Rigol's New DHO800 Oscilloscope unbox & teardown

[souldevelop]

Guys, my DHO914S arrives tomorrow and I'm ready to try DIY LA logic module PLA2216, it runs great on my MSO5000 oscilloscope.
And then I will talk about dismantling DHO914S look at the AFG module structure inside it, and by the way, I want to get a free DHO924S.
I also have to assemble VESA100 as a fixation bracket, please wait for my homework.

[Njk]

How is that a "bug"?

Well, let's define bug. It can be defined as something that should not be there. But if I were a bug, I would reply that I'm living there similar to all my ancestors from the entropy family while everyone else is a newcomer that shall be more humble on the first place (refer to the zoot suite issues back at the WW2 times). Or we can define bug as a difference between the expected and the actual behaviors. It also can be called a defect. The best way to make a defect-free product is to don't specify the expected behavior. We can provide it by our own but such defect claims will be baseless from the manufacturer's perspective. No documentation, no defects. In that particular case, the screen size of 7" is advertised but what exactly does it mean? It can be seen on the video that the picture does not occupy whole screen. There are dark areas at the borders. What if there are no pixels at all, only glass, so the working screen size is something about 6 5/9, not 7 (sorry I'm not a master of that weird math)? In that case, it can't be fixed by the firmware. Call it a bug or otherwise but it does not make the picture bigger.

[Dacian]

There are dark areas at the borders. What if there are no pixels at all, only glass, so the working screen size is something about 6 5/9, not 7 (sorry I'm not a master of that weird math)? In that case, it can't be fixed by the firmware. Call it a bug or otherwise but it does not make the picture bigger.

There is a capacitive touch glass on front of the 7" LCD screen and so there are no unused pixels with the black border being the touch glass that is outside of the 7" LCD screen area.
The complain was about the waveform area witch is only 400 pixels in height where LCD has 600 pixels with the 200 being used for the menu. They will like a full screen option for the waveform and consider that to be a bug.
I fully understand why Rigol decided to use 400 lines to display a 12bit (4096) waveform as it a simple divide by 10 and full screen on this 1024x600 screen will just not make sense.
If anyone wants to see a higher resolution waveform then they can just download the raw data and plot in excel or whatever graph application they prefer.
So the fact that a full screen waveform function does not exist is not a bug.
 

[hubertyoung]

hello souldevelop.
I'm happy that you purchased dho924s. I wonder if you know the adb command. I want to replace the complete rigol folder to unlock the highest configuration. Can you share the zip file of this folder? Thank you so much.

[Fungus]

There is a capacitive touch glass on front of the 7" LCD screen

It it a glass screen?

[Njk]

There is a capacitive touch glass on front of the 7" LCD screen and so there are no unused pixels with the black border being the touch glass that is outside of the 7" LCD screen area.

Understood. So the glass size is larger than 7". That's a design decision although it's not clear without the explainer, why not all glass area is used for picture.

The complain was about the waveform area witch is only 400 pixels in height where LCD has 600 pixels with the 200 being used for the menu. They will like a full screen option for the waveform and consider that to be a bug.
I fully understand why Rigol decided to use 400 lines to display a 12bit (4096) waveform as it a simple divide by 10 and full screen on this 1024x600 screen will just not make sense.
If anyone wants to see a higher resolution waveform then they can just download the raw data and plot in excel or whatever graph application they prefer.
So the fact that a full screen waveform function does not exist is not a bug.

So it's exactly the same as with DS1054Z. I also doubt it can be considered a bug, but what was definitely not good is that the scope draws the trace such that the waveform looks clipped while it's actually not as there is some headroom above and below the upper/lower margin on the screen. I wonder if the new model behaves similar

[NE666]

I fully understand why Rigol decided to use 400 lines to display a 12bit (4096) waveform as it a simple divide by 10 and full screen on this 1024x600 screen will just not make sense.

I think it's more likely to be attributable to the amount of shared code, specifically UI and UI elements/assets in this case, across the DHO range.  The menu bar size seems (largely) fixed across the range if you look at the DHO 1000 or 4000 series, and so what's left over for waveform display is purely a function of the total screen size and resolution of a particular model.

Given that, as you rightly point out, there's a lot of pixel mapping going on in the vertical deflection with 12-bit scopes as compared to 8-bit, and also given that HDMI outputs are becoming common, I'm curious as to whether higher resolutions would be viable and practical, when using external displays.  Or whether this has a massive impact on display update rates and hence why it's not being implemented in the DHO.

[Dacian]

It it a glass screen?

LCD screen is 7" and on top of that there is a capacitive glass touch sensor that is slightly larger as it is on almost any other similar device.

Understood. So the glass size is larger than 7". That's a design decision although it's not clear without the explainer, why not all glass area is used for picture.

It was obvious to me as soon as I saw the oscilloscope that there is a capacitive glass touch sensor on top of the LCD screen same as all other oscilloscopes with touch or any tablet. The screen and capacitive glass touch sensor are two separate components.

So it's exactly the same as with DS1054Z. I also doubt it can be considered a bug, but what was definitely not good is that the scope draws the trace such that the waveform looks clipped while it's actually not as there is some headroom above and below the upper/lower margin on the screen. I wonder if the new model behaves similar

The waveform may actually be clipped if the signal voltage is higher than what can be displayed on the screen.  Best guess is that they used 400 lines to display the 4096 points waveform with a 10x divider so there will be only maybe 10 lines outside of the display window that are captured.
You will just set the vertical resolution so that the signal will fit in the display window and then you can zoom in to it after the waveform is captured.
I guess you seen the video where Dave hits the BNC connector but he head no idea what voltage will be generated by that thus it is outside of the measurement range.  You will just change the V/div setting and repeat the test so that you can capture the entire waveform.  This is valid for any oscilloscope. 

[Dacian]

Given that, as you rightly point out, there's a lot of pixel mapping going on in the vertical deflection with 12-bit scopes as compared to 8-bit, and also given that HDMI outputs are becoming common, I'm curious as to whether higher resolutions would be viable and practical, when using external displays.  Or whether this has a massive impact on display update rates and hence why it's not being implemented in the DHO.

8bit is just 256 so easy to fit 1:1 on any screen.
For 10bit oscilloscopes 1024 is also possible to have 1:1 mapping if you have a full HD screen and then it may make sense to remove all menu items to show the waveform full screen to get that nice 1:1 mapping.
But for 12bit it is not cost effective to have an LCD with 4096 lines (that will be a huge vertical resolution) I think there are some 8K TV's or monitors but large size not suitable for a small oscilloscope and clearly not for a budget one.
As for external screen I think the output from HDMI will be the exact same 1024x600 and just a mirror of the internal screen.

If you need to see the full 12bit you will need to zoom in or save the raw waveform and use excel to plot on your computer but you will still not see the full 12bit unless you have an 8K screen so you still need to zoom in.
It is even worse on the horizontal time domain where you may have captured millions of points so there you need to zoom in no matter your screen resolution.   

[Howardlong]

Never really understood why boot time for scopes mattered. I have a RTO1024 as the main scope at work. 2-3+ minutes to boot up plus a recommended 1+ hour warm up time.

Not really apples to apples here but 99 times out of 100 you'd have your scope running and warmed up before doing anything.

For a multimeter however I'd understand.

Nope, scope goes on when I need to use it. Quick boot time is why my go-to scope on the bench is still an Agilent/Keysight MSO7104, despite there being plenty of other otherwise more competent scopes to hand.

What's the difference in the use case with a multimeter?

[Martin72]

I don't find the boot time that dramatic either.
I once worked with a Lecroy WS422 for a while, and when you turned it on, you could still get a coffee and go to the toilet, so you didn't miss anything.
Time is relative.

[NE666]

As for external screen I think the output from HDMI will be the exact same 1024x600 and just a mirror of the internal screen.

It will be for these units, as for the MSO5000, but that was my actual question - does it *need* to be this way?  Will the next logical step for manufacturers be to allow for different screen resolutions when using external displays?  It doesn't seem so unreasonable, now that these are effectively Android-based devices running an oscilloscope UI application.

I've a 3440x1440 native monitor sat on my workstation which of course isn't "12-bit native" but still would be nice to use, compared to a 7" display (if you're my age).

[Howardlong]

I don't find the boot time that dramatic either.
I once worked with a Lecroy WS422 for a while, and when you turned it on, you could still get a coffee and go to the toilet, so you didn't miss anything.
Time is relative.

That's precisely the kind of productivity loss I try to avoid!

When I come to a natural stop in my workflow, I'll make a cuppa. Otherwise I want to keep the momentum going.

Before you know if, that break waiting for your scope to boot turns into an excuse to do all manner of things completely unconnected with the job at hand.

I'm just talking from my own experience: the fewer excuses I have, the more productive I am thanks to an uninterrupted workflow.

But as I work for myself, I guess I have more incentive to be more productive.

[Martin72]

If a boot time of about 1 minute on average causes a noticeable loss of productivity, you should rethink your time management in general.

[EEVblog]

FYI, USB-C power consumption is 15V @37W with 4 channels and math running.

[Fungus]

Most 'scopes these days (including this one) run a variant of Linux so they take time to boot. They have file systems, USB drivers, etc., that probably shouldn't be reinvented from scratch (You want bugs? We've got bugs...!)

OTOH it could have a sleep mode. This is Android...

Maybe the FPGA, front end, etc., can't go to sleep so they'd have to leave most of it powered on, including the fan. I wouldn't want a noisy 20W device constantly running on my desk.

[EEVblog]

Photos of the HDMI output on a 4K monitor. Info says it's actually 3840x2160@60 
And it does look sharper than than the internal 1024x600.
I'm not sure if it's the scope upsacling, or it's the BENQ monitor?

[Fungus]

I'm not sure if it's the scope upsacling, or it's the BENQ monitor?

This suggests it's the 'scope:   


I'd bet against it having more pixels though. It's probably just that 1024x600 is very high definition on a 7" screen so you don't appreciate it.

[EEVblog]

On my 2K monitor, at first is chucked a wobbly and told me 1920x1080 with garbage but after a reboot I got 2560x1440
So it seems like the Rigol is detecting the resolution of the monitor and adjusting it's output?
I can say that it looks very nice and sharp. It's clearly not a fixed 1024x600 output.

[Fungus]

All the menus and fonts are being rendered at the higher resolution?

[StillTrying]

I can't see any extra resolution in the waveform area on HDMI, just bigger.
The Y sample levels look about the same as 8 bit, 25 levels per division, 200 over 8 div./400 pixels.
Is it really 12 bit.

[EEVblog]

All the menus and fonts are being rendered at the higher resolution?

They seem to be. Not sure abotu the waveform though, at least in terms of actual sampled bit rendering.

[BillyO]

I can't see any extra resolution in the waveform area on HDMI, just bigger.
The Y sample levels look about the same as 8 bit, 25 levels per division, 200 over 8 div./400 pixels.
Is it really 12 bit.

[StillTrying]

Dave has just said the same.

They seem to be. Not sure abotu the waveform though, at least in terms of actual sampled bit rendering.

That looks just like the normal 8 bit, 1 bit of noise.

[BillyO]

Dave has just said the same.

Sort of.  However, not displaying 4096 pixels does not imply the scope is not 12-bit.  Nor does it significantly reduce the merit of having a 12-bit front end.

To fully display 12-bits of vertical information you would need better than 8K resolution, assuming you want some other information on the display as well.  I don't think that would be a reasonable ask for a sub $400 scope.  At least today.  Plus, you'd need gigantic monitor to take advantage of that kind of resolution.   I have a 50" 4K TV and when I play full 4K video on it I have to stand within 0.5m to see the pixels.  Then I can't see what it is I'm looking at.

Personally, I think this whole race for more and more display resolution is a waste.  Double would be nice, but in my opinion 10X is just ridiculous for human eyes.  Maybe a hawk, but they don't use oscilloscopes much.