R&S to launch MXO5

[Martin72]

Only one year warranty...
Ah, you can buy additional warranty time, clever.

[chriswebb]

I can’t help but think these new R&S scopes are underbaked and being rushed to market.

[TheUnnamedNewbie]

Still waiting for them to release something fast for baseband analysis in high-speed serdes design. They promised me one 5 years ago, and I'm still waiting!

[pdenisowski]

I can’t help but think these new R&S scopes are underbaked and being rushed to market.

Well, we spent over 4 years developing our own ASIC for the MXO4, which now has the fastest waveform update rate in the world (plus the deepest standard memory, our patented digital trigger, etc. etc. etc.). 

Since the cat is already halfway out of the bag (and will be all the way out of the bag in about 8 more hours), the MXO5 leverages a lot of those years of R&D we put into the MXO4.  And work on the MXO5 began long before the MXO4 was released last year.

No test and measurement instrument is brought to market with every feature and option available on the first day.  Sure, we'd like to do that, but the reality is that we have plenty of customers who buy our scopes (and spec ans, sig gens, VNAs, power sensors, etc.) and use them successfully with the features and options that are available today. 

I spent over 20 years as a field applications engineer working with T&M customers, and I can assure you that almost every customer has something they'd like us to add to even our most mature products.

So if we're releasing "underbaked" products, I'm not sure what else we could be doing - but we're always looking for feedback!

[Martin72]

Cheaper or not, that is the question..
First I thought yes it must be, because of 100Mhz bandwith - But this is a eight-channel version...

[tv84]

Cheaper or not, that is the question..
First I thought yes it must be, because of 100Mhz bandwith - But this is a eight-channel version...

Martin, something like this will never be cheap (even if it was a 50MHz with 2-ch).

[chriswebb]

I can’t help but think these new R&S scopes are underbaked and being rushed to market.

Well, we spent over 4 years developing our own ASIC for the MXO4, which now has the fastest waveform update rate in the world (plus the deepest standard memory, our patented digital trigger, etc. etc. etc.). 

Since the cat is already halfway out of the bag (and will be all the way out of the bag in about 8 more hours), the MXO5 leverages a lot of those years of R&D we put into the MXO4.  And work on the MXO5 began long before the MXO4 was released last year.

No test and measurement instrument is brought to market with every feature and option available on the first day.  Sure, we'd like to do that, but the reality is that we have plenty of customers who buy our scopes (and spec ans, sig gens, VNAs, power sensors, etc.) and use them successfully with the features and options that are available today. 

I spent over 20 years as a field applications engineer working with T&M customers, and I can assure you that almost every customer has something they'd like us to add to even our most mature products.

So if we're releasing "underbaked" products, I'm not sure what else we could be doing - but we're always looking for feedback!

I am in the market for a 2GHz scope at the moment and I was hoping your MXO 5 would fit the bill, but I cannot hope that one day you will implement the features that I need. I have to make the decision on the features you have delivered. And sure I could probably buy your scope and get the features i need from other equipment, but why would I do that when I can get everything I require in single packages elsewhere? I dont need dedicated equipment for features I may only need occasionally.

I would also like to add that what you said does not necessarily contradict my original point, i.e. you are now trying to rush to market to compete with cheaper brands’ new scopes to recoup your R&D costs of your new hardware while in my eyes at least not providing a more complete software package. I really like your new hardware, and I really was hoping that this would be my new scope.

[Martin72]

Martin, something like this will never be cheap (even if it was a 50MHz with 2-ch).

Cheaper, not cheap.
But I quickly realized my mistake and so such a scope remains just a dream.
Theoretically.

[pdenisowski]

I am in the market for a 2GHz scope at the moment and I was hoping your MXO 5 would fit the bill, but I cannot hope that one day you will implement the features that I need. I have to make the decision on the features you have delivered. And sure I could probably buy your scope and get the features i need from other equipment, but why would I do that when I can get everything I require in single packages elsewhere? I dont need dedicated equipment for features I may only need occasionally.

Completely fair.  Are there specific features that you're looking for that we don't have on the MXO5? 

I would also like to add that what you said does not necessarily contradict my original point, i.e. you are now trying to rush to market to compete with cheaper brands’ new scopes to recoup your R&D costs of your new hardware while in my eyes at least not providing a more complete software package. I really like your new hardware, and I really was hoping that this would be my new scope.

I can absolutely, positively, 100% assure you that we aren't "rushing" anything to market to compete with "cheaper brands' new scopes" 

(and by "cheaper" I assume by this you mean Siglent, Rigol, Owon, etc.).

As a hobbyist who's bought both analog and digital scopes for his own personal use with his own money (I can post pictures ), I'm honestly very impressed by how far some of these "cheaper brands' new scopes" have come in the last decade or so.  But that's not the market that we are addressing the the MXO4 and MXO5.

Yes, it'd be great of someone who was about to buy a new Rigol DHO80x said: "You know, I might just spend the extra money and get a R&S MXO", but I get the feeling that doesn't happen very often.  (Although if you are such a person, please PM me!)

But I do appreciate the feedback and I'm very sincere when I say that we are always interested in hearing which features / capabilities are most needed by our (potential) customers, so please do let us know!

[Martin72]

I can absolutely, positively, 100% assure you that we aren't "rushing" anything to market to compete with "cheaper brands' new scopes"

Meanwhile, you can spend 20000 or more for a siglent or rigol scope, the models are there.
For the hobbyist with a somewhat higher budget, it would be interesting to know whether a successor to the RTB2000 is planned or in the starting blocks.
Edit:

"You know, I might just spend the extra money and get a R&S MXO"

At that time, I had resolved never to spend more than 2500€ on a scope.
Then came the SDS2000XHD and I put 3600 for it, that was already hard on the edge.
In the meantime I would go further if I could see a clear improvement in it, but....

https://www.batronix.com/versand/oszilloskope/Rohde-Schwarz-MXO44-242.html

No Sir...

[chriswebb]

Yes, it'd be great of someone who was about to buy a new Rigol DHO80x said: "You know, I might just spend the extra money and get a R&S MXO", but I get the feeling that doesn't happen very often.  (Although if you are such a person, please PM me!)

I definitely recognize that lurkers like me are an impossibly small market. I can't even get Keysight to directly sell me used equipment with only 90 day warranties just sitting in inventory for instance (edit: one of the big reasons for waiting for the MXO 5). I can definitely appreciate that you can't expect to make money and cater to such small numbers, and honestly when I originally posted I never expected anyone from R&S to actually see it let alone respond to it. I appreciate your taking the time to respond here. I really hope this launch goes well for R&S. I look forward to seeing where you guys go with these products.

[Stranger_danger]

I can’t help but think these new R&S scopes are underbaked and being rushed to market.

Well, we spent over 4 years developing our own ASIC for the MXO4, which now has the fastest waveform update rate in the world (plus the deepest standard memory, our patented digital trigger, etc. etc. etc.). 

Since the cat is already halfway out of the bag (and will be all the way out of the bag in about 8 more hours), the MXO5 leverages a lot of those years of R&D we put into the MXO4.  And work on the MXO5 began long before the MXO4 was released last year.

Since the cat is out of the bag, can you elaborate on the MXO5 ASIC?  You make the comment of spending 4 years to build your own for the MXO4 and starting this one well before that.

The MXO5 datasheet says it's 2.5GS/s and 12-bit ADC and 18-bit architecture in HD mode. Is this a R&S custom made ASIC/ADC chipset as well? and how does 18-bit storage work?

[pdenisowski]

when I originally posted I never expected anyone from R&S to actually see it let alone respond to it. I appreciate your taking the time to respond here. I really hope this launch goes well for R&S. I look forward to seeing where you guys go with these products.

Thanks!!!  And there are quite a few R&S folks -- including some of our oscilloscope design team members -- who are here on the EEVblog forum as well:  I'm just one of the few who posts as an R&S employee using my real name

And please feel free to reach out any time.  We honestly do appreciate all of the feedback and comments we get!

[pdenisowski]

Since the cat is out of the bag, can you elaborate on the MXO5 ASIC?  You make the comment of spending 4 years to build your own for the MXO4 and starting this one well before that.

The MXO5 datasheet says it's 2.5GS/s and 12-bit ADC and 18-bit architecture in HD mode. Is this a R&S custom made ASIC/ADC chipset as well? and how does 18-bit storage work?

We actually started the brainstorming for the MXO4 back in 2016 (!!!) and started active development in 2018. 

And yes, the ASIC in the MXO is our custom-designed MXO-EP (200 Gbit/s).  Among other things, this ASIC enables an update rate of up to 4.5 million acquisitions/s, which is the world's fastest update rate, as tested by Dave Jones himself.  Dave also did a teardown video if you're interested:

https://www.eevblog.com/forum/blog/eevblog-1545-worlds-fastest-oscilloscope-mxo4-teardown/

The MXO can use something called high-definition mode where bandwidth is traded off for bits of vertical resolution (lower BW -> more bits).  I actually made a video that provides a brief technical introduction HD mode and how it increases vertical resolution (compared to other methods). 



Not sure how much more I can say before the official launch (4 hours from now)

[Martin72]

Not sure how much more I can say before the official launch (4 hours from now)

It has something of the Apple events where they present the latest iPhone to their fans.
Wow, you can also use it to make phone calls...Awesome, what's next...?
It´s just another scope at the end.

[jjoonathan]

It's ok to not have hype, but it's ok to have hype too 

Those fast FFTs are pretty slick. Maybe even something about differential channels or a motor app?

Given the sample rates, I suspect the ADC is the same TI chip from the MXO4. On one hand, we all like to hear about crazy custom ADCs, but on the other, for hobby purposes it's nice to have obtanium inside. If a zillion channels becomes the standard, I have a sneaking suspicion that single-channel fails will land powerful scopes onto the benches of enthusiasts, so it's good for us if those are repairable 

Also, when I took a second look at the datasheet today I saw something I didn't see last time, namely the background calibration routine (7-4). Apparently it swaps redundant ADC cores in and out to achieve on-the-fly background calibration. Very cool!

[tv84]

Theoretically.

tv84: Stay away from the light!!!

Martin72: Next Xmas, maybe I... 

[nctnico]

Since the cat is out of the bag, can you elaborate on the MXO5 ASIC?  You make the comment of spending 4 years to build your own for the MXO4 and starting this one well before that.

The MXO5 datasheet says it's 2.5GS/s and 12-bit ADC and 18-bit architecture in HD mode. Is this a R&S custom made ASIC/ADC chipset as well? and how does 18-bit storage work?

We actually started the brainstorming for the MXO4 back in 2016 (!!!) and started active development in 2018. 

And yes, the ASIC in the MXO is our custom-designed MXO-EP (200 Gbit/s).  Among other things, this ASIC enables an update rate of up to 4.5 million acquisitions/s, which is the world's fastest update rate, as tested by Dave Jones himself.  Dave also did a teardown video if you're interested:

https://www.eevblog.com/forum/blog/eevblog-1545-worlds-fastest-oscilloscope-mxo4-teardown/

The MXO can use something called high-definition mode where bandwidth is traded off for bits of vertical resolution (lower BW -> more bits).  I actually made a video that provides a brief technical introduction HD mode and how it increases vertical resolution (compared to other methods). 

What would be super nice is a mode where you can have a fixed bandwidth setting. Recently I used my Yokogawa 12 bit scope to measure tiny signals (in the 50uV/div territory) and what helps a lot is being able to filter down to a couple of hundred Hz using a fixed setting. Now this Yokogawa design is 20 years old so nowhere near as low noise as modern oscilloscopes so it should be possible to achieve similar performance using higher bandwidths.

One of the things that annoys me when using high resolution modes is that you never really know with what bandwidth you are looking at a signal.

[2N3055]

Since the cat is out of the bag, can you elaborate on the MXO5 ASIC?  You make the comment of spending 4 years to build your own for the MXO4 and starting this one well before that.

The MXO5 datasheet says it's 2.5GS/s and 12-bit ADC and 18-bit architecture in HD mode. Is this a R&S custom made ASIC/ADC chipset as well? and how does 18-bit storage work?

We actually started the brainstorming for the MXO4 back in 2016 (!!!) and started active development in 2018. 

And yes, the ASIC in the MXO is our custom-designed MXO-EP (200 Gbit/s).  Among other things, this ASIC enables an update rate of up to 4.5 million acquisitions/s, which is the world's fastest update rate, as tested by Dave Jones himself.  Dave also did a teardown video if you're interested:

https://www.eevblog.com/forum/blog/eevblog-1545-worlds-fastest-oscilloscope-mxo4-teardown/

The MXO can use something called high-definition mode where bandwidth is traded off for bits of vertical resolution (lower BW -> more bits).  I actually made a video that provides a brief technical introduction HD mode and how it increases vertical resolution (compared to other methods). 



Not sure how much more I can say before the official launch (4 hours from now)

That HiDef mode is equivalent to ERES in LeCroy and Siglent and HiRes by other manufacturers.
R&S, like LeCroy and Siglent, provide pretty fine control of the process, which makes it useful.

I think 6bit "enhancement" (12->18bit) is bit of a stretch, but it is useful as a lowpass filter anyways..

[luudee]

Very exciting !

Can't wait for actual configurations & pricing and availability date !

I'm still in the market for an upgrade ...

EDIT: Found the answers I was looking for:  https://www.rohde-schwarz.com/th/products/test-and-measurement/oscilloscopes/rs-mxo-5-oscilloscope_334228.html

Cheers,
rudi

[Martin72]

Theoretically.

tv84: Stay away from the light!!!

Martin72: Next Xmas, maybe I... 

Starting price MX05: 17000€....
No my friend, that is worlds away from what I can spend, that would be at most something for the company.
The MX04 on the other hand is "significantly cheaper", but that would only come into question if I won the lottery.
Nope, I'll probably stay loyal to Siglent.

[pdenisowski]

That HiDef mode is equivalent to ERES in LeCroy and Siglent and HiRes by other manufacturers.
R&S, like LeCroy and Siglent, provide pretty fine control of the process, which makes it useful.

The biggest advantage of HD mode over other methods of improving vertical resolution is that you have explicit control over the bandwidth.  We do have a "high resolution" mode on some of our other scopes, but that mode doesn't allow you to specify the bandwidth.  I'll admit that I don't know enough technical details about other vendors' implementations to (reliably / credibly) compare and contrast.

I think 6bit "enhancement" (12->18bit) is bit of a stretch, but it is useful as a lowpass filter anyways..

Depending on what you're looking at, the improvement in resolution from HD mode can be anywhere between (OMG!) and almost unnoticeable.  That said, once you get used to having HD mode as an option, it's really hard to go back:  as time goes on I find myself using it more and more often.

(It's also a standard feature, so it doesn't cost extra)

[2N3055]

That HiDef mode is equivalent to ERES in LeCroy and Siglent and HiRes by other manufacturers.
R&S, like LeCroy and Siglent, provide pretty fine control of the process, which makes it useful.

The biggest advantage of HD mode over other methods of improving vertical resolution is that you have explicit control over the bandwidth.  We do have a "high resolution" mode on some of our other scopes, but that mode doesn't allow you to specify the bandwidth.  I'll admit that I don't know enough technical details about other vendors' implementations to (reliably / credibly) compare and contrast.

I think 6bit "enhancement" (12->18bit) is bit of a stretch, but it is useful as a lowpass filter anyways..

Depending on what you're looking at, the improvement in resolution from HD mode can be anywhere between (OMG!) and almost unnoticeable.  That said, once you get used to having HD mode as an option, it's really hard to go back:  as time goes on I find myself using it more and more often.

(It's also a standard feature, so it doesn't cost extra)

ERES is pretty much same (as it is implemented in hardware in LeCroy and Siglent, or you have one in Math for postprocessing at will).
I would have to see how it is implemented in MXO5 but ERES in Siglent for instance does not show BW reduction on screen. It is easy to calculate though, but I would still like it to be there in plain sight. I like when scope calculates and plainly shows those things for us, wherever is possible...

[Martin72]

I like when scope calculates and plainly shows those things for us, wherever is possible

LeCroys MAUI scopes do this(showing bandwith when using Eres)..
Could be something for the (siglent)wish list.

[nctnico]

That HiDef mode is equivalent to ERES in LeCroy and Siglent and HiRes by other manufacturers.
R&S, like LeCroy and Siglent, provide pretty fine control of the process, which makes it useful.

The biggest advantage of HD mode over other methods of improving vertical resolution is that you have explicit control over the bandwidth.  We do have a "high resolution" mode on some of our other scopes, but that mode doesn't allow you to specify the bandwidth.  I'll admit that I don't know enough technical details about other vendors' implementations to (reliably / credibly) compare and contrast.

Aha, so it does allow setting the bandwidth!  IMHO that is a major differentiation from eres / highres modes and makes it much more usefull. It would be nice if the marketing department could come up with a better term than HD mode. Even without giving additional bits, it will be highly useful for sure.
What is the lowest bandwidth you can set for 'HD mode'?