GDS-2000A series New economic oscilloscope by GW Instek comes to market

[Hydrawerk]

Hello, what do you think about the new GDS-2000A series? GW Instek web here. You can download manual or specification there. It seems to be a serious competitor for Rigol DS2000 series. The price for GDS-2000A is quite the same. 100 MHz, 2-channels  GDS-2102A is sold for $1,143 at Testequity.
Please do not confuse the GDS-2000 with GDS-2000A, they are totally different. See attached pictures. Yes, I do not know, why they chose the GDS-2000A name...
And features, as specified by the manufacturer.
*300MHz/200MHz/100MHz/70MHz Bandwidth ,2 or 4 Input Channel
*2GSa/s Real-time Sampling Rate and 100GSa/s Equivalent Time Sampling Rate
*2MegaPoints Record Length (When using one channel, probably...)
*1mV /div to 10V/div of Vertical Range
*1ns/div to 100s/div of Time Base Range
*80,000 wfm/s of Waveform Update Rate (I wish it was true...)
*8 inch 800*600 High Resolution TFT LCD Display (Very good!)
*Built-in Segmented Memory and Waveform Search Functions to Optimize the Efficiency of Record Length
*Zoom Window and Play/Pause can Rapidly Navigate the Waveforms
*36 Automatic Measurement Functions Offers Various Measurement Selections
*Optional 8 or 16 digital channel with Logic analyzer(MSO)
*Optional Function Generator (but probably only to 3 MHz)
*Flexible Remote Control Connectivity(Standard:USB ;Option:LAN/GPIB)

Finally, it has a better XY mode than Rigols... It has cursors in XY mode.
Update:
Photos and screenshots here. Please note that I was not an experienced user of this scope, when I took the pictures.  https://plus.google.com/photos/106264218831814439783/albums/5857196858625060337

Dave's video

Teardown made by Dave

[EEVblog]

It has 4 channels and mixed signal capability. The Rigol DS2000 is no match for that.

Dave.

[EEVblog]

I just noticed mention of the function gen  (the two unlabled connectors presumably, dual channel?)
But they have no specs on that... 

Dave.

[EEVblog]

Clearly aimed at the Agilent 2000X series.
Even has the multi channel demo signal connectors.

Dave.

[EEVblog]

Ah, dual channel 3MHz DDS function gen. It's in the datasheet.
LAN/SVGA is optional, just like on the Agilent.

Dave.

[Hydrawerk]

Even has the multi channel demo signal connectors.

According to the user manual, the "Trig Out" signal can be set to these connectors. So it should be quite easy to measure if it really has the 80,000 wfm/s of Waveform Update Rate. You would only need a frequency counter.
I wonder if this GW Instek is better made and if it has less bugs than Rigol DS2000... Who knows?
GW Instek is a company founded in 1975 with quite good raputation... http://www.gwinstek.com/en/Brandorigin_0.aspx
Rigol was founded in 1998... http://www.rigol.com/html/about/history.shtml

[mAJORD]

Yeah, the func gen and logic analyser appear to be plug in upgrade modules.

$245 for the func gen
$570 for 8c Logic

from tequipment.net

[mAJORD]

Can't seem to edit in tapatalk, and work has blocked eevblog (lol) so sorry about these posts..

Was going to say, the upgrades also appear cheaper than Agilents equivilent "soft" upgrades.

Only downside is no bw upgrades

[Hypernova]

Looks like serious competition to the 3000X, with the wavegen and MSO options it is feature wise very close yet way cheaper.

For a 4+16 MSO with wavegen at 100MHz:
$1,694+$830+$245=$2,769

[edit]

Just saw that it has color mode on the digital phosphor too, that's pretty nice. Not having colour mode was one of my disappointments with my 3000X especially after seeing it in action on the Lecroy's at work.

[Bored@Work]

I wonder if this GW Instek is better made and if it has less bugs than Rigol DS2000... Who knows?
GW Instek is a company founded in 1975 with quite good raputation... http://www.gwinstek.com/en/Brandorigin_0.aspx

Instek makes nice oscilloscopes, and everything is fine as long as they work. They let us badly down when we needed to get one repaired. They pointed to the distributor we bought from (authorized and all that), who wasn't at all equipped to service the oscilloscope and shipped the oscilloscope to yet another distributor who also failed to service it. From there it went to a third distributor, still not a real repair workshop. The couldn't even come up with a cost estimate. Month later it was returned unrepaired to the first distributor who returned it unrepaired to us. And in the end we were blamed for dareing we asked to get it repaired, and the first distributor had the galls to bill the shipping costs for that desaster to us.

During all this Instek "support" in Taiwan couldn't care less. The showed the typical Asian attitude, going into stelth mode the moment there was a problem.

We scraped the oscilloscope and blacklisted Instek.

[Nermash]

I am glad to see that they fixed IMHO the bigest failing point of their GDS 3000 line, which was abysmally small capture memory depth of 25 kpoints...

[andersm]

Instek makes nice oscilloscopes, and everything is fine as long as they work. They let us badly down when we needed to get one repaired.

Here in Finland we've gotten our Instek scope repaired a couple of times without problems.

[Hydrawerk]

Well, the GDS-2000A is probably so new, that nobody has bought it yet...

[Hydrawerk]

During all this Instek "support" in Taiwan couldn't care less. The showed the typical Asian attitude, going into stelth mode the moment there was a problem.

Well, do you think that Rigol, Owon or Tekway are better? Anyway, if I needed a oscilloscope for my company, where the instrument runs 12 hours a day, I would buy an Agilent, or so. But for hobby use the GW Instek could be OK...

[Chalky]

Has anyone seen these working?  Claiming 800x600 screen resolution, but all the screen images I can find look really crappy - not just shonky in-house fonts, it looks more like they're doing pixel doubling.

[marmad]

The problem with the GW-Instek DSOs seems to be getting your hands on one to try. On paper, both this and the GDS-3000 series seem very nice - but I couldn't even get GW-Instek to respond to a couple of emails when trying to locate a 3000 series to evaluate. Perhaps I wasn't a big enough customer? I never did figure out what the problem was.

In any case, even though their DSOs appear well-designed and feature-laden, if they really want to succeed in the lower end of the market, they really need to do some work on their distribution and customer-interface mechanisms - even before tackling after-sales support.

[digsys]

Agreed. I've tried SEVERAL times over the last 2 yrs to get responses from them on a few of their products.
Even applied for QTY pricing - as LONG AS they answered some questions ... NADA or bugger all reply.
I also don't like that the FEW resellers they have, ALL have pretty much EXACTLY the same price ! Sus.
I'll definitely be waiting until we get a few reports from brave? pioneers :-)

[mAJORD]

Has anyone seen these working?  Claiming 800x600 screen resolution, but all the screen images I can find look really crappy - not just shonky in-house fonts, it looks more like they're doing pixel doubling.

Found this on their site.. Some doofus saved it in JPG though instead of PNG   

http://www.gwinstek.com/webmanage/IMG/uploadfiles/DS0020.jpg

Well, do you think that Rigol, Owon or Tekway are better? Anyway, if I needed a oscilloscope for my company, where the instrument runs 12 hours a day, I would buy an Agilent, or so. But for hobby use the GW Instek could be OK...

I'd say the opposite vs those brands..   Instek are positioned more towards budget commercial than 'hobbiest' . Like a poor mans HP / Agilent. 

I wouldn't lump Rigol in with OWON either..

[marmad]

Combing through the data sheet and owner's manual, it mentions that the DSO can capture up to 2048 successive waveform segments with a time-tag resolution of 8ns.

Hmm... do they mean the same thing by those terms as I understand them to mean? If so, that implies a waveform update rate of 125M wfrm/s - the reciprocal of 8ns. For example, the Rigol DS2072 can capture up to 65000 successive waveform segments at a minimum time-tag resolution of 20us - the reciprocal of 50k wfrm/s.

[Chalky]

I call bullshit on GW-Instek (and Hantek & Owon).

I am very sus of these budget scope makers who have rushed out larger screen 'resolutions'.  I have had a good zoomed-in look at the images for both their 800x600 screens, the Owon is certainly using vertical pixel doubling for the waveform display.  The Instek seems to be doubling (or worse) as well, but some doubt must be given due to no decent images available yet.

Have a look at the attached images - every single waveform pixel is actually 2 dots tall, by 1 wide (compare to 1x1 graticule dots for reference) .  Now I realise that screen resolution does not equal measurement resolution, but if the scope is joining the dots, as per the default display mode, then I expect an expensive 800x600 scope to actual use that resolution for drawing the bl**dy lines!  This is not just 'oh we've used a 2-pixel thick line' - the displayed line only steps up or down 2 pixels at a time.  Seems to me the usual story - buy Rigol or better!

[mAJORD]

I'm not 100% sure how they normally go about it, but on such a square waveform (i.e the Instek screenshot) , I think you'll find the Rigol , and others for that matter would still show two pixel jumps on an 8 bit DAC. I can't see how you could interpolate those pixels below 2..  Maybe at a pinch on the second high transision there where it's a bit rounded off.

The fact the little noise blips on the top are 3 pixels high suggests it's not pixel doubling?? 

I think you'd need a Sinewave screenshot before even suspecting that to be honest 

[Chalky]

I'm not 100% sure how they normally go about it, but on such a square waveform (i.e the Instek screenshot) , I think you'll find the Rigol , and others for that matter would still show two pixel jumps on an 8 bit DAC. I can't see how you could interpolate those pixels below 2..  Maybe at a pinch on the second high transision there where it's a bit rounded off.

The fact the little noise blips on the top are 3 pixels high suggests it's not pixel doubling?? 

I think you'd need a Sinewave screenshot before even suspecting that to be honest 

Hi mAJORD, yeah I agree re the Instek 2000A & square wave, there might be something else going on there.  Have attached a sine wave showing more clearly.  If it's pixel doubling, it's the wrong approach, since the 8-bit ADC would then need 256x2=512 pixels to display, most of these scopes have less than that in visible waveform area on screen - so then it becomes not a 'nice' display issue, they are actually not able to represent all available information.  I think their displays are just terrible.  Contrast that with Rigol and Agilent who seem to be doing their line(between 2 samples)-to-bitmap rendering correctly.

[andersm]

Is it a real issue beyond looking ugly? For accurate readings you have to use the cursors anyway, trying to count pixels doesn't really work. The display issues could be solved by using anti-aliasing, but the used chipsets may not be up for that.

[tinhead]

I call bullshit on GW-Instek (and Hantek & Owon).

I am very sus of these budget scope makers who have rushed out larger screen 'resolutions'.  I have had a good zoomed-in look at the images for both their 800x600 screens, the Owon is certainly using vertical pixel doubling for the waveform display.  The Instek seems to be doubling (or worse) as well, but some doubt must be given due to no decent images available yet.

Have a look at the attached images - every single waveform pixel is actually 2 dots tall, by 1 wide (compare to 1x1 graticule dots for reference) .  Now I realise that screen resolution does not equal measurement resolution, but if the scope is joining the dots, as per the default display mode, then I expect an expensive 800x600 scope to actual use that resolution for drawing the bl**dy lines!  This is not just 'oh we've used a 2-pixel thick line' - the displayed line only steps up or down 2 pixels at a time.  Seems to me the usual story - buy Rigol or better!

i have to quote this.

First of all you can't simply take a unknown pictures from unknown setups and compare them.

Attached pictures from Tekway/Hantek, screen_1.png shows a typical screenshot, here with avg. wave. We can perfectly see
the 2x2 pixels when we zoom (screen_1_zoomed).png <- i've made a small scale so you can count pixels, one DIV is 50pix high.
But simply let snap that what the DSO is capable to do -> screen_1.png which is a noisy waveform. Can you see the diff
in the zoomed picture ? So yes, Tekway/Hantek can do 1x1 pixels on the screen.

[Hydrawerk]

The problem with the GW-Instek DSOs seems to be getting your hands on one to try.

There will be a trade fair in my city... http://www.amper.cz/ The Micronix company will probably display GW Instek products there. http://eshop.micronix.cz/merici-technika/elektricke-veliciny/osciloskopy/digitalni/stolni/gds-2102a.html
Well, do you think that Rigol is better than GW Instek?