DSO scopes with a good interface?

[tautech]

I find that most companies don't do a good job of listening unless you are their bread and butter...

Unless you have a dialogue with someone that has good contact with a manufacturer, well what do you expect ? 

Seriously, tell us what you want/need in a DSO UI and there will be a DSO that's very close to your requirements. 

[jasonbrent]

I find that most companies don't do a good job of listening unless you are their bread and butter...

Unless you have a dialogue with someone that has good contact with a manufacturer, well what do you expect ? 

Seriously, tell us what you want/need in a DSO UI and there will be a DSO that's very close to your requirements. 

tautech, I wish I could in quantitative ways... my intention here was to have a dialogue based on some observations of things I've seen painful in a singular device. Over my years I've come to realize that knowledge can be gained the hard way through time/trial/experience (and that is the best form of KNOWledge), but it can also come from the advice and experience of others. In this scenario, I simply desired inputs from others on their definition of "good interface".

I'm a noob here that's gone from an old analogue scope and a couple of cheap handheld DMMs to a reasonably well outfitted "lab" in a short period of time. My equipment greatly exceeds my knowledge and experience.... Working on a repair, I realized quickly that the Rigol interface was more about rote memorization than intuitive interactions.

* Little things, like the turning of the knob to control selection when multiple items are presented on the screen is completely backwards from my intuition; my brain says "clockwise should go down", but instead  the selection moves up.
* Selecting measurements using the left most buttons requires going to the "Measure" menu where you can get rid of measurements visually by going to the 3rd menu down, choosing "select item", then toggling them off,
* However, as soon as you add another one, the one you toggled off re-appears....
* To actually remove it you have to go to Measure->Clear->Delete.. but that doesn't really delete it, it just greys it out on the screen(!@!@#!). ]
* Choosing "Delete all", actually deletes them all... until you add another measurement, then 4 greyed out ones appear and the one you added appears on the right...
* .. So if your first 3 measurements were channel 2, you no longer need channel 2,  you delete them, turn off channel 2 and add a channel 1 measurement, 3 greyed out measurements appear followed by channel 1.

My intention isn't to just complain... functionally it's a great unit for the price point. I just wonder if there is another price point where these front-and-center niggles don't exist.

With any bit of technology, my preferred interface is simply one that doesn't get in the way of stream of consciousness.

</rambling....>

-j

[tautech]

tautech, I wish I could in quantitative ways...

And you just have ! 

I'm a noob here that's gone from an old analogue scope and a couple of cheap handheld DMMs to a reasonably well outfitted "lab" in a short period of time. My equipment greatly exceeds my knowledge and experience.... Working on a repair, I realized quickly that the Rigol interface was more about rote memorization than intuitive interactions.

* Little things, like the turning of the knob to control selection when multiple items are presented on the screen is completely backwards from my intuition; my brain says "clockwise should go down", but instead  the selection moves up.
* Selecting measurements using the left most buttons requires going to the "Measure" menu where you can get rid of measurements visually by going to the 3rd menu down, choosing "select item", then toggling them off,
* However, as soon as you add another one, the one you toggled off re-appears....
* To actually remove it you have to go to Measure->Clear->Delete.. but that doesn't really delete it, it just greys it out on the screen(!@!@#!). ]
* Choosing "Delete all", actually deletes them all... until you add another measurement, then 4 greyed out ones appear and the one you added appears on the right...
* .. So if your first 3 measurements were channel 2, you no longer need channel 2,  you delete them, turn off channel 2 and add a channel 1 measurement, 3 greyed out measurements appear followed by channel 1.

This is just basic UI poor design that I agree would drive me mad too. 

I've been very lucky to have had the 'ear' of some in Siglent where I've (and others) been able have changes implemented to equipment UI. This forum has been great to be able to bounce ideas about and come up with some very satisfying improvements to their range of gear.
Little things like what you mentioned about encoder's rotation not being intuitive .....when I was lucky enough to be asked to be a beta tester for the new 4ch X-E, the encoder was just that; backward  (for some operations, not all) AND at odds with all other recent Siglent DSO's ! Despite feedback from product managers it was pretty easy to show them of the inconsistency's across their range.

Having features/functions of a similar type bunched together in one main menu and then the finer tweaks of the same type within a sub-menu makes for an intuitive experience not a frustrating one especially for those coming from a CRO, more so a real analog one where all controls are physical and on the front panel.
That's where I've come from and that past experience helps guide one as to what to hope to expect from a DSO.

Keep looking and keep an open mind as not all DSO's offer the experience you've had.
As I said earlier, you don't have to spend a fortune to get a 'easy to use and good' DSO.

[jasonbrent]

@tautech -- ha, that started out non-qualitative, but I apparently re-edited enough times to tease some points out and failed to update that line. oops.

A portion of "day job" is, for a swath of my business, to ensure that it's "supportable". There are a lot of "-ility" words used (supportability, reliability, diagnosability, make-up-word-able-ility, etc.) but it basically means I look for things in our processes, our products, our people interactions that aren't intuitive, usable, consistent across product lines and get them fixed.... which means "niggly things can die in a fire." is a general self motto. 

I looked hard at the latest Siglent before picking up this DS1054Z, but a few years back I picked up a siglent (don't remember the model) and it DOAd within a week. I replaced it with a DS2072A that ultimately went to a friend as I put this hobby on hold. The experience with that one siglent made me gun-shy about their brand new product. The fact that they do listen to inputs from ${resellers, customers, etc.} is valuable knowledge.

I'm not stuck on a particular name brand in general, however my expectations and criticisms of an Agilent/Keysight/HP/Tek/R&S would be greatly higher than "WanHungLo". I would rather pay $120 for something that will last 6 years than $20 for something that will last 1 year though (numbers/time at extreme just to convey the point..).

... Useless bit of knowledge; in history past, I worked around a large american retailer^Wswindler beginning with wal and ending in mart. I became aware of their supplier purchase processes at that point in time... end result is for the consumer wanting to purchase a given product, let's call it PRODUCTX, if they bought it at this retailer it would be cheaper than another retailer and have the same product identifications.... However on the inside, PRODUCTX for this retailer would be made of lower cost parts that had a decreased reliability (plastic cogs instead of metal cogs, for example)...

I appreciate the discourse, time and insights!

-j

[tautech]

I looked hard at the latest Siglent before picking up this DS1054Z, but a few years back I picked up a siglent (don't remember the model) and it DOAd within a week. I replaced it with a DS2072A that ultimately went to a friend as I put this hobby on hold. The experience with that one siglent made me gun-shy about their brand new product. The fact that they do listen to inputs from ${resellers, customers, etc.} is valuable knowledge.

If you've been here a while you'd have seen even the 'A' brands have had their share of troubles.

Yep, and such is the speed at which this industry moves Siglent have had to come up to speed to ensure their brand is a viable choice.
Just as they've been flogged online for past POS, if you only knew what was going on privately !

There has been and are models in their range I will not stock now and into the future. Their good models give me and customers just no grief and of late any one I back publicly is due to good experiences, both customer's and mine.

How your supplier handles any issues you may have is really the key to a good brand experience and as so many seek the cheapest source (often overseas) so support quality is mostly sub-optimal.

I'll leave you be now to stew on all the feedback.

[Fungus]

Huh? How often do you switch 1x/10x? It's usually about once a year, if ever.

If that's the case then you're not using a scope very much or just doing similar simple and menial tasks !

10x, most common normal probe usage
1x, checking PSU rails for ripple, also very common probe use case.
1x, BNC connection, eg.  monitor AWG output while with another channel checking DUT amplitude/signal integrity/phase shift.

There's 4 connectors on the front of mine. It's not hard to designate channel 4 as "The Coax Channel", etc.

Often you want the 20MHz cutoff filter with the 1x probe setting. That's two things to change every time... unless you have a "1x channel".

[EE-digger]

I detest 1x/10x probes, primarily because the ones we have around are Rigol and the switch is cr-p.  Usually go to very small coax with whatever end connection is needed.

An associate recently lost several days of very expensive data collecting on a large system because the switch was flakey.  This didn't show until massive post processing showed abnormalities in the data.  Turned out to be the $^%& Rigol probe.

Scope should have been Agilent and the probes should have been Agilent with USA, JAPAN or GERMANY on the probes.  We had 3 of the 4 probes on a new MSOX3104A die in 3 months !  They were handled gently at all times and the scope was never suspended by any of them    These probes all said CHINA.

Agilent wanted the scope returned before they would replace the failed probes.  I had a battle with the support person, then their manager, who after 3 days of meditation, finally agreed to replace the probes only.

Sorry for the digression but it was probe related.

[Fungus]

Yep. All 'scopes should supply fixed 10x probes for primary use IMHO. A lot of problems could be avoided that way.

With multichannel scopes a 50:50 mix of fixed/switchable probes would be perfect.

(all we need is for newbies to know that the switchable ones are worse )

[nctnico]

I beg to differ. I like the GW Instek UI much better than the Agilent DSO7104A I had. Doing something simple like changing the probe multiplication factor and unit is real chore on the DSO7104A. The GW Instek OTOH even has a dedicated 10X button because that is the most used setting.

Huh? How often do you switch 1x/10x? It's usually about once a year, if ever.

If you don't have coded probes that the scope understands then it can be something that needs to be adjusted quite often when swapping between 50 ohm coax connections and probes.

Not just that but think about switching the range on a differential probe or current probe. Sure probes which can 'talk' to the scope make life easier but these don't work on scopes from a different brand or on different equipment. For example: I'm using low-Z probes on both a spectrum analyser and oscilloscope. And there is way more equipment you can use a probe on.

[rsjsouza]

tautech, I wish I could in quantitative ways...

And you just have ! 

I'm a noob here that's gone from an old analogue scope and a couple of cheap handheld DMMs to a reasonably well outfitted "lab" in a short period of time. My equipment greatly exceeds my knowledge and experience.... Working on a repair, I realized quickly that the Rigol interface was more about rote memorization than intuitive interactions.

* Little things, like the turning of the knob to control selection when multiple items are presented on the screen is completely backwards from my intuition; my brain says "clockwise should go down", but instead  the selection moves up.
* Selecting measurements using the left most buttons requires going to the "Measure" menu where you can get rid of measurements visually by going to the 3rd menu down, choosing "select item", then toggling them off,
* However, as soon as you add another one, the one you toggled off re-appears....
* To actually remove it you have to go to Measure->Clear->Delete.. but that doesn't really delete it, it just greys it out on the screen(!@!@#!). ]
* Choosing "Delete all", actually deletes them all... until you add another measurement, then 4 greyed out ones appear and the one you added appears on the right...
* .. So if your first 3 measurements were channel 2, you no longer need channel 2,  you delete them, turn off channel 2 and add a channel 1 measurement, 3 greyed out measurements appear followed by channel 1.

This is just basic UI poor design that I agree would drive me mad too. 

Interesting notes about the DS1000Z interface, which I concur are pretty stupid and should be considered true bugs. Its bigger brother DS4000 does not have any of that (although it has its own quirks).

The encoder rotation reminds me when I used a Mac laptop for the first time, where the direction of scroll is inverted when compared to Windows and Unix hosts.

I also agree the 1:10 switching is important and deserves a button or at least an easily accessible menu to it - still drawing a comparison with my Mac laptop experience, several important keys are missing (del, pgup/pgdn, ins) and other functions require quite non-intuitive key combinations.

Several years ago I was lucky that my oscilloscope probes were 10:1 fixed, as they saved its front end a few times. However, not everything that comes from the far east is crap - the probes that came with the DS4000 are RP3500A and quite good quality.

[tautech]

tautech, I wish I could in quantitative ways...

And you just have ! 

I'm a noob here that's gone from an old analogue scope and a couple of cheap handheld DMMs to a reasonably well outfitted "lab" in a short period of time. My equipment greatly exceeds my knowledge and experience.... Working on a repair, I realized quickly that the Rigol interface was more about rote memorization than intuitive interactions.

* Little things, like the turning of the knob to control selection when multiple items are presented on the screen is completely backwards from my intuition; my brain says "clockwise should go down", but instead  the selection moves up.
* Selecting measurements using the left most buttons requires going to the "Measure" menu where you can get rid of measurements visually by going to the 3rd menu down, choosing "select item", then toggling them off,
* However, as soon as you add another one, the one you toggled off re-appears....
* To actually remove it you have to go to Measure->Clear->Delete.. but that doesn't really delete it, it just greys it out on the screen(!@!@#!). ]
* Choosing "Delete all", actually deletes them all... until you add another measurement, then 4 greyed out ones appear and the one you added appears on the right...
* .. So if your first 3 measurements were channel 2, you no longer need channel 2,  you delete them, turn off channel 2 and add a channel 1 measurement, 3 greyed out measurements appear followed by channel 1.

This is just basic UI poor design that I agree would drive me mad too. 

Interesting notes about the DS1000Z interface, which I concur are pretty stupid and should be considered true bugs. Its bigger brother DS4000 does not have any of that (although it has its own quirks).

The encoder rotation reminds me when I used a Mac laptop for the first time, where the direction of scroll is inverted when compared to Windows and Unix hosts.

I also agree the 1:10 switching is important and deserves a button or at least an easily accessible menu to it - still drawing a comparison with my Mac laptop experience, several important keys are missing (del, pgup/pgdn, ins) and other functions require quite non-intuitive key combinations.

Several years ago I was lucky that my oscilloscope probes were 10:1 fixed, as they saved its front end a few times. However, not everything that comes from the far east is crap - the probes that came with the DS4000 are RP3500A and quite good quality.

There's much about scope usage that is 'personal' taste or methodology and if you think about how manufacturers might cater for the vast array of user preferences, it leads to quite a complex scenario.
10x, 1x probes are the tip of the iceberg and for entry level DSO's they're perfectly fine providing the user understands the shortcomings of each.
Spend a bit more and get to $1-3k and 10x probes are often standard and with auto-attenuation.
1x usage is usually much less demanding and any of the cheap probes available online with serve 1x usage perfectly well.

We all have such varying requirements of scope needs so how does any manufacturer attempt to cater for that ?
In the upper level DSO's functionality to set users preferences aren't uncommon but in the entry level marketplace there's little that supported this feature until the X-E's came out last year.
Now we can set the scope to our favorite settings, trace pos, V/div, timebase, input attenuation, measurements, trigger and pretty much all that you like to have as baseline settings when using a scope. User definable Default or with a couple of menu button presses, back to Factory Default.

[rsjsouza]

Exactly. Manufacturers will never again be off the hook when UI is involved. As more powerful processors become available, the customization demands become higher and customers will make their budget decisions based on that - especially when other products such as cellphones and tablets skew this perception.

[tautech]

Exactly. Manufacturers will never again be off the hook when UI is involved. As more powerful processors become available, the customization demands become higher and customers will make their budget decisions based on that - especially when other products such as cellphones and tablets skew this perception.

Funny you say that...........I've not long had W10 and while it's not hard to get used to after W7, the thing that really frustrated me the most was Office Picture Manager was dropped in later Office versions. 
WTF where they thinking ?
Actually after a little research and a few free minutes, I found there's a free MS Share Point Designer download and in which you can turn OFF all other parts and just have the Picture Manager.
Installed ! 

[Muttley Snickers]

We all have such varying requirements of scope needs so how does any manufacturer attempt to cater for that ?

They could simply ask what we want and a finished product would be a damn good start, for stand alone usage I still require a real time clock for basic event recording and mouse support should be a standard feature surely, particularly with all the numerous drop down lists now on various scopes.

[nctnico]

We all have such varying requirements of scope needs so how does any manufacturer attempt to cater for that ?

They could simply ask what we want and a finished product would be a damn good start, for stand alone usage I still require a real time clock for basic event recording and mouse support should be a standard feature surely, particularly with all the numerous drop down lists now on various scopes.

My forecast is that in a few years all oscilloscopes will have touch screens. In today's market touchscreen interfaces on oscilloscopes are still a bit of a hit &miss but it will improve.

[Muttley Snickers]

In my case a touch screen would be completely useless as most of my test equipment is set back on a deep bench so reaching over is almost out of the question, hence my need for a mouse.

[rsjsouza]

We all have such varying requirements of scope needs so how does any manufacturer attempt to cater for that ?

They could simply ask what we want and a finished product would be a damn good start, for stand alone usage I still require a real time clock for basic event recording and mouse support should be a standard feature surely, particularly with all the numerous drop down lists now on various scopes.

My forecast is that in a few years all oscilloscopes will have touch screens. In today's market touchscreen interfaces on oscilloscopes are still a bit of a hit &miss but it will improve.

Muttley, the "simply ask" is already a tall order: opinions will vary greatly among the customer base. I, for one, don't care at all for a mouse (no space on the bench, always loved a keypress instead of a virtual one... ) Obviously that, if processing power is enough, by all means make it an option. If not, then what would you sacrifice to implement that? Opinions would vary.

Nico, that is a great point where even the A brands are still struggling to meet the UI and responsiveness we are used in our tablets and cellphones.

[jasonbrent]

As a point of interest, Keysight doesn't seem to consider Rigol (at least the ones I have) a valid trade in for their 30% off the 3000T series unit.. at least not without an exception (which I'm waiting on). Too bad they don't actually tell you that in their T&Cs before potentially wasting mine and their time.

-j

[GlowingGhoul]

I'll just amplify what other's have already said. Keysight is the king of the UI. The responsiveness will spoil you for other scopes.

In the $3k range, the 3000T is going to give you the nicest experience because of the excellent touch interface.

[nctnico]

I'll just amplify what other's have already said. Keysight is the king of the UI. The responsiveness will spoil you for other scopes.

No wonder. Everyone can make a scope which scrolls 1Mpts left/right. Try and do the same with 10Mpts or more!

[rsjsouza]

I'll just amplify what other's have already said. Keysight is the king of the UI. The responsiveness will spoil you for other scopes.

No wonder. Everyone can make a scope which scrolls 1Mpts left/right. Try and do the same with 10Mpts or more!

That is true, although other responsiveness "treats" go in favour of Keysights.

[Lukas]

I'll just amplify what other's have already said. Keysight is the king of the UI. The responsiveness will spoil you for other scopes.

No wonder. Everyone can make a scope which scrolls 1Mpts left/right. Try and do the same with 10Mpts or more!

To me it seems like responsiveness simply isn't a top requirement for other manufacturers. If you really want to, you can decouple the zoom/pan action from the actual waveform rendering. I'd be perfectly happy if the scope would show me a pixelated waveform while it's crunching the data and eventually updates the display from memory. For moving waveforms up/down, there's really no excuse in any sort of lag since the software just has to tell the FPGA/ASIC to draw the waveform somewhere else on the screen.

OTOH, the otherwise decent RTM3000 don't even manage to scroll smoothly through menus. Heck, even my 6 year old galaxy nexus phone does much better.

[Fungus]

there's really no excuse in any sort of lag since the software just has to tell the FPGA/ASIC to draw the waveform somewhere else on the screen.

Yep, I really don't understand the reason for lag in things like the vertical control. It should really have the same latency as the screen update rate.

There has to be a reason (lag must annoy the firmware programmers as well!), but I don't know what it is.

I have a suspicion that internally the scope is split into two parts, one CPU processing the input controls and sending commands to another CPU which is controlling the display. The commands are ASCII, exactly the same as in the programming guides:

eg. http://int.rigol.com/File/TechDoc/20151218/MSO1000Z&DS1000Z_ProgrammingGuide_EN.pdf

If the commands are sent over an internal serial interface and need to be confirmed then it makes some sense that there could be lag.

This programming model might sound like lunacy to a video game programmer but it makes a lot of sense from a DSO point of view where you have to accept commands from various sources (eg. control knobs, Ethernet, USB)

FWIW I've tried moving the vertical position on a Rigol DS1054Z via telnet and it has about the same responsiveness as using the knob.

[jasonbrent]

Keysight approved my trade in for 30% off the 3014T. Now to decide....

[jasonbrent]

Thank you all for the feedback here. It was useful; I ordered a DSOX3014T+DSOXLAN a few moments ago as a result... now time to put a couple of other pieces up for sell on rebay to recoup some costs, lol!

-j