Author Topic: True analog scopes  (Read 50460 times)

0 Members and 1 Guest are viewing this topic.

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: True analog scopes
« Reply #275 on: December 22, 2022, 06:29:03 pm »
Skill and imagination resolves all these issues.

The same is true for analog scopes. I'm firmly in the DSO camp now but we got by for many years with analog scopes, only a select few were lucky enough to have access to a storage scope in the analog era. I remember writing specific code that would cause a microcontroller to do the same thing repeatedly so I could see what was going on on the scope. There are all sorts of tricks people used to get the job done. A DSO is superior now in almost every case, but for decades they did not exist and a lot of the early ones had serious shortfalls. People found ways to get the CRO to show them what they needed to see. It's understandable that people stick to what they know, learning new concepts and skills gets harder for everyone the older we get. If some new whiz bang technology replaces DSOs 40 years from now I think there's a good chance you'll be clinging to your then ancient DSO because it's what you know.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: True analog scopes
« Reply #276 on: December 22, 2022, 06:33:32 pm »
That's the thing with only using old equipment, you guys are missing out on so much capability now that memory is so cheap.
Not so much in the sub$1k bracket but many midrange save to internal memory if you don't have a USB stick installed. Then in menus so many seem to despise, within the File manager you can grab those captures remotely or transfer them to a later installed USB stick.

That's not new. My 25 year old DSO has an internal hard drive that can save and transfer waveforms, screenshots and other data. It was a very expensive high end instrument when new though.
 

Offline BillyO

  • Super Contributor
  • ***
  • Posts: 1847
  • Country: ca
Re: True analog scopes
« Reply #277 on: December 22, 2022, 07:02:57 pm »
Tautech does bring up a solid point, and what is IMO the MAIN reason to use a DSO over an analog scope, and that's the ability to capture single-shot events.
Well, I'm not sure this is quite true.  I and others that have posted on this thread have noted that many better CROs can do single sweep triggering.  I used to use that feature all the time and long before DSOs were ever dreamed of.  Sure the DSO alleviates the need for a camera, but the feature is not something that came about with the advent of the DSO.

Y'all need to quit using this as the main reason to have a DSO.  It's not.
Bill  (Currently a Siglent fanboy)
--------------------------------------------------
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: True analog scopes
« Reply #278 on: December 22, 2022, 07:35:56 pm »

That is a standard mistake in logical thinking. Just because X implies Y, it does not mean not-X implies not-Y.

I am not quite sure what you are trying to say with that. But if we must, the MAIN reason to use a DSO includes:

Your list with my response...
Single shot capture The USP for storage scopes; agreed
Zooming on a captured waveform Analogue scopes have delayed trigger. (Some digitising scopes cannot zoom - useless!)
Programable triggers of many kinds I've found those to be good when playing around, but insufficient for normal jobs
Serial decoding Digital domain tools are much better: protocol analysers and printf(). Some digitising scopes only decode what's on the screen, which is useless
MSO Scope plus logic/protocol analyser is usually sufficient. Where not, there's a fair chance a digitising scope won't be either
Statistics displayed on screen Some analogue scopes display some stats on screen. While digitising scopes appear to have more complex stats, their accuracy in the context of the specific waveform has to be verified. Too many scope's firmware has been found to have subtle bugs
FFT 8bit ADCs are limited toys w.r.t. FFTs. 14bit is much better, and even then it would be better to use a sound card for audio or dedicated spectrum analyser for RF. Some digitising scopes only do an FFT of what's on the screen, which is useless
Network connectivity In most cases a convenient feature rather than a sigificant benefit
Programability by scripting / remote control I don't know how you can usefully script looking at an analogue waveform to find something unexpected. For digital signals use a digital domain tool:  a decent logic analyser or protocol analyser will have far superior triggering and filtering
Much better display Analogue displays are usually sufficient for analogue waveforms. For digital domain analysis, a digitising scope's screen quickly becomes a limiting factor; better to use the better triggering/filtering in a logic/protocol analyser, or offline post-processing in a computer
Auto setup Ah, the "I don't know how to work it / I don't know what I'm expecting to see" button beloved of those that like the "Tesla full self-driving (beta)" way of using an instrument :)
Multiple acquisition modes, Math modes I've never found those necessary, but provided an 8bit ADC isn't a limitation then it could be an advantage
Mask tests useful for production tests, less so for exploratory design
Power analyses / other kinds of analyses I use a DMM or the relevant special tool for those
Bode plot Severely limited by 8bit DACs, not just a simple digitising scope, can be achieved other ways

Quote
The list just continues. There is so much the analog scope can not do. Especially the analog scopes available cheaply.

The cheap digitising scopes are severely limited; they look enticing, but are disappointing when you look in detail.

Your other points are strawman arguments addressed previously.

Quote
And THAT explains why nobody is producing analog scopes anymore.

No. The main reason is economic: it is now far cheaper to make a new digitising scope than a new analogue scope.

Good old analogue scopes are far cheaper than decent new digitising scopes.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: True analog scopes
« Reply #279 on: December 22, 2022, 07:55:13 pm »
Y'all need to quit using this as the main reason to have a DSO.  It's not.

It's the main reason I use a DSO. Yes it's possible to do single sweep with a CRO and a camera, but that's a big pain. Doing a single shot capture on a DSO is trivial, it's the biggest advantage that the average user will see the most benefit in. They other perks they offer are nice too, but single sweep is the big one.
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: True analog scopes
« Reply #280 on: December 22, 2022, 09:13:53 pm »
Y'all need to quit using this as the main reason to have a DSO.  It's not.

It's the main reason I use a DSO. Yes it's possible to do single sweep with a CRO and a camera, but that's a big pain. Doing a single shot capture on a DSO is trivial, it's the biggest advantage that the average user will see the most benefit in. They other perks they offer are nice too, but single sweep is the big one.

... Single sweep with the displayed waveform retained by more than the phosphors' persistence :)

Even so, it is somewhat surprising how useful that can be, if that's all you've got :)
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline TimFox

  • Super Contributor
  • ***
  • Posts: 8995
  • Country: us
  • Retired, now restoring antique test equipment
Re: True analog scopes
« Reply #281 on: December 22, 2022, 09:16:59 pm »
An especially useful feature of DSOs is the ability to display the waveform in a single sweep before the triggering event.
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: True analog scopes
« Reply #282 on: December 22, 2022, 09:21:02 pm »
An especially useful feature of DSOs is the ability to display the waveform in a single sweep before the triggering event.

Agreed, that's something which the unlamented analogue storage scopes couldn't do.

It is less compelling with repetitive waveforms, of course.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline mag_therm

  • Frequent Contributor
  • **
  • Posts: 783
  • Country: us
Re: True analog scopes
« Reply #283 on: December 22, 2022, 09:42:44 pm »
ARRR..., But we had the original Hioki Hi-Corders, 4 channel, they could capture sub microsecond.
I forget how much pre-trigger but quite a lot, seconds even.
The chart paper would be spread across the engineering office floor so everybody could offer their learned opinion.
 

Offline baldurn

  • Regular Contributor
  • *
  • Posts: 189
  • Country: dk
Re: True analog scopes
« Reply #284 on: December 22, 2022, 10:16:39 pm »

Your list with my response...
Single shot capture The USP for storage scopes; agreed
Zooming on a captured waveform Analogue scopes have delayed trigger. (Some digitising scopes cannot zoom - useless!)

This is just so much easier on the DSO. Delayed trigger also does not exactly do the same thing. Segmented capture? Searching for events matching a pattern? Dual view of zoomed out and zoomed in view?

Quote
Programable triggers of many kinds I've found those to be good when playing around, but insufficient for normal jobs

There is just stuff you can not do without. Runt triggering for example is not easy or impossible on many analog scopes. Means you are simply not going to get the job done. Or that you have do something contrived and complex instead of something easy on the DSO.

Quote
Serial decoding Digital domain tools are much better: protocol analysers and printf(). Some digitising scopes only decode what's on the screen, which is useless
MSO Scope plus logic/protocol analyser is usually sufficient. Where not, there's a fair chance a digitising scope won't be either

You can always get more (digital!) instruments but the point is that the modern DSO is very versatile. It will usually do the job. It might not be the best for every job, nothing is. It is a function the analog scope simply does not do (again).

Quote
Statistics displayed on screen Some analogue scopes display some stats on screen. While digitising scopes appear to have more complex stats, their accuracy in the context of the specific waveform has to be verified. Too many scope's firmware has been found to have subtle bugs

There is no comparison. The old cheap analog scopes do not display any stats at all (only more advanced expensive ones do). Talking about bugs in firmware is just handwaving. The stats a very useful. Not because you can't do without but because it is so easy and its just there. No need to calculate something yourself when the machine does it.

Quote

FFT 8bit ADCs are limited toys w.r.t. FFTs. 14bit is much better, and even then it would be better to use a sound card for audio or dedicated spectrum analyser for RF. Some digitising scopes only do an FFT of what's on the screen, which is useless

I just compared FFT on my new Siglent SDS2104X HD with my spectrum analyzer. It was damn impressive. But even with the cheaper entry level scope SDS1104X-E you get 1 Mpts FFT which will help the average HAM operator to verify his transmitter just fine. I also find it very useful for teaching.

Could you use a sound card? Possibly but when you are done spending the day setting that up, the guy with a DSO just pressed a few buttons on his DSO to the job done. Not to mention that the sound card is yet another digital solution, it is limited to audio frequencies and usually requires additional hardware anyway.

Quote
Network connectivity In most cases a convenient feature rather than a sigificant benefit
Programability by scripting / remote control I don't know how you can usefully script looking at an analogue waveform to find something unexpected.

Not useless just because you have no need for networking. There has to be a reason that all modern instruments come with it. What might it be?

Quote
For digital signals use a digital domain tool:  a decent logic analyser or protocol analyser will have far superior triggering and filtering[/b]

Again the DSO is a versatile tool. It might not be the best but it can do a lot of stuff for free. Even the entry level scopes have this stuff. And sometimes it is just important to have the analog and digital signals on the same screen - nothing does that better than the DSO. The analog scopes does not do it at all.

Quote
Much better display Analogue displays are usually sufficient for analogue waveforms. For digital domain analysis, a digitising scope's screen quickly becomes a limiting factor; better to use the better triggering/filtering in a logic/protocol analyser, or offline post-processing in a computer

"sufficient" does not equal "good". Lets just admit that the small CRTs are horrible and that the new breed of entry level scopes with 10 inch touch screens are just so nice that you would think it came straight from Startrek... are there uses where an even bigger screen is better? Of course, that is why all new DSO are network connected so you can bring that right to your huge 30 inch desktop.

Quote
Auto setup Ah, the "I don't know how to work it / I don't know what I'm expecting to see" button beloved of those that like the "Tesla full self-driving (beta)" way of using an instrument :)

That is right. Beginners love that button. Sure sometimes they love it too much. Someone here claimed the analog scope is so easy because every lever is visible and right there. Well tell you what - the beginner just presses that one hated button and usually he gets what he wants. Not sure why it should be hated when its helps people.

Quote
Multiple acquisition modes, Math modes I've never found those necessary, but provided an 8bit ADC isn't a limitation then it could be an advantage
Mask tests useful for production tests, less so for exploratory design
Power analyses / other kinds of analyses I use a DMM or the relevant special tool for those
Bode plot Severely limited by 8bit DACs, not just a simple digitising scope, can be achieved other ways

That is pure rubbish. Bode plot is _very_ useful. It seems to be doing the job excellently for most purposes and 8 bit does not matter at all. Again the versatile tool DSO does it, but is not the only instrument or maybe the best instrument, but why should that take anything away if it solves the job?  Never seen an analog scope that could do bode plot.

Quote
Quote
The list just continues. There is so much the analog scope can not do. Especially the analog scopes available cheaply.

The cheap digitising scopes are severely limited; they look enticing, but are disappointing when you look in detail.

Your other points are strawman arguments addressed previously.

Quote
And THAT explains why nobody is producing analog scopes anymore.

No. The main reason is economic: it is now far cheaper to make a new digitising scope than a new analogue scope.

Good old analogue scopes are far cheaper than decent new digitising scopes.

The new entry level DSOs from Siglent and Rigol et al are fantastic devices. You are not going to find an analog scope + something that replaces all of what the DSO also does for anything less. That would be true even if the analog scope was free which is why everyone ends up with a DSO in the end.
 
The following users thanked this post: tautech

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: True analog scopes
« Reply #285 on: December 22, 2022, 10:52:11 pm »
I just compared FFT on my new Siglent SDS2104X HD with my spectrum analyzer. It was damn impressive. But even with the cheaper entry level scope SDS1104X-E you get 1 Mpts FFT which will help the average HAM operator to verify his transmitter just fine. I also find it very useful for teaching.

The reason I am never impressed with DSO FFTs is that they always lack the basic functionality of a true spectrum analyzer, like a noise marker and correction for RBW.  These things would be trivial to add, but for whatever reason DSO designers decline to do so.  My favorite mistake is when they allow averaging with FFTs, and then proceed to average the input to the FFT instead of the output.  And then we get to throwing away the phase results.

A 20 year old DSO with FFT is better than most DSO FFTs today.  Back then they didn't know to leave stuff out.

 

Offline alm

  • Super Contributor
  • ***
  • Posts: 2903
  • Country: 00
Re: True analog scopes
« Reply #286 on: December 22, 2022, 10:56:57 pm »
Never seen an analog scope that could do bode plot.
Any sweep function generator with any half-way decent scope with a second channel or external trigger input can do it:

For more bandwidth you could use an RF sweeper and an RF diode detector probe.

Quote
The new entry level DSOs from Siglent and Rigol et al are fantastic devices. You are not going to find an analog scope + something that replaces all of what the DSO also does for anything less. That would be true even if the analog scope was free which is why everyone ends up with a DSO in the end.
The Analog Discovery probably comes the closest, and can be bought a lot cheaper than decent entry-level DSOs during sales or on the used market. It contains a DSO, but not a very good one. It is quite nice for bode plotting and FFT due to its 14 bit resolution.

I think few people will argue that a modern DSO isn't much better than most analog scopes for at least 99% of use cases. But an analog scope is still infinitely better than not having a scope, and can still do a lot of useful work. Unless you have a financial interest in selling DSOs, then obviously it's a lost sale either way. In some economies, analog scopes might be available cheaply, while entry level DSOs may not be affordable.

A 20 year old DSO with FFT is better than most DSO FFTs today.  Back then they didn't know to leave stuff out.
Lecroy used to be decent with this (showing phase, math on math, etc). Have they removed it from their more recent models?
« Last Edit: December 22, 2022, 10:59:37 pm by alm »
 

Offline BillyO

  • Super Contributor
  • ***
  • Posts: 1847
  • Country: ca
Re: True analog scopes
« Reply #287 on: December 22, 2022, 11:01:49 pm »
.. they always lack the basic functionality of a true spectrum analyzer, like a noise marker and correction for RBW.  These things would be trivial to add, but for whatever reason DSO designers decline to do so.
Well if they did, how would they convince you to buy their spectrum analyzer?
Bill  (Currently a Siglent fanboy)
--------------------------------------------------
 

Offline BillyO

  • Super Contributor
  • ***
  • Posts: 1847
  • Country: ca
Re: True analog scopes
« Reply #288 on: December 22, 2022, 11:21:30 pm »
Never seen an analog scope that could do bode plot.
How do you think they were done in 1976?

I'd use my scope and an S-100 computer system running CP/M and wrote my own BASIC program to crunch the numbers and print the graphs.

DSOs are great, but we got by just fine without them.  We designed and built aircraft, spaceships, cars, particle accelerators, stereos, TVs, F1 cars, Lamborghini's, satellites, radio transceivers, X-ray machines, CT-scanners,  computers, etc.. 

We also wrote real software in C and assembler that ran word processors, spreadsheets, and engineering applications on 8-bit computer with 65,536 bytes of memory.

Today we have DSOs and people writing buggy, bloated slow as f**k code that requires a million times the processor power and memory to do essentially the same thing.  We're now just learning again how to put men on the moon.

Progress.

Don't get me wrong, I love me DSOs - all 4 of them, but rather than wait a minute for them to boot and then spend another 2 minutes getting lost in badly written menu systems, when I have to have a look at something right now, I usually head for my old Tek 465.
Bill  (Currently a Siglent fanboy)
--------------------------------------------------
 
The following users thanked this post: precaud

Offline baldurn

  • Regular Contributor
  • *
  • Posts: 189
  • Country: dk
Re: True analog scopes
« Reply #289 on: December 22, 2022, 11:23:10 pm »
Any sweep function generator with any half-way decent scope with a second channel or external trigger input can do it:

I can tell you that trick is nothing like the real thing.
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: True analog scopes
« Reply #290 on: December 23, 2022, 12:49:48 am »
Overall your points have some validity, but most don't stand up to detailed examination. Using strawman arguments doesn't help your contentions.


Your list with my response...
Single shot capture The USP for storage scopes; agreed
Zooming on a captured waveform Analogue scopes have delayed trigger. (Some digitising scopes cannot zoom - useless!)

This is just so much easier on the DSO. Delayed trigger also does not exactly do the same thing. Segmented capture? Searching for events matching a pattern? Dual view of zoomed out and zoomed in view?

Of course it isn't the same thing. Nonetheless the many of the advantages can be obtained by using analogue scopes with digital domain tools. In most cases the digitising scopes triggering filtering and searching digital domain information is worse that the alternative.

Use the right tools for the job; a jack-of-all-trades tool is often poorer than the combination of appropriate tools.

Quote
Quote
Programable triggers of many kinds I've found those to be good when playing around, but insufficient for normal jobs

There is just stuff you can not do without. Runt triggering for example is not easy or impossible on many analog scopes. Means you are simply not going to get the job done. Or that you have do something contrived and complex instead of something easy on the DSO.

Runt pulses are part of ensuring signal integrity, and a scope is the right tool for that. Once signal integrity is assured, flip to debugging in the digital domain with logic/protocol analysers.

It is possible, with skill, to use an LA to trigger on runt pulses.
It is often possible to create the conditions under which runt pulses are repetitive, and therefore not requiring storage scopes.

Quote
Quote
Serial decoding Digital domain tools are much better: protocol analysers and printf(). Some digitising scopes only decode what's on the screen, which is useless
MSO Scope plus logic/protocol analyser is usually sufficient. Where not, there's a fair chance a digitising scope won't be either

You can always get more (digital!) instruments but the point is that the modern DSO is very versatile. It will usually do the job. It might not be the best for every job, nothing is. It is a function the analog scope simply does not do (again).

I disagree that a DSOs "versatility" means it can usually do the job. It may be able to do your jobs, especially if they are relatively undemanding. My professional experience is littered with cases where tools ran into a brick wall and other tools had to be used instead. Often I had to create a my own custom tool for the job at hand. Using combinations of tools is standard operating procedure. But then I suppose if all you have is a digitising scope, then everything "looks like a nail".

"Swiss army knife" tools (i.e. tools with a principal purpose and other things added) tend to be substandard in all but their principal purpose.

I prefer to use the best tools for any given job.

Quote
Quote
Statistics displayed on screen Some analogue scopes display some stats on screen. While digitising scopes appear to have more complex stats, their accuracy in the context of the specific waveform has to be verified. Too many scope's firmware has been found to have subtle bugs

There is no comparison. The old cheap analog scopes do not display any stats at all (only more advanced expensive ones do). Talking about bugs in firmware is just handwaving. The stats a very useful. Not because you can't do without but because it is so easy and its just there. No need to calculate something yourself when the machine does it.

Strawman argument; nobody claimed grotty analogue scopes displayed stats.

The "old advanced expensive" ones are often cheaper than most modern digitising scopes. To take one example, a fully working 345MHz Tek 2465 or 485 costs £300-£450. What does a new >350MHz digitising scope cost?

Quote
Quote

FFT 8bit ADCs are limited toys w.r.t. FFTs. 14bit is much better, and even then it would be better to use a sound card for audio or dedicated spectrum analyser for RF. Some digitising scopes only do an FFT of what's on the screen, which is useless

I just compared FFT on my new Siglent SDS2104X HD with my spectrum analyzer. It was damn impressive. But even with the cheaper entry level scope SDS1104X-E you get 1 Mpts FFT which will help the average HAM operator to verify his transmitter just fine. I also find it very useful for teaching.

Including noise floor, intermod products, phase noise, harmonics and other spurii? I doubt it.

See David Hess' points about RBW etc.

Quote
Could you use a sound card? Possibly but when you are done spending the day setting that up, the guy with a DSO just pressed a few buttons on his DSO to the job done. Not to mention that the sound card is yet another digital solution, it is limited to audio frequencies and usually requires additional hardware anyway.

None of that invalidates my statement, and I don't understand what point you are trying to make.

Quote
Quote
Network connectivity In most cases a convenient feature rather than a sigificant benefit
Programability by scripting / remote control I don't know how you can usefully script looking at an analogue waveform to find something unexpected.

Not useless just because you have no need for networking. There has to be a reason that all modern instruments come with it. What might it be?

Ah, youngsters. I wonder how long it will be before you understand how marketing works (hint: checkbox specifications)

How, exactly, do you use scripting as described with analogue waveforms? I don't think scripting is as useful as you imagine (production tests excepted).

Quote
Quote
For digital signals use a digital domain tool:  a decent logic analyser or protocol analyser will have far superior triggering and filtering[/b]

Again the DSO is a versatile tool. It might not be the best but it can do a lot of stuff for free. Even the entry level scopes have this stuff. And sometimes it is just important to have the analog and digital signals on the same screen - nothing does that better than the DSO. The analog scopes does not do it at all.

Unconvincing argument.

Often there simply isn't space to have all the signals on the screen simultaneously. Combinations of tools ameliorate that by using "complex" filtering and triggering to not display irrelevant data.

Analogy (in the nowledge that all analogies are dangerous!): I use a car to travel on roads and ships to travel on water. An amphibious car would "do it all", but it would not be the best choice.

Quote
Quote
Much better display Analogue displays are usually sufficient for analogue waveforms. For digital domain analysis, a digitising scope's screen quickly becomes a limiting factor; better to use the better triggering/filtering in a logic/protocol analyser, or offline post-processing in a computer

"sufficient" does not equal "good". Lets just admit that the small CRTs are horrible and that the new breed of entry level scopes with 10 inch touch screens are just so nice that you would think it came straight from Startrek... are there uses where an even bigger screen is better? Of course, that is why all new DSO are network connected so you can bring that right to your huge 30 inch desktop.

... and let's just admit that a scope's screen isn't up to the job of displaying lots of complex digital signals, whereas a logic or protocol analyser's superior triggering and filtering will remove the unimportant stuff and allow you to concentrate on the useful stuff.

You use the "DSOs may not be good but they are sufficient" argument elsewhere. "Physician, heal thyself".

Quote
Quote
Auto setup Ah, the "I don't know how to work it / I don't know what I'm expecting to see" button beloved of those that like the "Tesla full self-driving (beta)" way of using an instrument :)

That is right. Beginners love that button. Sure sometimes they love it too much. Someone here claimed the analog scope is so easy because every lever is visible and right there. Well tell you what - the beginner just presses that one hated button and usually he gets what he wants. Not sure why it should be hated when its helps people.

Too often it gives them what they ask for, but they haven't asked the right question.
Too often it gives them what they ask for, but they don't understand the limitations of the answer.
Too often it gives them what they want, not what they need.
Too often it discourages sanity checking and/or verifying the validity of a result.

There is no substitute for thought and understanding.
In the case of "full self driving (beta)" there is no subsitute for being awake and actively monitoring the vehicle's actions.

Quote
Quote
Multiple acquisition modes, Math modes I've never found those necessary, but provided an 8bit ADC isn't a limitation then it could be an advantage
Mask tests useful for production tests, less so for exploratory design
Power analyses / other kinds of analyses I use a DMM or the relevant special tool for those
Bode plot Severely limited by 8bit DACs, not just a simple digitising scope, can be achieved other ways

That is pure rubbish. Bode plot is _very_ useful. It seems to be doing the job excellently for most purposes and 8 bit does not matter at all.

That first sentence is yet another strawman argument; unimpressive.

The 8bits can be a killer, e.g. when assessing filter stop-band performance behaviour.

Quote
Again the versatile tool DSO does it, but is not the only instrument or maybe the best instrument, but why should that take anything away if it solves the job?  Never seen an analog scope that could do bode plot.

See examples in other people's posts, and stop using strawman arguments.

Quote
Quote
Quote
The list just continues. There is so much the analog scope can not do. Especially the analog scopes available cheaply.

The cheap digitising scopes are severely limited; they look enticing, but are disappointing when you look in detail.

Your other points are strawman arguments addressed previously.

Quote
And THAT explains why nobody is producing analog scopes anymore.

No. The main reason is economic: it is now far cheaper to make a new digitising scope than a new analogue scope.

Good old analogue scopes are far cheaper than decent new digitising scopes.

The new entry level DSOs from Siglent and Rigol et al are fantastic devices. You are not going to find an analog scope + something that replaces all of what the DSO also does for anything less. That would be true even if the analog scope was free which is why everyone ends up with a DSO in the end.

Strawman arguments and non-sequiteurs. Unimpressive.
« Last Edit: December 23, 2022, 12:59:27 am by tggzzz »
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: True analog scopes
« Reply #291 on: December 23, 2022, 12:58:10 am »
The new entry level DSOs from Siglent and Rigol et al are fantastic devices. You are not going to find an analog scope + something that replaces all of what the DSO also does for anything less. That would be true even if the analog scope was free which is why everyone ends up with a DSO in the end.
The Analog Discovery probably comes the closest, and can be bought a lot cheaper than decent entry-level DSOs during sales or on the used market. It contains a DSO, but not a very good one. It is quite nice for bode plotting and FFT due to its 14 bit resolution.

I currently have two Digilent Analogue Discoveries, like them and wheel them out when I need their capabilities. (Must sell one of them :) )

The 14-bit ADC/DAC was a key benefit making me think they are worthwhile. 8bits was, um, uninticing.

Quote
I think few people will argue that a modern DSO isn't much better than most analog scopes for at least 99% of use cases. But an analog scope is still infinitely better than not having a scope, and can still do a lot of useful work. Unless you have a financial interest in selling DSOs, then obviously it's a lost sale either way. In some economies, analog scopes might be available cheaply, while entry level DSOs may not be affordable.

That's pretty much my position, but I would significantly reduce that 99% figure because I use combinations of tools. I have more than a hammer in my toolbox :)
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: True analog scopes
« Reply #292 on: December 23, 2022, 01:03:50 am »
Never seen an analog scope that could do bode plot.
How do you think they were done in 1976?

I'd use my scope and an S-100 computer system running CP/M and wrote my own BASIC program to crunch the numbers and print the graphs.

DSOs are great, but we got by just fine without them.  We designed and built aircraft, spaceships, cars, particle accelerators, stereos, TVs, F1 cars, Lamborghini's, satellites, radio transceivers, X-ray machines, CT-scanners,  computers, etc.. 

We also wrote real software in C and assembler that ran word processors, spreadsheets, and engineering applications on 8-bit computer with 65,536 bytes of memory.

Today we have DSOs and people writing buggy, bloated slow as f**k code that requires a million times the processor power and memory to do essentially the same thing.  We're now just learning again how to put men on the moon.

Progress.

Don't get me wrong, I love me DSOs - all 4 of them, but rather than wait a minute for them to boot and then spend another 2 minutes getting lost in badly written menu systems, when I have to have a look at something right now, I usually head for my old Tek 465.

Just so.

Except I prefer 2465 and 485 to a 465 :)

Too many youngsters fall into the trap of "I don't know how to build pyramids, therefore they didn't know how to build pyramids (hence aliens must have done it)".
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: True analog scopes
« Reply #293 on: December 23, 2022, 01:06:24 am »
Any sweep function generator with any half-way decent scope with a second channel or external trigger input can do it:

I can tell you that trick is nothing like the real thing.

Ditto using a digitising scope alone, vs a scope plus logic/protocol analysers and printf() statements.

Physician, heal thyself.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline BillyO

  • Super Contributor
  • ***
  • Posts: 1847
  • Country: ca
Re: True analog scopes
« Reply #294 on: December 23, 2022, 01:17:38 am »
Except I prefer 2465 and 485 to a 465 :)
As would I if I had either.

Too many youngsters fall into the trap of "I don't know how to build pyramids, therefore they didn't know how to build pyramids (hence aliens must have done it)".
Yep, and there is a growing (or is that groaning?) number of them that think the earth is flat.  :palm:

Heaven forbid they solve an engineering approach for themselves instead of pouting that their scope won't do it for them.

Anything to be anti-boomer.
« Last Edit: December 23, 2022, 01:19:42 am by BillyO »
Bill  (Currently a Siglent fanboy)
--------------------------------------------------
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 29809
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: True analog scopes
« Reply #295 on: December 23, 2022, 02:24:22 am »
Love to stop and chat but too darn busy with a Xmas rush selling these infernal modern DSO's....higher priced ones too !
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Offline vk6zgo

  • Super Contributor
  • ***
  • Posts: 7854
  • Country: au
Re: True analog scopes
« Reply #296 on: December 23, 2022, 02:26:54 am »
An especially useful feature of DSOs is the ability to display the waveform in a single sweep before the triggering event.

Agreed, that's something which the unlamented analogue storage scopes couldn't do.

It is less compelling with repetitive waveforms, of course.

Analog 'scopes with delay lines can display the waveform in a recurrent waveform before the trigger event, although it is seldom essential.
 

Offline vk6zgo

  • Super Contributor
  • ***
  • Posts: 7854
  • Country: au
Re: True analog scopes
« Reply #297 on: December 23, 2022, 03:20:38 am »
Any sweep function generator with any half-way decent scope with a second channel or external trigger input can do it:

I can tell you that trick is nothing like the real thing.
When I learnt about "Bode plots", it was done by plotting transfer functions on graph paper.
This would give unrealistically sharp corner frequencies.
They were normally done for amplitude and phase response.

When we measured amplitude-v-frequency response of real devices, we used a RF or audio sweeper & an oscilloscope, or an "all in one" sweeper like a Polyskop SWOB.

We called it what it was----a "amplitude/frequency response test".

Back the day, VNAs were "rare as hen's teeth" so we didn't normally have the occasion to do "phase/frequency" testing.

Do the common modern DSOs perform both of these types of test?
 

Offline TimFox

  • Super Contributor
  • ***
  • Posts: 8995
  • Country: us
  • Retired, now restoring antique test equipment
Re: True analog scopes
« Reply #298 on: December 23, 2022, 03:45:33 am »
An especially useful feature of DSOs is the ability to display the waveform in a single sweep before the triggering event.

Agreed, that's something which the unlamented analogue storage scopes couldn't do.

It is less compelling with repetitive waveforms, of course.

Analog 'scopes with delay lines can display the waveform in a recurrent waveform before the trigger event, although it is seldom essential.

Analog scopes with delay lines allowed the time at the trigger event to be displayed, but very, very little before it.
 

Offline BillyO

  • Super Contributor
  • ***
  • Posts: 1847
  • Country: ca
Re: True analog scopes
« Reply #299 on: December 23, 2022, 04:01:04 am »
Love to stop and chat but too darn busy with a Xmas rush selling these infernal modern DSO's....higher priced ones too !
That's all there is to sell .. that's what gets sold.

But, as said before, I love my DSOs.  They will always be my go-to for serious work.  No argument.  But I'm not tossing my CRO anytime soon.

I'm willing to admit that my reflexes are better habituated to a (Tektronix) CRO, but that's what I learned on and spent most of my career as an ET with.  That said, the advantages of a modern DSO are seductive for sure.  And if I have something more serious to do, that's where I go.  However, if the moment calls for a reflexive quick peek at an event or waveform I can best leverage a 40YO Tek to get the job done.  Both technologies have a place in my life.

I can also do the timing and adjust the points, synchronize the carbs and prep the plugs on my '74 RD350.  Many here would be at a total loss and never be able to enjoy the feel of riding an untamed, raw twin cylinder 2-stoke bike of epic reputation.  I pity those that cannot appreciate where they, or their world, came from.

Let me lay out a question.  My Tek 465 has outlived it's era by ~30Y+ but still provides purpose and usefulness.  Can I expect my "improved" SDS2104XP to do the same?  My 12YO UNI-T UT2102CEL has outlived it's usefulness and it's 1/4 the age of the 465. :-//
Bill  (Currently a Siglent fanboy)
--------------------------------------------------
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf