1Mwf/s * 1kpts = 1Gpts of data.
Neither can you refresh the screen at 50000Hz, because it's a screen with a maximum pixel clock etc. So you need to buffer all your waveform data in a memory and combine it ('digital phosphor').

So if you update the screen from hardware, it can be done much faster.
So okay, you want 1Mwf/s; fine. Only update it at 10Hz? Then you need the memory to store 1/10 of 1M waveforms; and if you choose those to be 1k long, that's 100Mpts of sample memory right there.
Of course memory is expensive, so updating the screen more frequent means smaller buffers.

It's a very big coincidence that digital phosphor scopes have very high memory depths. The buffer calculations I did, is basically segmented memory.
'digital phosphor' (seems like a tektronix trademark) is nothing more than a large segmented memory buffer and graphics engine to process it. Competitors have it as well.Application note from Agilent, page 2&3.
R&S say:
"The final bottleneck for high acquisition rates within a digital oscilloscope is the graphical display of the waveforms. The RTO ASIC, therefore, also includes a dedicated graphics engine that prepares the pixel representation of the accumulated waveforms for display. In order to adapt to the high data throughput of the overall ASIC, the RTO utilizes several graphical engines in an interleaved approach. "

The more frequent you can update the screen, the less waveforms you need to remember for each the update, the smaller it can be. Thus you save costs. So, the fastest you can go depends on your hardware, i.e. the ASIC and the TFT screen.
I also believe you need to constantly send data to some TFT screens anyway, especially if there is no logic of memory on there. So why not send a fresh image everytime, anyway.

Agilent are doing it very smart, they integrate the menu's on the ASIC so the CPU is no longer important in the oscilloscope process (just like Dave explained in the review and teardowns). I wonder if they also use a special displays that can be driven on high speeds, as that will decrease the required segmented memory size.

I'm still interested what Uni-T came up with. I have a feeling most Chinese companies use off the shelf parts. But if they can pull it off, that would be very interesting.
 

Offline hlavac

  • Frequent Contributor
  • **
  • Posts: 533
  • Country: cz
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #45 on: July 14, 2012, 11:47:36 pm »
That Unique scope looks like the UNI-T piece of crap I have (UT2102C):

It has an annoying firmware bug in trigger level (it thinks it sets it by millivolts but they are volts really). I think most of these mysterious clones have that bug.
Good enough is the enemy of the best.
 

Offline adamhixon

  • Contributor
  • Posts: 6
Siglent SDS1022c vs SDS1062c
« Reply #46 on: July 15, 2012, 02:09:22 am »
I see on eBay the Siglent SDS1022c is actually a few dollars more than the Siglent SDS1062C.    From what I can tell the SDS1062C is the 60MHz version of the SDS1022c.  Shares a manual and what not.  Anyone know why the 25MHz one is a bit more expensive?  Is it superior in some other way that I'm not seeing?

Considering my first hobby scope by the way.  No rush on my purchase though.  - Thanks
 

Offline T4P

  • Super Contributor
  • ***
  • Posts: 3706
  • Country: sg
    • T4P
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #47 on: July 15, 2012, 03:31:21 am »
Hold on mr.scope guy,
if you think UNi-T's bench equipment is good i suggest you search everywhere on eevblog forums for UT801,803,804
1)UT801 : POS, overpriced and it's basically a ut33 inside
2)UT803 : this one goes for a very high price, which is a 61D inside, meh POS lunchbox
3)UT804 : Classic Lunchbox! POS! It's a 71C inside
4)UT805 : Need i say more?
 
The following users thanked this post: Xenoamor

Offline rohitdesa

  • Contributor
  • Posts: 7
  • Country: in
    • NutsandBoltsandFlyingSparks
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #48 on: July 15, 2012, 05:01:47 am »
I'm closely following this discussion. I'm thinking of picking up my first scope too, most likely online from AliExpress. Budget is about USD400. Atten/Siglent/Rigol all have 100MHz, 1Gs/s scope that fit into this price bracket so I'd love to see a shootout/multiple-video review.

Right now I have a Hantek DSO2090 PC scope. Yes I know some people on the forum will scoff at a PC scope, but I was still a student at that time on a tiny budget, and the '2090 worked pretty well. Yes, naturally, signal integrity and accurate waveform reproduction are nowhere close to Agilent/LeCroy, but having said that, I can say that as a student-hobbyist I haven't come across anything that the DSO2090 hasn't been able to satisfactorily measure. It was excellent for university robotics projects and general hacking/tinkering in the lab. My only grouse is that its trigger menu is a little too simple. Only rising and falling edges; no pulse trigger, not even a holdoff (I learned to 'use' the holdoff function from an earlier blog video - cheers Dave!). Maybe you could include this one in the shootout too?
Cheers!
Rohit
 

Offline T4P

  • Super Contributor
  • ***
  • Posts: 3706
  • Country: sg
    • T4P
Re: EEVblog #310 - Cheap Siglent and Agilent scopes
« Reply #49 on: July 15, 2012, 07:12:20 am »
Owon's SDS7102 is also under 400USD
That said dave has said why PC scopes are a POS for their price
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf