Rigol's New DHO800 Oscilloscope unbox & teardown

[Martin72]

I'll add something here, solder there, change the software...
Evaluation boards with housing.

[.RC.]

What I do not understand is why these manufacturers do not lock down their products so the software cannot be hacked, so people can not simply modify the software and they suddenly have the faster model.

I know why the crippled machines are still mostly the same build as the faster ones, but if I was the manufacturer, I would be making damn sure everything was encrypted or whatever so the cheaper machine can not be made better then the manufacturer wants.

I would be in the game to make lots of the evil stuff called profit, if people want a faster machine, they pay for it.

Unless of course the mindset is, they will sell more of the basic ones if the end user can "improve" it.  If it can not be improved, there is nothing to think people will automatically go buy the more expensive faster one, there might be an intermediate better model by another manufacturer they might buy instead.

[iMo]

I'll add something here, solder there, change the software...
Evaluation boards with housing.

This DHO800 kit looks like a great platform for the hackers and modders.
I think Rigol does that to keep the technical people happy.. Good so..
Btw I wonder what is the mean time between failures for that USB-C connector..

[gjvdheiden]

What I do not understand is why these manufacturers do not lock down their products so the software cannot be hacked, so people can not simply modify the software and they suddenly have the faster model.

I know why the crippled machines are still mostly the same build as the faster ones, but if I was the manufacturer, I would be making damn sure everything was encrypted or whatever so the cheaper machine can not be made better then the manufacturer wants.

I would be in the game to make lots of the evil stuff called profit, if people want a faster machine, they pay for it.

Unless of course the mindset is, they will sell more of the basic ones if the end user can "improve" it.  If it can not be improved, there is nothing to think people will automatically go buy the more expensive faster one, there might be an intermediate better model by another manufacturer they might buy instead.

I suspect it’s all deliberate. This model will also sell to schools. They don’t tincker with it. Some end users don’t fancy hacking it. So there is still profit in those two categories. Also l, these hacks make the lab equipment very popular. People who are enthusiastic about the brand from home lab experience might recommend it at work.

Rigol also saves some money to develop a _more_ secure system. Remember that this also impacts servicing and upgrading. I think it is a deliberate business decision. Might be even designed to be hackable. Keeps maintaining the software easier too.

[svetlov]

for alternative cooling, you can look in this direction - since the oscilloscope has a VESA 100 mount
with threaded bushings installed, you can install an additional fan in this place - a quick selection gives 120 fans - the seats do not match, but the space on the case allows you to make fan new holes after marking - and also, ideally, print your own protective cover on a 3D printer - yes and can be printed in Rigol honeycomb style
despite the fact that the fan operates to blow out - from the grille outwards - the flow direction can be changed if desired or left as is after tests

 https://www.silverstonetek.com/en/product/info/fans/AirPenetrator120i/
https://www.silverstonetek.com/upload/sstedm/ap120i/AP120i-Product_Sheet-EN.pdf

[switchabl]

I'll add something here, solder there, change the software...
Evaluation boards with housing.

This DHO800 kit looks like a great platform for the hackers and modders.
I think Rigol does that to keep the technical people happy.. Good so..
Btw I wonder what is the mean time between failures for that USB-C connector..

A number of people on this thread have said similar things. I don't really understand why but maybe I am missing something?

I can see that with adb root you can tinker with the Android system and sideload your own apps. But when it comes to the parts that actually matter, the complicated architecture (Zynq <-> PCIe <-> SoC <-> RAM) and the undocumented ADC and front end ICs are surely going to take a very broad skillset and a lot of determination to reverse engineer. Especially compared to something like the entry level Siglents, where everything is just off-the-shelf parts connected to a single Zynq and a good deal of work has already been done (https://github.com/360nosc0pe/yocto)?

[Fungus]

for alternative cooling, you can look in this direction - since the oscilloscope has a VESA 100 mount

Yep, that's exactly what I've been thinking.

[Fungus]

What I do not understand is why these manufacturers do not lock down their products so the software cannot be hacked

Because they're smarter than you.

(and have sales numbers to back it up)

[nctnico]

I'll add something here, solder there, change the software...
Evaluation boards with housing.

This DHO800 kit looks like a great platform for the hackers and modders.
I think Rigol does that to keep the technical people happy.. Good so..
Btw I wonder what is the mean time between failures for that USB-C connector..

A number of people on this thread have said similar things. I don't really understand why but maybe I am missing something?

I can see that with adb root you can tinker with the Android system and sideload your own apps. But when it comes to the parts that actually matter, the complicated architecture (Zynq <-> PCIe <-> SoC <-> RAM) and the undocumented ADC and front end ICs are surely going to take a very broad skillset and a lot of determination to reverse engineer. Especially compared to something like the entry level Siglents, where everything is just off-the-shelf parts connected to a single Zynq and a good deal of work has already been done (https://github.com/360nosc0pe/yocto)?

Sorry, but this project is at 0.01% of all work needed (and it is based on Yocto which sucks badly). Oscilloscope firmware with a decent number of features is by far the most complicated piece of firmware that runs in test equipment. You are talking many years worth of work (fulltime!).

If you want added functionality, then acquire the data through LAN and process on a computer. You can use a small box with a Rasberry pi for that if you want to have a small size solution.

[Fungus]

Btw I wonder what is the mean time between failures for that USB-C connector..

It looks like the mounting pins go right through the PCB so it's probably a lot longer than you think.

If you're paranoid you can attach one of these to it and replace as needed:


Bonus: The cable comes out sideways so you gain some bench space.

[sebyon]

What I do not understand is why these manufacturers do not lock down their products so the software cannot be hacked, so people can not simply modify the software and they suddenly have the faster model.

Basically, they need to weigh up the cost X to encrypt an upgrade system that will be abused by Y amount within the total of their market Z that will either:

• Reduce uptake by hobbyists by Y units
• Be cracked anyway by someone anyway, resulting in X amount lost

Schools, universities and workplaces (typically) won't crack them anyway, as they don't want to void warranty and it could make certain accreditations they hold at risk. They make most of their money in these markets anyway. Market size of hobbyists (Y) is a significant size smaller than the total market (Z). Additionally, there is a unique advantage of increasing sales within the Y market if they can hack it. So the total cost advantage gained of increased Y sales is probably larger than the potential of having reduced sales but people within the Y market purchasing bigger models.   

[firewalker]

Why would you buy a 1000Z instead of this?

Educational equipment? I must make a large oscilloscope order for a school lab. I don;t know-believe the DHO800 is suitable.

Alexander.

[dreamcat4]

Btw I wonder what is the mean time between failures for that USB-C connector..

It looks like the mounting pins go right through the PCB so it's probably a lot longer than you think.

If you're paranoid you can attach one of these to it and replace as needed:


Bonus: The cable comes out sideways so you gain some bench space.

I would also add to what Fungus suggested: you put the right angle dongle onto the Rigol scope. + Plus then glue it in place with a reversible silastic too (ie a medium durometer type that is removable, or similar like a poor hot glue that comes away). This will add some extra support and durablity.

[switchabl]

A number of people on this thread have said similar things. I don't really understand why but maybe I am missing something?

I can see that with adb root you can tinker with the Android system and sideload your own apps. But when it comes to the parts that actually matter, the complicated architecture (Zynq <-> PCIe <-> SoC <-> RAM) and the undocumented ADC and front end ICs are surely going to take a very broad skillset and a lot of determination to reverse engineer. Especially compared to something like the entry level Siglents, where everything is just off-the-shelf parts connected to a single Zynq and a good deal of work has already been done (https://github.com/360nosc0pe/yocto)?

Sorry, but this project is at 0.01% of all work needed (and it is based on Yocto which sucks badly). Oscilloscope firmware with a decent number of features is by far the most complicated piece of firmware that runs in test equipment. You are talking many years worth of work (fulltime!).

I am well aware of what it takes and I don't see this project going anywhere in terms of a full oscilloscope implementation. I was just referring to the reverse engineering part. They seem to have the ADCs and basic peripherals working at least (it's basically the cheapest Zynq + GSa/s ADC dev board you can buy now). I don't really expect anyone to achieve even that with the new Rigol.

[EEVblog]

Now in the sub $400 price bracket that other manufacturers will be forced to follow:
12bit
Touch screen
HDMI
VESA mount
USB-C (arguable downside for some, big feature for others)

I'd call that game changer. As always YMMV.

Yes if you look at it like that it makes sense..
Except...
Did Keysight (or R&S or Tektronix) made any reply in that segment lately?
They didn't. They don't care for sub 500USD scopes market.

They never have and they never will. And there is nothign wrong with that. Rigol and then Siglent, and a few other bit player have always had the really low end market to themselves. Nothings changed.

I do know Siglent has response to DHO1000 series by Rigol.
As for 800/900 it remains to be seen.

Rigol have the advntage wiht their custom ASIC's. It's likely Siglent can't price match that.

Race to the bottom is not always the best answer.

It's not, but it certainly gives customers a decent low priced option.

But I'm not saying it would not be nice that it does start a fierce competition, with awesome hardware coming from all sides for more affordable prices...
It's just I won't hold my breath.

Rigol may very well have the market to themselves.
The other makers are not going to cut their own throats. Siglent for example never matched the Rigol prices, Rigol has always been slightly lower priced.
Rigol are doing this because they have long term invested in the ASIC's that enable the lower pricing at a decent profit margin. You can bet you bottom dollar that Rigol are not doing this at a loss.
And the others may not be able to match it, they might have to stock to the "sure they have 12bit, but we offer 8bit and all this and that"
The marketing ferris wheel keep going around.

[EEVblog]

What I do not understand is why these manufacturers do not lock down their products so the software cannot be hacked, so people can not simply modify the software and they suddenly have the faster model.

Basically, they need to weigh up the cost X to encrypt an upgrade system that will be abused by Y amount within the total of their market Z that will either:

• Reduce uptake by hobbyists by Y units
• Be cracked anyway by someone anyway, resulting in X amount lost

Schools, universities and workplaces (typically) won't crack them anyway, as they don't want to void warranty and it could make certain accreditations they hold at risk. They make most of their money in these markets anyway. Market size of hobbyists (Y) is a significant size smaller than the total market (Z). Additionally, there is a unique advantage of increasing sales within the Y market if they can hack it. So the total cost advantage gained of increased Y sales is probably larger than the potential of having reduced sales but people within the Y market purchasing bigger models.

Yes, that's it.
It's a classic marketing ploy. The more people talk about your product, ultimately the more you are going to sell and the more visible it will be in the minds of everyone.
And the ones that talk the most on forums, blogs, and in Youtube videos etc are the keen enthusiasts who like to hack and get a bargain. It's not Joe Boring Day Job Engineer at company X.

[EEVblog]

I've got a couple of suitable Delta fans on order, I'll see if they make a difference. Once is much thicker and should barely fit.

[tv84]

What I do not understand is why these manufacturers do not lock down their products so the software cannot be hacked, so people can not simply modify the software and they suddenly have the faster model.

I had promised myself but.... arghhhh.... 

1. Who is the scope manufacturer that has successfully locked its scopes?

2. Developing a good secure system takes lots of money, skill and time. Security is expensive. So, you have to think about tradeoffs... At which level does one place his "lock bar"?

3. My reverse skills have not reached the final answer but I tend to believe (conspiracy spoiler alert) that, in fact, explicitly or not, part of all of this is inside Rigol commercial/expansion tactics.

[Nikki Smith]

I  teardown DHO914S The point is on the AFG module, if it is possible to copy it and install it to DHO900 or DHO800?
AFG core chip is AD9744.

Wow! that module is so much more complex than I was hoping. Thank you for taking the trouble to photograph it for us 

I guess that cloning that board would be a LOT of work, but are there any expensive components? The most expensive I can identify is the AD9744 14-bit DAC which is £20 from Mouser, US$6.68 from LCSC (and US$4.42 as a QFP).

FWIW the cost between the cheapest DHO804 and the cheapest DHO914S is €360 or £310 (Rigol EU, inc 20% VAT). The 914S does include hardware support for a logic analyser as well as the AFG module, plus the software-locked bandwidth increase.

[souldevelop]

I  teardown DHO914S The point is on the AFG module, if it is possible to copy it and install it to DHO900 or DHO800?
AFG core chip is AD9744.

Wow! that module is so much more complex than I was hoping. Thank you for taking the trouble to photograph it for us 

I guess that cloning that board would be a LOT of work, but are there any expensive components? The most expensive I can identify is the AD9744 14-bit DAC which is £20 from Mouser, US$6.68 from LCSC (and US$4.42 as a QFP).

FWIW the cost between the cheapest DHO804 and the cheapest DHO914S is €360 or £310 (Rigol EU, inc 20% VAT). The 914S does include hardware support for a logic analyser as well as the AFG module, plus the software-locked bandwidth increase.

I thought the same way you did, but seeing such a complex composition I now seem to have lost my cloning value.

[faveri97]

Hello,

I have a suspicion that the Rigol 12 Bit Scopes are processing the data if possible. This can be seen from the fact that the wave files do not contain 12 but already 16 bit values.
That's why you should make sure that the highest sampling rate is forced when measuring, so that hopefully you get the raw data.
Unfortunately, my request for data has not yet been successful:

"Can someone please post two wavefile in bin format with the following settings:
1 mV/div 0.05 ms/div 1 Mpts and
1 V/div 0.05 ms/div 1 Mpts.
And accompanying screenshots.
Each with an open input and full bandwidth.

It is important that 1.25 GSa/s is used so that the device cannot sugarcoat the data.

If it's not too much work, corresponding files with 20 MHz bandwidth would be very nice."

Best regards
egonotto

Rigol finally returned my DHO914S to me, and I captured the requested screenshots and data files.

[EEVblog]

I  teardown DHO914S The point is on the AFG module, if it is possible to copy it and install it to DHO900 or DHO800?
AFG core chip is AD9744.

Wow! that module is so much more complex than I was hoping. Thank you for taking the trouble to photograph it for us 

Yeah, bit more complex than I expected too.

[RAPo]

I tried to convert the bin to csv with the MSO5000 bintest program, no output. Does the DHO9145S need its own bintest?

Rigol finally returned my DHO914S to me, and I captured the requested screenshots and data files.

[Fungus]

1. Who is the scope manufacturer that has successfully locked its scopes?

R&S?

[Doctorandus_P]

I am in the process of buying an scope within a few weeks, and was wondering why the Siglent SDS1104X-E suddenly got a 10% discount this month (both at Eleshop, Welectron and Batronix) and this may be the reason. (Or maybe Siglent is planning to release an upgrade in a few months?)

12 bit is a significant upgrade compared to 8-bit, but everything else seems to be so much "cost optimized" that I only briefly considered this Rigol scope. And also, when I look at the Batronix prices, It's EUR594 for the Rigol compared to EUR 460 for the Siglent (Both 4 channels 100MHz versions).

This Rigol would need a bunch of very serious reviews before I would even consider buying it.
One thing I really dislike on Rigol is their measurements are done from on-screen data. That is really crippling your 12-bit ADC. How good is FFT on this 12-bitter compared to Siglent? Other functions, wave capture rate, measurement functions etc. How much are those being hampered by the "cost optimization" efforts? (To be fair, I'm not sure if the Siglent uses real data or screen pictures for the on-screen measurements, because the Siglent SDS1104X-E is near the top of what I am willing to spend on a scope, and I would have to make do with whatever it delivers anyway. (Siglent / Keysight / R&S are out of my range).

And things like the first products being shipped wit such a crappy power supply (See start of this thread) does not give much confidence. If this goes on then the Rigols would be just as bad as the fnirsi crap in just a few more iterations.