Rigol DHO804 Test and Compare Thread

[Martin72]

Hi,
Measured the DS1054Z today at work..
"Usable" values:

Rigol DS1054Z: Width: 115mm / Height : 70mm /Diagonal: 135mm
Rigol DHO804:  Width: 150mm / Height: 60mm / Diagonal : 160mm

This is probably called an optical illusion.

[Fungus]

Hi,
Measured the DS1054Z today at work..
"Usable" values:

Rigol DS1054Z: Width: 115mm / Height : 70mm /Diagonal: 135mm
Rigol DHO804:  Width: 150mm / Height: 60mm / Diagonal : 160mm

This is probably called an optical illusion.

That's about what I expected - DHO is wider because there's no side menus but vertical is less because the bars at top/bottom are bigger.

The DHOs have 12% more overall area... but really it's a toss-up because vertical is arguably more important than horizontal.

OTOH the DHOs have nearly twice as many vertical pixels so they easily win in terms of information displayed.

Edit: Not true, they only have 25% more vertical pixels (600, not 800 as I thought)

[TurboTom]

Interesting find -- the DHO900 (and probably the 800 as well) invokes automatic boxcar averaging if the sampling rate/sample memory combination is lower than the speed the physical sampling engine is running at. I recorded a "null" signal (input terminated 50 Ohms) into a 1kSa buffer at 1V/Div and maximum sample rate. Then I exported the buffer via CSV and visualized it via a spreadsheet program since this way, I've been able to retrieve sample data and won't get confused by the sinx/x interpolation.

The bin spacing clearly shows that we've got 4096 bins (12 bit) of sample depth and an input range at the 1V/Div sensitivity setting of ~9.5Vpp, so in total approx. 0.75 vertical divisions beyond the visible screen range.

After that, I changed the time base to result in a sampling rate of 100kSa/s and did the same test -- this time observing a much closer bin spacing which is somewhat more difficult to analyze, but it appears now we've got close to 20500 bins (~14.3 bit) of resolution.

So the arrangement of the shadow memory has got to be much more complex than I initially assumed here https://www.eevblog.com/forum/testgear/rigol-dho804-test-and-compare-thread/msg5108949/#msg5108949. Probably the shadow memory is much wider and the 12-bit wide sampling memory area to be displayed gets copied/added at a certain position to result in the actual / decimated sample rate that's selected. This way, the boxcar averaging happens more or less automatically and virtually doesn't take any additional time, at least if it's done in multiples of 2.

[Njk]

Hi,
Measured the DS1054Z today at work..
"Usable" values:

Rigol DS1054Z: Width: 115mm / Height : 70mm /Diagonal: 135mm
Rigol DHO804:  Width: 150mm / Height: 60mm / Diagonal : 160mm

This is probably called an optical illusion.

Hi Martin,
Thanks for the data. So my illusion is proven correct. It is vertically smaller. Although TV-like "aspect ratio". OHOH I have old analog scope where the screen is of 42x60mm. No problem. But it's true CRT

[TurboTom]

...
That's about what I expected - DHO is wider because there's no side menus but vertical is less because the bars at top/bottom are bigger.

The DHOs have 12% more overall area... but really it's a toss-up because vertical is arguably more important than horizontal.

OTOH the DHOs have nearly twice as many vertical pixels so they easily win in terms of information displayed.

DS1000Z: 800 x 480
DHO800 / 900: 1024 x 600

This is 25% more vertical resolution than the DS1000Z, which easily explains the somewhat crowded appearance of the DHO800 / 900's screen vs. the DHO1000 / 4000 series...

[Martin72]

There's a pop-out side panel with measurements on it. Any measurement can be expanded to see stats by clicking the arrow at the bottom of the measurement.
Notice how much faster/easier it is to do than menu diving and using a twisty knob to turn things on/off.

Seconds counts..
Yes, that's not bad - But it would only be consistent if you could add several parameters once you have already activated the measuring panel.
This is not possible, you always have to call the measuring menu for each new parameter.

[nctnico]

...
That's about what I expected - DHO is wider because there's no side menus but vertical is less because the bars at top/bottom are bigger.

The DHOs have 12% more overall area... but really it's a toss-up because vertical is arguably more important than horizontal.

OTOH the DHOs have nearly twice as many vertical pixels so they easily win in terms of information displayed.

DS1000Z: 800 x 480
DHO800 / 900: 1024 x 600

This is 25% more vertical resolution than the DS1000Z, which easily explains the somewhat crowded appearance of the DHO800 / 900's screen vs. the DHO1000 / 4000 series...

Resolution doesn't matter. Screen size does. A higher resolution (as in smaller dot size) gives a nicer image though. In some of my own projects that need a nice graphic interface I choose a display with a higher resolution so the UI is prettier to look at.

[Martin72]

Last play before it´s getting serious...

Testing the decodings...
With the batronix board, RS232, SPI and I²C.
First the positive:
The table, I love how clearly you can see the values through the white/gray, that's a real benefit, I'd love to see it on my Siglent too.
Definitely a plus point.
And:
You can move the decoded values up/down (see RS232 and 232_1), very nice to make it better visible.
Then less great:
The menu...You go in via the selection button at the bottom left of the screen, so far so good.
But as soon as the decoder menu is hidden, you have to go the same way again, there is no dedicated button for it.
Also, the equipment for the respective decoders is rather sparse, as with the FFT.
And there is no copy to trigger function, only the other way around.
And I kept falling for the "Roll Mode" earlier:
Default is from 50ms/div the roll mode active, I find not bad.
However, one is then immediately thrown out of the decoding and rigorously:
No warning, it is simply deactivated and you have to reactivate it.
SPI and RS232 were easy to set, with I²C it is a bit more difficult, you can't get a halfway stable trigger.
But I'll deal with it more deeply when I test the demo board completely on the Rigol.

[Martin72]

Resolution doesn't matter. Screen size does. A higher resolution (as in smaller dot size) gives a nicer image though. In some of my own projects that need a nice graphic interface I choose a display with a higher resolution so the UI is prettier to look at.

And it really is, it just looks better and sharper than the other 7" scopes I know.
The same effect can be seen on the 10" models, it just looks better, so the UI.

[tv84]

Resolution doesn't matter. Screen size does. A higher resolution (as in smaller dot size) gives a nicer image though. In some of my own projects that need a nice graphic interface I choose a display with a higher resolution so the UI is prettier to look at.

And it really is, it just looks better and sharper than the other 7" scopes I know.
The same effect can be seen on the 10" models, it just looks better, so the UI.

Sincerely I have an hard time swallowing these DHO pictures. The menus seem rendered in 4K but the signal looks like Minecraft. 

Well, maybe it's just me.. 

[Fungus]

DS1000Z: 800 x 480
DHO800 / 900: 1024 x 600

My bad, I thought it was 800 pixels vertical.

[Martin72]

Sincerely I have an hard time swallowing these DHO pictures. The menus seem rendered in 4K but the signal looks like Minecraft. 
Well, maybe it's just me.. 

No.
I see it exactly the same way.
The presentation of the graphical elements benefits a lot from it, everything just looks sharper/better.
But the actual signal does not benefit from it.

[Fungus]

No.
I see it exactly the same way.
The presentation of the graphical elements benefits a lot from it, everything just looks sharper/better.
But the actual signal does not benefit from it.

The UI graphics and text are scaled vectors, they can be antialiased.

The traces aren't, they can't.

[Martin72]

This does not change the fact that it is visually striking.
But that's not a big minus point now.
I have now ordered USB adapter, A/B.
This gets on my nerves, this eternal swap because only one port A is available.

[Martin72]

Can you give us some of your personal pros/cons for comparison of MSO5000 and DHO800/900?
I'm also wondering why u decided to get this new DHO, because of the smaller size or fancy 12-bit resolution?

Just out of curiosity, that alone is worth it to me.
I don't like to talk about things I don't know.
I think a comparison to the MSO5074 is still premature, I'll comment on that when I've tested more.

[ebastler]

Sincerely I have an hard time swallowing these DHO pictures. The menus seem rendered in 4K but the signal looks like Minecraft. 

Maybe Android will let you install crude bitmap fonts to make the display look more familiar.

[2N3055]

Sincerely I have an hard time swallowing these DHO pictures. The menus seem rendered in 4K but the signal looks like Minecraft. 

Maybe Android will let you install crude bitmap fonts to make the display look more familiar.

First of all, I know you're joking, but made me want to comment.. I know, shocker, right... 

"Crude" fonts relay same information as super high resolution anti-aliased fonts.

Fonts on very low resolution screen on MSOX3000T are very easy to read and clear. It is waveform that is thick and even that is looking quite good when you enable HiRes. And that is 800x480 pixel 8.5"screen.

All of that is nice but is irrelevant, same as watching some really stupid movie on super hi res 8K screen. Still stupid movie not worth watching. There are many movies I would rather watch on VHS than really bad movie on super resolution.
Race car won't go faster if you paint racing stripes on it... Or coolest painting imaginable...

Super high resolution and/or large screen does not provide more information... It simply presents opportunity to  provide more.
If more information is provided depends on how it is used. Large thick window edges take space, taskbars, icons etc..
And then it is matter of organizing info and how it is presented. etc etc..

Fonts cannot be too small, either.. Scope is not a phone so you stick it in your face. Scope is used at arm's length on a desk, or on a shelf... Physical font size of cca 3mm on the screen is smallest you can comfortably read..
And that is for a person with OK eyesight.  We need to include people that does not have perfect eyesight.
As a side note, on latest FW, SDS2000X HD can change font size for that reason. Some other scopes do that too.

I prefer that scope should minimize usage of screen for anything that is not related to waveforms, or other useful data..

[Martin72]

I prefer that scope should minimize usage of screen for anything that is not related to waveforms, or other useful data..

I would also like to have this with my Siglent.
As with Windows, for example.
Fades out after a few seconds, is back again at the touch of a finger/mouse.

[tautech]

I prefer that scope should minimize usage of screen for anything that is not related to waveforms, or other useful data..

I would also like to have this with my Siglent.
As with Windows, for example.
Fades out after a few seconds, is back again at the touch of a finger/mouse.

Menu fade/hide is a 2 edged sword especially when set to a fast hide as it's often gone before you can take a screenshot. 
We disable menu hide before dispatch so new users don't have it disappear while finding their way around a new DSO.
Later when they are conversant with the UI and featureset they can easily re-enable menu hide or just continue to touch or click a vacant area on the display to hide it.

[Fungus]

We disable menu hide before dispatch so new users don't have it disappear while finding their way around a new DSO.

They enable it by default? 

[ebastler]

I prefer that scope should minimize usage of screen for anything that is not related to waveforms, or other useful data..

I certainly agree that any additional screen detail should (a) not waste space, and (b) not distract from the data and curves. But I don't think Rigol have done too much of that with the DHO800. My first reaction to seeing screenshots from the DHO800 was indeed "They have gone a bit overboard with the decorations". But having looked closer, I don't think so:

The height of the settings displays at the top and bottom of the screen is determined by the need to provide large-enough touch areas there. (And yes, they take up a relatively large part of the small screen.) Given that height requirement, I think Rigol have done a nice job packing useful information into these areas and structuring it well -- and in a way that does not disctract from the net content.

Regarding the anti-aliased, nice-looking fonts: Why not? The screen resolution and grading capability is there anyway (driven by the need to display information-rich curves), so anti-aliased fonts come for free. They do not take up more space than crude bitmap fonts in lower resolution, and they do make the information easier to read.

[tautech]

We disable menu hide before dispatch so new users don't have it disappear while finding their way around a new DSO.

They enable it by default? 

Enabling menu hide is what experienced users need as it saves considerable time however it's a suboptimal choice for the new user of a scope.
However when I brought it to Siglents attention they rightly changed the default setting in later firmware.

Still we have 5 auto hide settings from 3-60s before the menu auto hides or we can leave OFF to be hidden manually with a touch or click of a mouse.
Much depends on how you want to use the scope and the range of options for how the menu interacts with the display where it can float over the display or be embedded and compress the active display.

Choices are good and the more the better.
Edit to add better images.

[2N3055]

Regarding the anti-aliased, nice-looking fonts: Why not? The screen resolution and grading capability is there anyway (driven by the need to display information-rich curves), so anti-aliased fonts come for free. They do not take up more space than crude bitmap fonts in lower resolution, and they do make the information easier to read.

To make it clear I don't mind if things are "prettier"... I'm a geek too and like "cool looking" design.
But I'll take "blocky fonts" and 4 math channels any day over scope that is "cool looking" but has 1 math channel with only +/- operations...
But I don't mind blocky (retro looking) fonts either if they do the job..

I see people getting downright giddy how cool it looks and not even noticing when it is missing basic features...
Form over substance and utility.

And that is fine if I need a movie prop... 

But all I see that super high resolution, anti aliased, fully graphically designed display of screen elements was given priority over waveform update rate.. Phosphorus emulation is worse than DS1000Z and WFMs/s also...
Useless segmented mode is only made for screen display... Etc... Whole purpose of segmented mode is to capture bunch of data blocks with minimum blind time and analyze it (measurements, decode) later. Without that, it has same analytic capabilities as DS1000Z that also had same useless "Recording mode".

To explain it further, while working with a scope I was in a situation where I did say: "darn, if this screen was taller (physically, as in 20cm instead of 14cm) so I can fit 20 decode lines instead of 10 I have now, I would do this job better, faster or whatnot".
Never, ever, I was in a situation that I said : "man, if only these menus were prettier and fonts were nicer looking, I would be able to measure more data at the same time"...

To summarize, I'm not against "cool" and sophisticatedly looking GUI. I'm against that being replacement for actual capabilities.

If you ask me, Rigol should have pushed for 10" screens even with lower resolution than they have now (to make it cheap enough to make to keep the product price range), not use use panoramic display ratio and therefore provide larger screen real estate to show information, even if fonts were a bit blocky...

But... Rigol knows all this. They know that DHO800/900 with lover resolution 10" screen would be perceived as same as DHO1000  (hint: they are pretty much same except screen and slightly higher sampling rate which makes no difference for a 100MHz scope). 800 and 900 are artificially created to make product for that range and make scope lineup differentiation.
Actually DHO900 with lover res 10" screen would make DHO1000 pretty much dead in a water, sales vise ...
Same as there is no DHO1000MSO (HDO2000 ??) at the time. Why? We know motherboard has it all already.

At this moment, they want to sell you DHO4000 if you want a 12 bit MSO in that range...
As competition responds with their models and pricing reshuffling, expect changes... And the game goes on...

[ebastler]

I see people getting downright giddy how cool it looks and not even noticing when it is missing basic features...
Form over substance and utility.
[...]
To summarize, I'm not against "cool" and sophisticatedly looking GUI. I'm against that being replacement for actual capabilities.

Fully agree; choosing a scope based on its looks while ignoring the capabilities is not wise. Although, on second thought, that approach is probably not harmful for the (large?) share of hobbyists who only use very basic DSO functionality anyway -- essentially using it like a CRO which takes up less space, plus maybe the ability to stop and capture a trace. I am pretty sure that Rigol gets an above-average share of those users, and there's no harm in that. (Except for the missed opportunity to discover more advanced scope capabilities...)

If you ask me, Rigol should have pushed for 10" screens even with lower resolution than they have now (to make it cheap enough to make to keep the product price range), not use use panoramic display ratio and therefore provide larger screen real estate to show information, even if fonts were a bit blocky...

Yes indeed. Their screen design works fine on the 10" screen, I think, so there is nothing wrong with the choice of fonts, frames, colors per se. It's just that the settings displays and controls take up a rather large part of the 7" screen, which causes usability compromises some will be happy to accept while others will not.

But... Rigol knows all this. They know that DHO800/900 with lover resolution 10" screen would be perceived as same as DHO1000  (hint: they are pretty much same except screen and slightly higher sampling rate which makes no difference for a 100MHz scope). 800 and 900 are artificially created to make product for that range and make scope lineup differentiation.
Actually DHO900 with lover res 10" screen would make DHO1000 pretty much dead in a water, sales vise ...

No doubt, a lot of these design choices are marketing-driven. And I don't blame Rigol (or the other scope manufacturers) for that. Hobbyists benefit from this "market segmentation" approach, since they get a lot of functionality at an attractive price point, while the manufacturers still find differentiators to motivate commercial customers to pay much more. Screen size and bandwidth are nicely obvious differentiators which pretty much all manufacturers have relied on.

I am curious to see how Siglent will play this with the SDS1000X HD vs the 2000X HD. With the most recent spec changes they have positioned them even closer to each other, so it seems that they will either need to make the 1000X HD relatively expensive, or will lose a significant share of potential 2000X HD customers to it.

[2N3055]

I see people getting downright giddy how cool it looks and not even noticing when it is missing basic features...
Form over substance and utility.
[...]
To summarize, I'm not against "cool" and sophisticatedly looking GUI. I'm against that being replacement for actual capabilities.

Fully agree; choosing a scope based on its looks while ignoring the capabilities is not wise. Although, on second thought, that approach is probably not harmful for the (large?) share of hobbyists who only use very basic DSO functionality anyway -- essentially using it like a CRO which takes up less space, plus maybe the ability to stop and capture a trace. I am pretty sure that Rigol gets an above-average share of those users, and there's no harm in that. (Except for the missed opportunity to discover more advanced scope capabilities...)

If you ask me, Rigol should have pushed for 10" screens even with lower resolution than they have now (to make it cheap enough to make to keep the product price range), not use use panoramic display ratio and therefore provide larger screen real estate to show information, even if fonts were a bit blocky...

Yes indeed. Their screen design works fine on the 10" screen, I think, so there is nothing wrong with the choice of fonts, frames, colors per se. It's just that the settings displays and controls take up a rather large part of the 7" screen, which causes usability compromises some will be happy to accept while others will not.

But... Rigol knows all this. They know that DHO800/900 with lover resolution 10" screen would be perceived as same as DHO1000  (hint: they are pretty much same except screen and slightly higher sampling rate which makes no difference for a 100MHz scope). 800 and 900 are artificially created to make product for that range and make scope lineup differentiation.
Actually DHO900 with lover res 10" screen would make DHO1000 pretty much dead in a water, sales vise ...

No doubt, a lot of these design choices are marketing-driven. And I don't blame Rigol (or the other scope manufacturers) for that. Hobbyists benefit from this "market segmentation" approach, since they get a lot of functionality at an attractive price point, while the manufacturers still find differentiators to motivate commercial customers to pay much more. Screen size and bandwidth are nicely obvious differentiators which pretty much all manufacturers have relied on.

I am curious to see how Siglent will play this with the SDS1000X HD vs the 2000X HD. With the most recent spec changes they have positioned them even closer to each other, so it seems that they will either need to make the 1000X HD relatively expensive, or will lose a significant share of potential 2000X HD customers to it.

There is nothing wrong with instrument being simple. I dislike hype and false expectations.
Frankly, very many people still uses analog CRT scopes and get by with that. In comparison, DHO800 is space ship...

As for SDS1000X HD vs 2000 that one I'm also curious how it's going to play out.
Analytically they are going to be quite close.
But 2000 has twice the sampling troughput (2x2GS/s), 500 Mhz max BW, Zone triggering and MSO and AWG built in...
It is quite a lot more, really...
But hey, they are looking identically, save for the label... 

Jokes on the side, I look at it differently: this is going to be most powerful 1000 series scope ever built....