Rigol HDO1000 and HDO4000 12bit oscilloscopes launched in China

[2N3055]

21uV AC RMS noise at full 250M memory 100uV range 1ms/div. No difference on 200uV and 500uV ranges because they are physically the same.
17uV at 1M memory which meets the 18uV spec.
Attached plot is with input internal 50ohm terminated. Same with 1M and external 50ohm terminator.
Curiously the measurement stats stop at 1000 samples.

From datasheet:

[4]: 100 μV/div, 200 μV/div, and 500 μV/div are a magnification of 1 mV/div setting. For vertical
accuracy calculations, use full scale of 8 mV for sensitivity setting.

also you said : Under 1mV is bandwidth limited to 20MHz ....

That is interesting. SDS2000X HS is NOT BW limited at 500uV /div. Could it be that Rigol is doing some kind of eres for those 3 magnified ranges? That would be clever trick but trick nevertheless.

[rf-loop]

21uV AC RMS noise at full 250M memory 100uV range 1ms/div. No difference on 200uV and 500uV ranges because they are physically the same.

Seems like they knocked it out of the park then.

(so far out that they added a 100uV range)

[4]: 100 μV/div, 200 μV/div, and 500 μV/div are a magnification of 1 mV/div setting. For vertical
accuracy calculations, use full scale of 8 mV for sensitivity setting.

Naturally Siglent can vertical zoom. And down to 1µV/div - if someone think it is somehow fun in this park... (only difference is that we do not have full screen zoom but splitted)
But still it looks like Rigol is not bad if look low frequency "1/f" noise based to now displayed picture..

[2N3055]

21uV AC RMS noise at full 250M memory 100uV range 1ms/div. No difference on 200uV and 500uV ranges because they are physically the same.

Seems like they knocked it out of the park then.

(so far out that they added a 100uV range)

Rigol is doing some kind of interpolation/eres below 1mV/DIV. Clever and looks to be working OK.

But, they didn't knock anything out of the park. SDS2000X HD noise figures are for 500 MHz BW.
With only 20MHz limiter on, it has 22uV of RMS noise at 1mV/div
If I do the same trick on SDS200X HD  and add ERES3 I get 18.5uV RMS at 500uV/div.
And since SDS2000X HD supports vertical zoom, here is 2uV/div display in all of it useless glory...

Anyways, it looks quite decent noise vise. Good job Rigol, finally. Now you need to do all of them with that frontend.
I'm looking forward to see how scope (as a whole) actually works. Maybe they are on the right track this time, looks promising..

[EEVblog]

From datasheet:
[4]: 100 μV/div, 200 μV/div, and 500 μV/div are a magnification of 1 mV/div setting. For vertical
accuracy calculations, use full scale of 8 mV for sensitivity setting.

That doesn't explain why the noise floor is very different on the 1mV range vs 500u/200u/100u.

[2N3055]

From datasheet:
[4]: 100 μV/div, 200 μV/div, and 500 μV/div are a magnification of 1 mV/div setting. For vertical
accuracy calculations, use full scale of 8 mV for sensitivity setting.

That doesn't explain why the noise floor is very different on the 1mV range vs 500u/200u/100u.

What was the noise floor at 1mV/div if I may ask? 85-95uV RMS full bandwidth and 25-30uV RMS with 20MHz BW?

My second comment might explain it: 100,200 and 500uV is interpolated from 1mV. And BW limited at 20MHz. You can do both if you do ERES kind of interpolating/filter transform. That would explain both and numbers are really close to what I can simulate here with other scope.   
That would be considered a trick, but a very usable one. They just need to explain what are they doing in datasheet.

Dave, could I make a suggestion?  In datasheet I read spec about dynamic range of instrument that is very confusing. One thing to test would be overdrive recovery, i.e what is non distorted range "outside screen"  vertically. Old MSO5000 chipset had decent overdrive recovery and range, hope they didn't compromise that here to squeeze as little noise as they can.

Best,
Sinisa

[thm_w]

Annoyingly, I can't seem to find a way to turn of sinx/x interpolation, and it does it based on the stopped displayed data. So when you freeze waveform and then attempt to zoom in you can't get to see the quantisation level like I can on my R&S2000 I used a comparison

Could have something to do with the UltraAcquire mode? Maybe that turns off sinx/x.

Max. Waveform Capture Rate
50,000 wfms/s (Vector Mode)
1,500,000 wfms/s (UltraAcquire Mode)

[Fungus]

5 times better... that is little over 2 bits. IIRC the ENOB of the Rigol MSO5000 is around 5 or so. So this new '12 bit' scope would have an ENOB around 7 to 8 bit.

FWIW the datasheet specifically claims more than 8

[EEVblog]

From datasheet:
[4]: 100 μV/div, 200 μV/div, and 500 μV/div are a magnification of 1 mV/div setting. For vertical
accuracy calculations, use full scale of 8 mV for sensitivity setting.

That doesn't explain why the noise floor is very different on the 1mV range vs 500u/200u/100u.

What was the noise floor at 1mV/div if I may ask? 85-95uV RMS full bandwidth and 25-30uV RMS with 20MHz BW?

Sorry, I goofed that due to an annoyign bug that switches off 20MHz BW when you switch from 1mV to 500uV and then back again. So the BW limit was switched off.
1ms/div 1Mpoint, 1mV/div 20MHz BW noise is 17.8uV AC RMS
No BW limit is 55uV AC RMS

[EEVblog]

FYI 1mV range has 240MHz bandwidth on the 200MHz model.

[EEVblog]

Unboxing

[2N3055]

From datasheet:
[4]: 100 μV/div, 200 μV/div, and 500 μV/div are a magnification of 1 mV/div setting. For vertical
accuracy calculations, use full scale of 8 mV for sensitivity setting.

That doesn't explain why the noise floor is very different on the 1mV range vs 500u/200u/100u.

What was the noise floor at 1mV/div if I may ask? 85-95uV RMS full bandwidth and 25-30uV RMS with 20MHz BW?

Sorry, I goofed that due to an annoyign bug that switches off 20MHz BW when you switch from 1mV to 500uV and then back again. So the BW limit was switched off.
1ms/div 1Mpoint, 1mV/div 20MHz BW noise is 17.8uV AC RMS
No BW limit is 55uV AC RMS

Thanks a lot for for answering.. That is a decent result.

P.S: my estimates were for full BW on 800Mhz version...

[EEVblog]

P.S: my estimates were for full BW on 800Mhz version...

I only have the 200MHz version. I don't know of any hack for this yet.

[EEVblog]

I totally missed that this has a battery pack connector sliding door on the back!

[Fungus]

I totally missed that this has a battery pack connector sliding door on the back!

I was sure you were going to mention those slide out feet/supports on the back as well. You usually go for that sort of thing.

Looks like they've done a LOT of work on the UI. This is a complete rewrite/change of OS.

Can't wait to see the 1000 series. It definitely looks like it could be the new champion, the start of a new whole generation of sub-$1000 'scopes.

[2N3055]

Dave, fact check, you are wrong on memory management, SDS2000X HD has manual memory management. "zoom out" comment is incorrect..

[Fungus]

It looks like the statistics keeps the last 1000 readings, not that it "stops" after 1000. The numbers on screen are still updating even though the counter is stopped at 1000.

That's quite handy - you don't have to keep resetting so much. It also fits with the "stats are always on" philosophy of the 'scope. At any time you can just open up the stats display and the numbers will be there instantly. No need to wait or hit reset.

They've thought about it...

Is there a way to change the "1000" to a different value? I could see myself wanting a bit less, eg. stats over the last two seconds.

[hhappy1]

I enjoyed the video.
Input dc bias and offset adjustment are separate. It's very good.

[Kleinstein]

There is nothing wrong with using the BW limit and Eres (digital averaging/filtering of consecutive samples) for the < 1mV/div modes. The noise makes those ranges without the BW limit hardly usefull.
It is also OK to have only the 20 MHz limit in hardware - though the higher BW modes could make good use of enabling the 200 / 400 MHz limit to reduce aliasing. Fine adjustables filtering could be done in the digital domain with "Eres" mode.

The settings menue shows the block diagram with a seprate amplifier for the 50 ohms input. It would be interesting if this is really the case, or just a 50 ohms termination. A different amplifier may be lower noise.  So the noise test could show a difference between internal 50 ohms mode and an external 50 ohm termination.

[EEVblog]

That's quite handy - you don't have to keep resetting so much. It also fits with the "stats are always on" philosophy of the 'scope. At any time you can just open up the stats display and the numbers will be there instantly. No need to wait or hit reset.

Yes, I remarked how I liked that it just there ready to pull down. The font changing to a larger one when minimised was a nice touch.

Is there a way to change the "1000" to a different value? I could see myself wanting a bit less, eg. stats over the last two seconds.

Not that I could find.
A "last n seconds" rolling stats would be a very nice function, with the number changing based on the sample rate and memory depth that changes the capture rate. I don't think any scope has this?

[EEVblog]

Dave, fact check, you are wrong on memory management, SDS2000X HD has manual memory management. "zoom out" comment is incorrect..

They must have changed it on the HD model then, because that wasn't the case for previous Siglents I've looked at, as seen in my video on the topic. In fact it was the inspiration for the video at the time.

[EEVblog]

I totally missed that this has a battery pack connector sliding door on the back!

I was sure you were going to mention those slide out feet/supports on the back as well. You usually go for that sort of thing.

I actually did but cut that part out in the edit when I was trying to get the video under 30min.

Looks like they've done a LOT of work on the UI. This is a complete rewrite/change of OS.
Can't wait to see the 1000 series. It definitely looks like it could be the new champion, the start of a new whole generation of sub-$1000 'scopes.

I haven't been keeping up to date with the Rigols, is this different to the other high end units that have came out recently? Is it entirely new?
In any case I rather like it.

[EEVblog]

The settings menue shows the block diagram with a seprate amplifier for the 50 ohms input. It would be interesting if this is really the case, or just a 50 ohms termination. A different amplifier may be lower noise.  So the noise test could show a difference between internal 50 ohms mode and an external 50 ohm termination.

That would be interesting. Should be obvious on the teardown.

[Fungus]

... there's a lot of little touches in there.

eg. The little "slide across to the next channel's parameters" animation when you change between channels at 6:12.

Totally unnecessary, but nice... 

(just watching the video again)

I also like the trigger slope indicator on the front panel. Done with LEDs!



Plus you get a special button to switch between rising/falling trigger.

Again: Totally unnecessary (just look at the wave on screen to see if it's rising or falling!), but...

[2N3055]

It looks like the statistics keeps the last 1000 readings, not that it "stops" after 1000. The numbers on screen are still updating even though the counter is stopped at 1000.

That's quite handy - you don't have to keep resetting so much. It also fits with the "stats are always on" philosophy of the 'scope. At any time you can just open up the stats display and the numbers will be there instantly. No need to wait or hit reset.

They've thought about it...

Is there a way to change the "1000" to a different value? I could see myself wanting a bit less, eg. stats over the last two seconds.

Statistics should be configurable, probably is. 1000 points is arbitrary number, sometimes too large, sometimes too small. Most scopes are doing running stats (i.e. last x measurements) that's nothing special, but how it should be.
More interesting is how they do measurements.

[2N3055]

Dave, fact check, you are wrong on memory management, SDS2000X HD has manual memory management. "zoom out" comment is incorrect..

They must have changed it on the HD model then, because that wasn't the case for previous Siglents I've looked at, as seen in my video on the topic. In fact it was the inspiration for the video at the time.

Yes that was added later.. But have been so for some time. That video is more that 2 years old now.... Time flies...!!