Modern digital scopes: real-time sampling or equivalent-time sampling

[tggzzz]

Ahem, this document is from 1996! It's almost 20 years old! The technology that HP has offered back then is no longer relevant!

Well it was your choice of document, so naturally I referred to the points you were making!

And the question if you see an interpolated signal on the screen or not has absolutely nothing to do with ETS or RTS. It simply depends on your scope, i.e. can you switch off interpolation or is it forced enabled. For example, on LeCroy scopes interpolation is optional that has to be enabled by the user, as their philosophy is that a scope shall always display a signal unaltered by default. Other vendors have different mindsets.

But again, this has nothing to do with ETS vs RTS.

Yes, but I was referring to the points in "your" reference.

I appreciate that we have many older EE's in this forum who have grown up with analog scopes and who never got really into DSOs, but really, it's not 1986 any more, and scope technology and prices have changed a lot since then. And there's a reason why scopes these days up to 65Ghz (or 33GHz if you're Tek) are RTS scopes.

I started using DSOs when they first became available. I recognised their advantages and disadvantages, as I do with any tool. Nowadays if you have money to spend, digital is clearly the way to go.

Analogue scopes still have two advantages for beginners: the UI is usually simpler to use (fewer options, not hidden behind menu buttons), and there is less chance for misleading marketing specmanship.

[Wuerstchenhund]

Ahem, this document is from 1996! It's almost 20 years old! The technology that HP has offered back then is no longer relevant!

Well it was your choice of document, so naturally I referred to the points you were making!

The point I was making was that there were some issues using ETS only scopes for Eye Diagrams, and I did highlight that this document was written at a time when RTS scopes had much lower sample rates.

I certainly didn't suggest to use HP's 1996 technology instead!

And the question if you see an interpolated signal on the screen or not has absolutely nothing to do with ETS or RTS. It simply depends on your scope, i.e. can you switch off interpolation or is it forced enabled. For example, on LeCroy scopes interpolation is optional that has to be enabled by the user, as their philosophy is that a scope shall always display a signal unaltered by default. Other vendors have different mindsets.

But again, this has nothing to do with ETS vs RTS.

Yes, but I was referring to the points in "your" reference.

See above. It seems you didn't really get the point I was making. The document describes a problem with ETS scopes at a time RTS sample rates were lower than today. If anything, the advantage of RTS over ETS has only increased since then.

Analogue scopes still have two advantages for beginners: the UI is usually simpler to use (fewer options, not hidden behind menu buttons), and there is less chance for misleading marketing specmanship.

I disagree with both. The UI has fewer options but isn't necessarily simpler for beginners (the amount of options doesn't necessarily make a UI better, in fact, the lack of any supporting facilities like a help functionality or task-dependent user guidance can make it pretty difficult for a starter). And I remember things back old times, like when an unnamed company overestimated the capabilities of the trigger in some of their analog scopes. It wasn't all good back then.

On the other side, analog scopes teach beginners practices of which many are no longer appropriate when using a proper modern DSO. I believe it is much better they learn how to handle a modern tool properly right from the start instead of learning outdated methods on a museum piece. I mean, we don't teach students of medicine how to properly set bloodsuckers or do a bloodletting any more, do we?

[David Hess]

I agree but not all DSOs made today have sufficient real time sampling rates to support their input bandwidth without aliasing even with input signals which are completely below their Nyquist frequency.

I'm not really aware of any mainstream one that isn't?
Tek set the benchmark 15+ years ago with the TDS200 series RTS scopes.

I would say that most of the time it is not required for analog design work including digital signal integrity.  I am actually disappointed in the current crop of 100 MHz 250 MS/s DSOs because of their lack of ETS.

But 250MSPS is just enough to reconstruct the waveform if you use Sin X/x interpolation. In theory I believe that in most cases you can't get a huge amount more real information out of a 100MHz bandwidth limited signal once you pass the 2.4x mark, i.e. 240MSPS for 100MHz bandwidth. But that does depend on the filter type used etc.
Either way, it think it's pretty marginal.
That is likely why Rigol have simply dropped ETS on the 1000Z.

I was hoping someone would bring this up.

I certainly agree that sin(x)/x reconstruction is completely sufficient to reconstruct an unaliased waveform.

There is an EEVBlog discussion here from 2010 with a video which illustrates the problem at about 46 seconds in:

https://www.eevblog.com/forum/chat/rigol-ds1000e-series-possible-errorfail-in-sin%28x%29x-interpolation/

I disagree with the analysis of the cause in that discussion.  The sin(x)/x reconstruction makes it more visible but what is being shown is the result of aliasing caused by non-linearity and sampling error in the digitizer (and oscilloscope front end) itself.

Tek's real time DSOs suffer from the same problem to one extent or another (which Agilent loves to harp on) but higher sampling rates whether real time or equivalent time ameliorate it unless they come at the expense of greater distortion.  Equivalent time sampling rates are usually so high that it becomes insignificant.

The LeCroy white paper Wuerstchenhund linked to does not discuss this specifically but touches on it in connection with noise from interleaving and filtering to remove it.  The technology from SP Devices that pascal_sweden linked includes linearization after digitizing which would help prevent it as well and may be the same thing LeCroy was referring to as their special filtering.  I suspect a lot of high end DSOs do this to one extent or another.  The interleaved ADCs Rigol is using (or at least the TI ones like the ADC08D500) include self calibration which helps to prevent this.

[David Hess]

because I read the paper in detail and the only difference they identify is filtering after interleaving which is not unique to LeCroy

Who else does it exactly the same then? Tek not, according to the documents I've seen and what I remember from the Tek DSOs I've used. So if you have some reference please share it.

We cannot know if Tektronix or Agilent does it the same way even if they discuss it publicly because LeCroy does not describe how their filter works in detail or what it is really doing.  LeCroy even admits that others do filtering similar to theirs:

The main implementation difference between LeCroy and other vendors is the filtering of the final interleaved RIS trace. In other words, some vendors simply acquire the waveform segments and interleave them together.

Which says that some vendors do not and instead do what LeCroy is doing whatever that is.

and they do not even include that difference in their summary.

No, they don't, because unlike say Agilent's or Tek's our-vs-theirs comparisons where the main aim is to make the own product look good, the aim of this document is to describe what RIS does, what its limits are and how the implementation varioes between vendors.

The document reminds me exactly of a similar Agilent document which discusses distortion from interleaved digitizers and then shows how poorly a Tektronix oscilloscope performs in that regard.

Distinguishing their products from their competitors is just good marketing.

[David Hess]

Cost, well, a 100GHz ETS scope is slightly cheaper than a 65GHz RTS scope, but we're talking about prices in the region of $50k to $100k and more. Chances are that if you're working on such high complexity projects then you can afford a proper scope as well.

For everything in the lower bandwidth ranges (i.e. below 5GHz) ETS isn't a sensible options. There are no ETS only scopes available in that bandwidth range, so you'd either have to settle for one of the few USB ETS scopes or hunt for a museum piece on ebay, which will be old, big, loud, noisy, power hungry, and chances are good spare parts are no longer available.

I think there is some confusion here with the term ETS.  The original post mentioned the Tektronix TLS216 which uses only real time sampling as far as I know and the Philips PM3340 which is a sequential equivalent time sampling oscilloscope optimized for high bandwidth.

As you point out, sequential equivalent time DSOs are almost extinct having been replaced with very high end real time DSOs.  However many lower cost real time DSOs also support random equivalent time sampling.

I appreciate that we have many older EE's in this forum who have grown up with analog scopes and who never got really into DSOs, but really, it's not 1986 any more, and scope technology and prices have changed a lot since then. And there's a reason why scopes these days up to 65Ghz (or 33GHz if you're Tek) are RTS scopes.

I have always liked DSOs at least since the Tektronix 2230.

The reason that practically all DSOs use real time sampling now is that fast digitizers, memory, and the logic to support them have decreased in price until the incremental cost of RTS versus random ETS at a slower sampling rate is small.  Many RTS DSOs still support random ETS.  Only the lowest costs ones and the ones with very high real time sample rates compared to their bandwidth forgo it.

[EEVblog]

I was hoping someone would bring this up.
I certainly agree that sin(x)/x reconstruction is completely sufficient to reconstruct an unaliased waveform.

Tek on the issue:
http://www.tek.com/dl/55W_17589_2.pdf

I have not read it all, but they basically summarise that there is no practical difference between Sin X/x and ETS.

From that Lecroy paper:

Setting up the DSO to Enhance Validity of SinX Interpolation
All interpolation methods gain in validity as the ratio of the sample rate to the bandwidth
grows. Interpolation will always improve as the sample rate is made higher. Some rules
of thumb are in order. Linear interpolation works very well only when the ratio of the
sample rate to the highest frequency component is at least 10 to 1. SinX interpolation
works very well only when this ratio is greater than 2:1 - 3:1 is a good ratio with 4:1
usually working almost perfectly.
Using the LeCroy WaveMaster 8620A as an example, SinX interpolation is almost
perfectly valid at the highest channel sample rate of 20 GS/s. This is because the
bandwidth of the scope is 6 GHz, with such a sharp dropoff in response that the signals
are greatly attenuated at and above 7 GHz. Since the Nyquist rate at 20 GS/s is 10 GHz,
Nyquist's criterion is met and SinX interpolation is highly effective. In effect, the
bandwidth limitation of the scope ensures that interpolation is always valid at 20 GS/s.

Bottom line is that all the manufacturers seem to say a similar thing, and I have a more detailed paper mathematically proving 2.4x minimum in some way (can't find it now). So for all but the most critical applications, a 250MSPS 100MHz scope with Sin X/x should be just fine.

[tggzzz]

Analogue scopes still have two advantages for beginners: the UI is usually simpler to use (fewer options, not hidden behind menu buttons), and there is less chance for misleading marketing specmanship.

I disagree with both. The UI has fewer options but isn't necessarily simpler for beginners (the amount of options doesn't necessarily make a UI better, in fact, the lack of any supporting facilities like a help functionality or task-dependent user guidance can make it pretty difficult for a starter).

My experience of teaching beginners is that having everything visible and logically arranged on the front panel helps guide them towards making appropriate use of the features. The problem with "hidden" options is you don't know they are there until after you've looked for them.

As for manuals, well the acronym "RTFM" is there for a good reason T'was ever thus. As for DWIM techniques (sorry, "task dependent help"), to some extent they are necessary as a result of having by hidden controls

But there are no absolutes here, and the comments only apply to beginners.

And I remember things back old times, like when an unnamed company overestimated the capabilities of the trigger in some of their analog scopes. It wasn't all good back then.

Strawman argument! T'was ever thus!

On the other side, analog scopes teach beginners practices of which many are no longer appropriate when using a proper modern DSO. I believe it is much better they learn how to handle a modern tool properly right from the start instead of learning outdated methods on a museum piece. I mean, we don't teach students of medicine how to properly set bloodsuckers or do a bloodletting any more, do we?

Neither do we teach them to drive cars using a Bugatti Veyron (I think that's the latest hot car). No, we teach drivers on gutless new clunkers.

[edavid]

I was hoping someone would bring this up.
I certainly agree that sin(x)/x reconstruction is completely sufficient to reconstruct an unaliased waveform.

Bottom line is that all the manufacturers seem to say a similar thing, and I have a more detailed paper mathematically proving 2.4x minimum in some way (can't find it now). So for all but the most critical applications, a 250MSPS 100MHz scope with Sin X/x should be just fine.

You missed David Hess's important point that this is only valid for an unaliased waveform.  Since the Rigol DS1000Z doesn't have a sharp enough antialiasing filter in front of the ADC (as shown by the bandwidth measurements that people have posted), the 250MSPS sample rate is not adequate to support 100MHz bandwidth.

[edavid]

The PM3340 was a 2GHz scope with 250MSa/s real-time sampling rate and 14bit vertical resolution which also offered 2GSa/s in ETS mode. It was a good scope at it's time (1989; I had a PM3343 back then which was the 200MHz version of the PM3340) but by today's standards it's a boat anchor. It's also difficult to fix with lots of unobtainium parts. The high vertical resolution made (and still makes) especially the PM3343 sought after for audio work, though.

The PM3340 manual shows that it has a 10 bit ADC, not 14 bit.

Are you sure it was 250MSPS?  It's hard to believe they could build even 10bit  250MSPS ADCs back then.

[Wuerstchenhund]

Who else does it exactly the same then? Tek not, according to the documents I've seen and what I remember from the Tek DSOs I've used. So if you have some reference please share it.

We cannot know if Tektronix or Agilent does it the same way even if they discuss it publicly because LeCroy does not describe how their filter works in detail or what it is really doing.  LeCroy even admits that others do filtering similar to theirs:

The main implementation difference between LeCroy and other vendors is the filtering of the final interleaved RIS trace. In other words, some vendors simply acquire the waveform segments and interleave them together.

Which says that some vendors do not and instead do what LeCroy is doing whatever that is.

It doesn't mean other vendors do the same. It is a non-exclusive statement, i.e. that the author doesn't know if other vendors do the same.

At least the big names don't seem to use filtering in their ETS modes.

The document reminds me exactly of a similar Agilent document which discusses distortion from interleaved digitizers and then shows how poorly a Tektronix oscilloscope performs in that regard.

Only that Agilent often stretches the truth to the extreme and goes to immense lengths to make other products look bad.

Distinguishing their products from their competitors is just good marketing.

It is, but there is a difference between construing a synthetic situation to make the competitor look bad and stating some  facts which are easy to check. This document doesn't even say that the competition is inferior, all they say is that there are differences and that the user should know how the specific ETS implementation works independent of who manufactured the scope.

[Wuerstchenhund]

The reason that practically all DSOs use real time sampling now is that fast digitizers, memory, and the logic to support them have decreased in price until the incremental cost of RTS versus random ETS at a slower sampling rate is small.  Many RTS DSOs still support random ETS.  Only the lowest costs ones and the ones with very high real time sample rates compared to their bandwidth forgo it.

That's exactly it 

[Wuerstchenhund]

My experience of teaching beginners is that having everything visible and logically arranged on the front panel helps guide them towards making appropriate use of the features. The problem with "hidden" options is you don't know they are there until after you've looked for them.

True, but that is no different for a DSO. What you call "hidden menu" (which btw aren't really hidden on most scopes, it's often more of a case of 'can't be bothered to learn how to use my tool properly') is usually only required if you need advanced functionality.

As for manuals, well the acronym "RTFM" is there for a good reason T'was ever thus. As for DWIM techniques (sorry, "task dependent help"), to some extent they are necessary as a result of having by hidden controls

"RTFM" applies to any tool, even an analog scope (where else would you learn about it's limitations?).

And I remember things back old times, like when an unnamed company overestimated the capabilities of the trigger in some of their analog scopes. It wasn't all good back then.

Strawman argument! T'was ever thus!

Not really. You seem to think that vendors only got creative with their specifications when DSOs came along (and your signature supports that impression) but that doesn't conform with reality, which is that the specs of your average big brand DSO are as reliable as they were for analog scopes.

On the other side, analog scopes teach beginners practices of which many are no longer appropriate when using a proper modern DSO. I believe it is much better they learn how to handle a modern tool properly right from the start instead of learning outdated methods on a museum piece. I mean, we don't teach students of medicine how to properly set bloodsuckers or do a bloodletting any more, do we?

Neither do we teach them to drive cars using a Bugatti Veyron (I think that's the latest hot car). No, we teach drivers on gutless new clunkers.

I'm sorry but like in most cases where someone comes up with a car analogy this one is silly, too. It doesn't matter if you drive a Yugo or a Veyron, all cars have a steering wheel and the same set of basic controls (throttle, brake, clutch for stick shifts, blinker, gear/transmission lever). Both cars work exactly the same (the energy from a piston combustion engine goes to the transmission and from there to the wheels). A Veyron might be of much higher performance than a Yugo but essentially you drive it the same, especially on public roads. In short: if you can drive a shitty Yugo perfectly then you can drive a Veyron, too. No need to re-learn driving.

If it has to be about cars, a better analogy would be teaching a driving student using horse and buggy.

That's not necessarily true with analog scopes and DSOs. The operating principles (direct analog display vs sample and digital storage) are completely different, which means there are different things to consider when using a DSO than if an analog scope was used. This also means that treating a DSO like an analog scope often won't cut it and give poor/false results. Some basic controls are the same on analog and digital scopes but that's about it, and as soon as you want to have a closer look at a complex signal your analog scope leaves you with maybe some primitive CRT storage mode and (if you're lucky) some basic cursors with readouts while a good DSO offers you a full suite of signal analysis, maths, decoding etc. So unlike a driver that moves from a Yugo to a Veyron, even an EE who has many years or decades of experience in using analog scopes will have to learn quite a bit of new stuff when moving to a DSO, and much more if it's one of the more advanced ones (although the hardest part is giving up old habits from the analog scope days).

When I train apprentices and students I want to prepare them to be able to make the best use of tools they will be working with when designing the Next Greatest Thing(tm), not to become curator in a T&M museum. I can see that an analog scope can still be useful for many simpler tasks, and there's nothing wrong with that. But in this day and age it no longer should be the entry for someone who wants to learn using an oscilloscope properly.

[Wuerstchenhund]

The PM3340 manual shows that it has a 10 bit ADC, not 14 bit.

I had a look at some old documents. The scope I had was a PM3320A, not PM3343 (not sure the latter even exists).

I couldn't find the specs on a quick search but you're probably right that it was 10bit only.

Are you sure it was 250MSPS?  It's hard to believe they could build even 10bit  250MSPS ADCs back then.

Yes, 200MHz bandwidth and 250MSa/s sample rate.

[David Hess]

The PM3340 manual shows that it has a 10 bit ADC, not 14 bit.

I had a look at some old documents. The scope I had was a PM3320A, not PM3343 (not sure the latter even exists).

I couldn't find the specs on a quick search but you're probably right that it was 10bit only.

Are you sure it was 250MSPS?  It's hard to believe they could build even 10bit  250MSPS ADCs back then.

Yes, 200MHz bandwidth and 250MSa/s sample rate.

I remember wanting a PM3382/PM3384/PM3392/PM3394.  The PM3394 is a 4 channel, 200 MHz, combination analog/DSO with a 200 MS/s real time sample rate which is shared in somewhat between channels.  They support real time and random equivalent time sampling.  Feature wise they seemed better than the Tektronix equivalents.

The Philips PM3340 is a sequential digital sampling oscilloscope with a 2 GHz input bandwidth, 10 bit vertical resolution, and based on the specifications generates a 512 point record in less than 25 milliseconds which works out to a sampling rate greater than 20 kSamples/second.

It has built in delay lines after the trigger pickoffs so it can capture the triggering edge unlike most sampling oscilloscopes.  This will be what actually limits the input bandwidth.

There was a short discussion on EEVBlog about it here:

https://www.eevblog.com/forum/buysellwanted/philips-3340-2ghz-scope-%28uk%29/

[tggzzz]

My experience of teaching beginners is that having everything visible and logically arranged on the front panel helps guide them towards making appropriate use of the features. The problem with "hidden" options is you don't know they are there until after you've looked for them.

True, but that is no different for a DSO. What you call "hidden menu" (which btw aren't really hidden on most scopes, it's often more of a case of 'can't be bothered to learn how to use my tool properly') is usually only required if you need advanced functionality.

As for manuals, well the acronym "RTFM" is there for a good reason T'was ever thus. As for DWIM techniques (sorry, "task dependent help"), to some extent they are necessary as a result of having by hidden controls

"RTFM" applies to any tool, even an analog scope (where else would you learn about it's limitations?).

Of course, as I previously noted the words I have now emphasised in my original statement.

And I remember things back old times, like when an unnamed company overestimated the capabilities of the trigger in some of their analog scopes. It wasn't all good back then.

Strawman argument! T'was ever thus!

Not really. You seem to think that vendors only got creative with their specifications when DSOs came along (and your signature supports that impression) but that doesn't conform with reality, which is that the specs of your average big brand DSO are as reliable as they were for analog scopes.

You are falsely inferring things I didn't write. Again, note the words I have now emphasised in my original statement.

On the other side, analog scopes teach beginners practices of which many are no longer appropriate when using a proper modern DSO. I believe it is much better they learn how to handle a modern tool properly right from the start instead of learning outdated methods on a museum piece. I mean, we don't teach students of medicine how to properly set bloodsuckers or do a bloodletting any more, do we?

Neither do we teach them to drive cars using a Bugatti Veyron (I think that's the latest hot car). No, we teach drivers on gutless new clunkers.

I'm sorry but like in most cases where someone comes up with a car analogy this one is silly, too.

My mechanical analogy is as stupid and unhelpful as your previous medical analogy. By "too", are you referring to your medical analogy?

It doesn't matter if you drive a Yugo or a Veyron,
...

I've snipped a long diatribe about something you claim is silly, for the simple reason that all the analogies in this thread are misleading and unhelpful.  Your elaborating them does not change that.

When I train apprentices and students I want to prepare them to be able to make the best use of tools they will be working with when designing the Next Greatest Thing(tm), not to become curator in a T&M museum. I can see that an analog scope can still be useful for many simpler tasks, and there's nothing wrong with that. But in this day and age it no longer should be the entry for someone who wants to learn using an oscilloscope properly.

If you knew my career history, you would realise just how off-beam your presumptions are.

You seem to be making a habit of feeling something warm, grey, wrinkled, 2ft(!) in diameter, and hypothesising an elephant.

[David Hess]

I was hoping someone would bring this up.
I certainly agree that sin(x)/x reconstruction is completely sufficient to reconstruct an unaliased waveform.

Bottom line is that all the manufacturers seem to say a similar thing, and I have a more detailed paper mathematically proving 2.4x minimum in some way (can't find it now). So for all but the most critical applications, a 250MSPS 100MHz scope with Sin X/x should be just fine.

You missed David Hess's important point that this is only valid for an unaliased waveform.  Since the Rigol DS1000Z doesn't have a sharp enough antialiasing filter in front of the ADC (as shown by the bandwidth measurements that people have posted), the 250MSPS sample rate is not adequate to support 100MHz bandwidth.

My more significant point is that the situation is actually worse than that and I linked to that EEVBlog discussion and video to show exactly why:

https://www.youtube.com/watch?feature=player_detailpage&v=W7Opur1Xbvs#t=44

Assume for the moment that the signal source is perfect or the analog antialiasing filter is perfect and no frequency components higher the the Nyquist frequency are present at the input to the digitizer.  If the signal source is close to but below the Nyquist frequency, then nonlinearity and sampling clock error in the digitizer will produce harmonic distortion and that will be above the Nyquist limit.  They will also produce mixing between the input signal and sampling clock which will produce sidebands and some of those will be above the Nyquist frequency.

The result is shown in the time domain in the above video as what I like to call "wobbulation".  I suspect the same thing on one of the Rigol oscilloscopes with an index graded display makes for waveform thickening which is mistaken for noise.  An older Rigol in envelope mode will show the same apparent noise.

A simple way to verify this is to single shot capture a clean high frequency sine wave and run an FFT on it which will display spurs in the passband caused by aliasing.

Page 12 of this Agilent application note discusses this:

http://www.newark.com/pdfs/techarticles/agilent/EvaluationgOscilloscopeRatesFidelity.pdf

I was hoping someone would bring this up.
I certainly agree that sin(x)/x reconstruction is completely sufficient to reconstruct an unaliased waveform.

Tek on the issue:
http://www.tek.com/dl/55W_17589_2.pdf

I have not read it all, but they basically summarise that there is no practical difference between Sin X/x and ETS.

I do not disagree with this provided that the oversampling ratio is high which is what ETS provides or the sin(x)/x interpolated waveforms are averaged.  The very end of the Tektronix application note you linked says this very thing:

... An interesting point is that the average of interpolated waveforms (*) and the average of ET mode waveforms produce a virtually identical result! This tends to contradict misconceptions that sin(x)/x interpolation does not accurately reproduce high speed digital signals.

I have taken advantage of both techniques as appropriate on my old Tektronix 2440 which suffers greatly from interleave distortion by design although it is not particularly worse than modern DSOs operating at similar (500 MS/s) sample rates.  It is just annoying and something I would look to minimize if I was in the market for a newer DSO.

(*) Sin(x)/x interpolated waveforms.

[Wuerstchenhund]

[lots of stuff not worth quoting]

It seems you've now resented replace arguments in an adult discussion with childish diatribe, aggressiveness and personal attacks. Although it's unfortunate (but not completely surprising), to keep the peace in this forum I will simply end the discussion with you.

[Wuerstchenhund]

I remember wanting a PM3382/PM3384/PM3392/PM3394.  The PM3394 is a 4 channel, 200 MHz, combination analog/DSO with a 200 MS/s real time sample rate which is shared in somewhat between channels.  They support real time and random equivalent time sampling.  Feature wise they seemed better than the Tektronix equivalents.

Yes, Philips/Fluke ("The T&M Alliance") did have some nice scopes at that time. I remember the PM338x/339x models. We had some at work (together with some PM305x/PM307x analog scopes), and most of our engineers seemed to prefer the Philips scopes over the Tek 465's we also had (I did).

The Philips PM3340 is a sequential digital sampling oscilloscope with a 2 GHz input bandwidth, 10 bit vertical resolution, and based on the specifications generates a 512 point record in less than 25 milliseconds which works out to a sampling rate greater than 20 kSamples/second.

I vaguely remember that the PM3340 had a very low real-time sample rate, but at that time that was pretty much standard for (sequential/random) ETS scopes.

I think it's a shame that "the T&M Alliance" didn't work out.

[David Hess]

I remember wanting a PM3382/PM3384/PM3392/PM3394.  The PM3394 is a 4 channel, 200 MHz, combination analog/DSO with a 200 MS/s real time sample rate which is shared in somewhat between channels.  They support real time and random equivalent time sampling.  Feature wise they seemed better than the Tektronix equivalents.

Yes, Philips/Fluke ("The T&M Alliance") did have some nice scopes at that time. I remember the PM338x/339x models. We had some at work (together with some PM305x/PM307x analog scopes), and most of our engineers seemed to prefer the Philips scopes over the Tek 465's we also had (I did).

This era was before my time so the best I could do was drool over the product brochures and catalogs.  I was happy enough to have a Lavoie LA-265A (30 MHz dual trace delayed sweep) which is a clone of a Tektronix 545A.  Since it has leather handles on top, I consider it portable although portability is limited to crushing things with its weight.

I remember advertising at the time comparing the PM338x/339x models to the Tektronix 2232 series with the Philips oscilloscopes having twice the sample rate, twice the bandwidth, twice the number of channels, and math/FFT support.  I see them show up on Ebay occasionally in what looks like reasonable shape.

I have been told that Hitachi and/or Philips analog oscilloscopes had sharper traces than Tektronix did on their 465 series but this was after Tektronix had started using a scan expansion mesh in their CRTs for greater deflection sensitivity which would explain that.  Personally I have not noticed a significant difference but I have not had a Fluke or Hitachi available to make a direct comparison.

The Philips PM3340 is a sequential digital sampling oscilloscope with a 2 GHz input bandwidth, 10 bit vertical resolution, and based on the specifications generates a 512 point record in less than 25 milliseconds which works out to a sampling rate greater than 20 kSamples/second.

I vaguely remember that the PM3340 had a very low real-time sample rate, but at that time that was pretty much standard for (sequential/random) ETS scopes.

This was certainly true of the sampling oscilloscopes and of course continues to be the case.  The Tektronix 7854/7T11A runs at about 50 kSamples/second in analog or digital mode and the 7854 alone (400 MHz bandwidth) runs at about 500 kSamples/second (10 bits though) although the former could be considered "real time" in analog mode while the later was not even close despite having 10 times and sample rate and an order of magnitude less bandwidth.  Nobody would mistake the 7854 for a modern DSO.

Their follow on high bandwidth 11k series digital sampling oscilloscopes ran at about 200 kSamples /second.

The contemporary real time digital storage oscilloscopes were 40 to 100 MSamples/second but quickly rose to 500 MSamples /second and faster in the 2440, 11k, and later TDS series DSOs.  I think they all supported random equivalent time sampling as required except maybe for some of the oddball TDS models.

I think it's a shame that "the T&M Alliance" didn't work out.

I never learned the story behind that.  I knew there was some relationship between Fluke and Philips but never any details.

Now of course Fluke, Tektronix, and Keithley are all owned by Danaher Corporation.  I have heard interesting stories about that like how the Tektronix DMM916 line of handheld multimeters was making the Fluke 97 series look bad so Tektronix received an offer that they could not refuse and discontinued them.

[tggzzz]

[lots of stuff not worth quoting]

It seems you've now resented replace arguments in an adult discussion with childish diatribe, aggressiveness and personal attacks. Although it's unfortunate (but not completely surprising), to keep the peace in this forum I will simply end the discussion with you.

Pot. Kettle. Black.

But ending this to-and-fro is the right decision.

[edavid]

I think it's a shame that "the T&M Alliance" didn't work out.

I never learned the story behind that.  I knew there was some relationship between Fluke and Philips but never any details.

http://www.fundinguniverse.com/company-histories/fluke-corporation-history/

Among Parzybok's most prolific [sic] moves was the 1993 purchase of the testing and measuring device division of N.V. Philips, the Netherlands-based electronics giant. Fluke had entered a partnership with Philips in 1987. The alliance gave Fluke new products to sell in its U.S. market, and also allowed it to begin selling its own gear through Philips distribution channels in Europe. The partnership was also responsible for the development of the ScopeMeter, which became a big seller for Fluke. Fluke finally decided to end the partnership by paying $41.8 million to simply buy the Philips division, which added about 900 employees to its payroll and roughly $125 million in annual revenues.

[Wuerstchenhund]

http://www.fundinguniverse.com/company-histories/fluke-corporation-history/

Thanks for the link. It was very interesting to read.

Among Parzybok's most prolific [sic] moves was the 1993 purchase of the testing and measuring device division of N.V. Philips, the Netherlands-based electronics giant. Fluke had entered a partnership with Philips in 1987. The alliance gave Fluke new products to sell in its U.S. market, and also allowed it to begin selling its own gear through Philips distribution channels in Europe. The partnership was also responsible for the development of the ScopeMeter, which became a big seller for Fluke. Fluke finally decided to end the partnership by paying $41.8 million to simply buy the Philips division, which added about 900 employees to its payroll and roughly $125 million in annual revenues.

It's a real shame that Fluke didn't continue Philips benchtop scopes. I guess they only really wanted the portable gear like the ScopeMeter.

[EEVblog]

You missed David Hess's important point that this is only valid for an unaliased waveform.  Since the Rigol DS1000Z doesn't have a sharp enough antialiasing filter in front of the ADC (as shown by the bandwidth measurements that people have posted), the 250MSPS sample rate is not adequate to support 100MHz bandwidth.

I knew about David's point and was ignoring it for the argument that a scope should have an adequate filter in place.
I have not seen the result for the Rigol, but ok, if it doesn't then the issue becomes relevant.

[marmad]

I am actually disappointed in the current crop of 100 MHz 250 MS/s DSOs because of their lack of ETS.

What "crop" are you referring to? The single DS1000Z model from Rigol, sold as three FW-delineated versions?

The result is shown in the time domain in the above video as what I like to call "wobbulation".  I suspect the same thing on one of the Rigol oscilloscopes with an index graded display makes for waveform thickening which is mistaken for noise.

Yes, we already know you weren't happy with the DS1000E you bought. It still doesn't explain how whatever nonlinearity and sampling clock errors you feel were present in the 5x dual-ADC-chip DS1000E series were magically transferred to the redesigned, single quad-ADC-chip DS1000Z model.

I knew about David's point and was ignoring it for the argument that a scope should have an adequate filter in place.
I have not seen the result for the Rigol, but ok, if it doesn't then the issue becomes relevant.

I'm not sure how a low-cost 100MHz DSO could have an adequate enough filter for a 125MHz Nyquist frequency. A Gaussian frequency response would only be around -5db at 125MHz, and even with a flat-response, it would only be approx. -9db at 125MHz. Perhaps the 50MHz model's frequency response is adequate. The DS1000Z series is, for all practical purposes, a 1/2 channel 50/75/100MHz DSO and a 3/4 channel (50?)25MHz DSO.

[pascal_sweden]

Isn't the BW 100 MHz for ALL 4 channels?

The digital sample rate is divided when you use more than one channel, but my understanding is that the BW is 100 MHz for ALL 4 channels.

Note that analog BW and digital sample rate are two different things.

Analog BW = 100 MHz on ALL channels, even if you use all 4 channels at once.

Digital sample rate is 1 GS/s when you use 1 channel, 512 MS/s when you use 2 channels, 250 MS/s when you use 4 channels.

For a 100 MHz BW, 250 MS/s is enough to represent the original signal as it complies with the Nyquist theorem.

The Nyquist theorem states that a signal must be sampled at a rate greater than twice the highest frequency component of the signal to accurately reconstruct the waveform; otherwise, the high-frequency content will alias at a frequency inside the spectrum of interest.

More detailed information about the relation between Bandwith, Sample Rate and the Nyquist Theorem is available here:
http://www.ni.com/white-paper/2709/en/

What can an anti-aliasing filter do to impact this? I like to read this in a formal white paper to get better understanding.