Question about analog oscilloscopes vs digital oscilloscopes

[David Hess]

Yes im trying to see if an analog scope has any use when you have a decent digital scope.

i already have a decent digital scope on its way, Im just trying to build my learning lab without having to spend too much. So if an analog scope is useless when i have a decent DSO than i can cross it out on the list so it can go towards getting other things instead.

Since you already have a DSO on the way, I would consider an analog oscilloscope superfluous unless it fulfills a specific but general need like high bandwidth.  No DSO feature can make up for lack of bandwidth when it is needed.  What is the least expensive new DSO with 350 MHz or more of bandwidth?

There have been a few times where I have used an analog oscilloscope to do things that few or no inexpensive modern DSOs could do but the situation is rare enough to be generally ignored.  Examples include reliable wideband RMS measurements, alternate triggering (some inexpensive DSOs can do this), displaying small snivets, and large range offset measurements.

Challenge accepted, i will try to find some way to make my DSO stackable. Perhaps i could put it on my server rack or even on my desktop PC case .

I have never had a problem stacking my DSOs as illustrated by this photograph although currently in that spot I have a 2232 analog/DSO stacked on a 2247A analog oscilloscope which has considerable automatic measurement capability.  Also in reference to alternate triggering which I *have* found to be useful in the past, you could always buy two separate inexpensive DSOs which is sort of what I ended up doing although not for that specific reason.

[nctnico]

Just wait until the question ends up on your plate! And no, there wasn't a simple answer for the problem at hand.

And what help using an oscilloscope in this case ?
You know the contact is arcing....You don't have a lot of solutions...choose the right relay or contacts, eliminate all LdI/dt, reduce the nominal voltage, reduce the current, use a 2 stages off circuit, ....

The question in this case was: how much can a particular relay handle and is it possible do something to make it handle more. This information was necessary to create an economically optimal design for a mass produced circuit which also needed to be absolutely reliable. Using a DSO was essential to get screenshots of what kind of voltages and currents are flowing during arcing; I literally cooked a few relays during these tests so repeatability at the limits was hard to achieve. I can't go into much details (NDA) but it was an interesting thing to measure and just using a DMM would not have been sufficient to get all the information I wanted.

BTW A differential probe is an easy way to isolate the scope's inputs from circuits misbehaving.

For me, your explanation is nonsense.

The reason is simple: because I can't give out all the details to make sense of it all. You just have to accept that from other people on a forum every now and then.

The point however is that I needed to make measurements on a non-predictable system which is where a DSO in single shot mode really shines. Capture once and study the event in detail with the help of long memory and zoom. An added bonus is the ability to make screenshots with cursor measurements and math traces to put in a report.

[Wuerstchenhund]

... which is that an analog scope (a tool you correctly said is for looking at waveforms) doesn't give you any numbers

Nonsense. OK, it doesn't spoonfeed you with digits, but the numbers are there for the moderately intelligent. Or, if you aren't sufficiently intelligent to determine the numbers, then you aren't intelligent enough to use a scope in the first place!

In short, it solely relies on user interpretation. Thanks for confirming 

Besides, I wouldn't exactly consider clinging to yesteryear's test gear a sign of intelligence 

Parallax has the benefit that it directly and visibly has to be understood. OTOH, it is very difficult to spot and correct problems with DSOs internal algorithms.

No, it isn't, at least not if you have a rudimentary understanding as to what's going on.

You acknowledge that in...

No, I don't. In fact, people were pretty quick identifying the various bugs in their Rigol (or Siglent) gear. The issue wasn't identifying the bugs, the isse was that the products in question were clearly not mature enough to be released, and the slow response of the manufacturer in fixing the bugs.

But as someone who only superficially follows the various Rigol threads I can't say how mature the measurement functionality in Rigol scopes is, but personally I wouldn't touch a product, which at the end of the day is supposed to be a test instrument you want to rely on for your measurements, is thrown to market with such a large number of bugs and with the hardware idiocracies listed in the "Yaigol" thread. But I understand that the low price tag can be very tempting.

So, it appears that your "DSOs are always best" kind of statements are implicitly presuming that the DSO is actually a high-end professional-grade expensive DSO from a reputable manufacturer.

No, my "these days investing money in anything else than a DSO is silly" kind of statements are implicitly presuming that the DSO isn't made by some crappy B-brand which regulary releases products full of bugs but by a reputable big brand. That neither implies "high-end" nor "expensive", unless you consider entry-level scopes like the R&S HMO1200 or Keysight DSOX2000 "high-end". Or an old HP 54600 Series scope for that matter, which despite being roughly 20yrs old, still tops most analog scopes. And these scopes regularly go for a couple of hundred dollars these days.

Also, let's talk about prices. That Tek 475, which actually *was* high-end back then, did cost around $1700 in the mid-70's, that's roughly $10k in today's money. The broad base of scopes however weren't 200Mhz high-end scopes, these were mostly simpler models in the 20Mhz to 60Mhz range, with generally inferior specs. A lot of them were pretty shit even when they came out, like some of the USSR made low end scopes that were sold under various names around the globe, or (as a younger example) the various analog scopes made by GW Instek, Atten and other outfits.

What that means is that even what once was a high-end analog scope has already difficulties keeping up with a somewhat decent entry-level DSO. Now imagine how much worse the other crop of analog boat anchors compare against them.

I've seen a lot of scopes throughout the decades, and even back then in the analog scope days not everything was great, and these old scopes didn't get better now that they are antiques.

In that case I probably wouldn't argue. Neither would I argue that a new Rolls Royce isn't better than a second-hand Ford Escort.

But a Ford Escort will probably get you to your destination just as well - and much better if you can't afford a Rolls Royce! That's the kind of balance and nuance missing from your statements.

Just that the equivalent to the analog scope isn't a Ford Escort, it's pretty much horse and buggy, and (like the analog scope) it lacks all technological achievements made that are available in its successor, i.e. powerful brakes, heating, crash-safe body, occupant restraining systems, rubber tires, electrical lighting, cruise control, stability and tracktion control, on-board entertainment systems, satnav, whatever (like the analog scope's lack of even the most basic measurement and analysis capabilities, or even very basic stuff like proper storage). And I'm not sure horse and buggy can generally be seen as a viable means of transport these days.

The Escort equivalent would be an earlier DSO like the HP 54500 Series, because like the Escort it can pretty much do all the basics that are expected from a car today, it just comes with inferior performance (acceleration, economy, crash performance) and lacks most of the options that are available in newer cars (ESP, ABS even depending on the model, side airbags, pre-crahs warning and so on), and it looks old.

But, and this is why car analogies are seldomly appropriate, I'm not sure a new DSOX2000 can be considered the equivalent of a Rolls Royce (that would be something like the Keysight DSO-Z or the LeCroy LabMaster, both scopes that come with a 6 digit price tag). A modern day entry level DSO is more like cars like the Skoda Fabia, VW Up or Dacia Sandero.

Anyways, I know we're unlikely to agree on this, but at the end of the day that doesn't matter. The fact is that the analog scope is dead, period. All major manufacturers have stopped producing them, most of them a long time ago (Iwatsu held on for longest due to contractual requirements to keep them available). There hasn't been any new development in analog scopes in more than two decades, and there won't be any new models coming out. It's dead, period. Reality is that the world of engineering has long moved on to DSOs, which in on way or another have more than fully replaced DSO throughout the various industries, and this isn't changed just because some few hold-outs still cling to their old boat anchors. That is how it is.

Considering that analog scopes are dead, it's silly to buy one for anything else than curiosity if you have a somewhat decent DSO already. As for the common example of the broke student/hobbyist, an analog scope can still be fine if it's free/very cheap (<$100 for better models), however, considering that older DSOs like the HP 54600 Series can often be found for a couple of hundred dollars, saving up a bit longer is in my opinion the better choice, rather than investing even an lower amount in a scope that has been obsolete 2 decades ago.

[tggzzz]

I'll ignore your strawman arguments and where you stretch analogies beyond reasonable limits, and concentrate on...

...
Considering that analog scopes are dead, it's silly to buy one for anything else than curiosity if you have a somewhat decent DSO already. As for the common example of the broke student/hobbyist, an analog scope can still be fine if it's free/very cheap (<$100 for better models), however, considering that older DSOs like the HP 54600 Series can often be found for a couple of hundred dollars, saving up a bit longer is in my opinion the better choice, rather than investing even an lower amount in a scope that has been obsolete 2 decades ago.

... and so eventually you, rather grudgingly, come to some balance and implicitly acknowledge the points other people have being making.

I have used many top-of-the range HP digitising scopes (and other equipment) in the past, and have worked with the person that designed many of their digitisers, Tom Hornak (incidentally, his Czech name is usually misspelled: it should have an upside down circumflex on the n). I do have a clue in this area.

Being retired, I currently use a 350MHz Tek 485 and a 60MHz HP54621 - of the type you mention. I am familiar with both.

IMNSHO, the HP54621 is useless for digital circuits - with modern jellybean logic its bandwidth is simply too low to assess signal integrity and timing violations. OTOH the 485 is OK, and even a 475 is adequate. Nor would I use any scope to debug logical operation? No, a scope is th ewrong tool for that. Once I had assured signal integrity, I flip to debugging a logic signal using logic-based tools.

For mosying around an unfamiliar circuit when debugging, the 485 is faster.  In addition, the 54621 can be downright misleading in some circumstances, for example checking signal envelopes.

OTOH, the HP54621 is better at catching turn-on transients - but even my Analog Discovery is adequate for that!

Now, let me dispel some of your assumptions and implications...

We are not offering opinions to professional engineers; they have better sources of information and ought to be able to assess their requirements without our help. We are offering opinions to beginners and amateurs, and we should act accordingly.

Most analogue scopes you see around (e.g. 10/20MHz TV scope) aren't worth more than £20, if that. I've never said or implied otherwise.  I've advised several people against buying such things. But I know several people that are reluctant to afford even £50 and are only interested in audio and mechatronics. For them such a scope may be the best way to contiune with electronics - especially when you consider that if they have to wait too long, they may well lose interest in the subject.

However, a decent working analogue scope, e.g. Tek 465 or later, are available for £100, and they are worth having. And IMNSHO they are more useful than a low-end DSO.

[rstofer]

Now, let me dispel some of your assumptions and implications...

We are not offering opinions to professional engineers; they have better sources of information and ought to be able to assess their requirements without our help. We are offering opinions to beginners and amateurs, and we should act accordingly.

And in the case of starving students, "saving up" isn't an alternative.  If the scope is going to be of benefit, it needs to be timely.  There will be opportunities down the road to upgrade.
Yes, I know that colleges have labs - I went there once.  It's just so much more convenient to experiment when time is available - usually the middle of the night, as I recall...

[nctnico]

IMNSHO, the HP54621 is useless for digital circuits - with modern jellybean logic its bandwidth is simply too low to assess signal integrity and timing violations.

You keep hammering on this but for the parts to achieve that performance they'll need very good decoupling and driving ideally terminated PCB traces. And even if those conditions are met you'll also need to have the right probes to measure the effects. But before all that you have to ask whether all of that is actually necessary and quite frankly: I don't think so. If you put a gate here en there as auxilary logic to drive a led or something like that there isn't going to be any trouble. Things are different if you have a large board with 100 or so logic chips but I don't think such boards are designed for commercial purposes nowadays.

[tggzzz]

Now, let me dispel some of your assumptions and implications...

We are not offering opinions to professional engineers; they have better sources of information and ought to be able to assess their requirements without our help. We are offering opinions to beginners and amateurs, and we should act accordingly.

And in the case of starving students, "saving up" isn't an alternative.  If the scope is going to be of benefit, it needs to be timely.  There will be opportunities down the road to upgrade.
Yes, I know that colleges have labs - I went there once.  It's just so much more convenient to experiment when time is available - usually the middle of the night, as I recall...

Precisely. I'm annoyed I forgot to mention timeliness.

[tggzzz]

IMNSHO, the HP54621 is useless for digital circuits - with modern jellybean logic its bandwidth is simply too low to assess signal integrity and timing violations.

You keep hammering on this but for the parts to achieve that performance they'll need very good decoupling and driving ideally terminated PCB traces. And even if those conditions are met you'll also need to have the right probes to measure the effects. But before all that you have to ask whether all of that is actually necessary and quite frankly: I don't think so. If you put a gate here en there as auxilary logic to drive a led or something like that there isn't going to be any trouble. Things are different if you have a large board with 100 or so logic chips but I don't think such boards are designed for commercial purposes nowadays.

I "keep hammering" on about it because I keep seeing people fail in this way - as I said, the last time was ~22hours ago!

Of course appropriate probes are needed. What's new? You also need an appropriate scope.
Of course decoupling ought to be "good". What's new? But someone has to verify that the everything is "good enough" - and not everybody gets it right, especially beginners.
Of course 100 chip systems are few and far between. What's new? But nowadays many systems are made from multiple boards cobbled together, with all sorts of ensuing problems.

I've even seen people say "there's no problem" (on a 60MHz scope), only to see a very visible problem (a static hazard) on a 350MHz scope. Bad design? Sure. So what; that's why we have scopes.

[rsjsouza]

IMNSHO, the HP54621 is useless for digital circuits - with modern jellybean logic its bandwidth is simply too low to assess signal integrity and timing violations.

You keep hammering on this but for the parts to achieve that performance they'll need very good decoupling and driving ideally terminated PCB traces. And even if those conditions are met you'll also need to have the right probes to measure the effects. But before all that you have to ask whether all of that is actually necessary and quite frankly: I don't think so. If you put a gate here en there as auxilary logic to drive a led or something like that there isn't going to be any trouble. Things are different if you have a large board with 100 or so logic chips but I don't think such boards are designed for commercial purposes nowadays.

I "keep hammering" on about it because I keep seeing people fail in this way - as I said, the last time was ~22hours ago!

If you are referring to me, I agree with nctnico. As I said before, IME the sub nanosecond measurements you mentioned a few posts ago are not very relevant for hobbyist-level applications. To get an accurate measurement of the the slew rates you talked about you will definitely need something much more expensive than entry-level oscilloscopes and alligator clip probes - again, all this is applicable to hobbyist-level projects. Peace.

[tggzzz]

If you are referring to me, ...

Only if you were building a CNC machine in Hackspace  yesterday!

No,"seeing a person make the mistake" is "in the flesh", not "virtual", unfortunately.

[rsjsouza]

If you are referring to me, ...

Only if you were building a CNC machine in Hackspace  yesterday!

No,"seeing a person make the mistake" is "in the flesh", not "virtual", unfortunately.

Ah, ok. Our discussion happened at about the same timespan you mentioned and our experiences diverge... 

[Martin.M]

As I said before, IME the sub nanosecond measurements you mentioned a few posts ago are not very relevant for hobbyist-level applications. To get an accurate measurement of the the slew rates you talked about you will definitely need something much more expensive than entry-level oscilloscopes and alligator clip probes - again, all this is applicable to hobbyist-level projects. Peace.

hello friends,

excuse, I am the hobbyist here  a sheet metal worker, collecting all kind of scopes, mostly old and very old systems.

In my opinion the testing of sub nanoseconds slew rates are requiering the knowlegde what I am doing there. This can be high upside from the knowledge of a hobbyist.
I would select a 7104, plugged with 7A29 (this is a Tek oldie of 1GHz, I will store my shot with a camera) and a nice Fet probe to do that, but when I took the hook on a near to trustful rectangle generator (284) to see what my scope can do for me I see sometimes that is not a hundret per cent rectangle. There is some effects of reflectometry as a part of my signal. So I need at first the best possible setup to make shure that I will see what I will get.
The knowledge of this effects is imho upside from hobbyists, you as EE can understand what is "true" on the display and what is "false". A fast DSO will show us the same, I am pretty shure in that 

So I think the most hobbyists will not get a good measurement of slew rates, also when they use a high rated DSO. So it`s very good that this tests are not very relevant for hobbyist-level applications.
Today it is possible for a hobbyist to buy a 7104, or the 7844 Dual beam, at the time of this systems is was a dream, they was the high laboratory class and ultra expensive.

greetings
Martin

[tggzzz]

A DSO and analogue scope with the same bandwidth should indeed show the same signal.

You can find a good old 350MHz analogue scope for EUR280 (Tek485). An equivalent DSO is significantly more expensive. That's an advantage to the analogue scope, but of course a DSO would have different advantages. Whether the advantages are benefits depends on individual's requirements and position. My objection is to people that ignore (and in some cases deny) the advantage and benefit, and presume everybody else is in the same position as they are. Imbalanced..

I'll note that for many purposes, particularly with jellybean digital logic, a "low impedance Z0" probe is technically as good as a FET probe - and is more robust and probably cheaper. The major exception is with differential digital signals where a differential FET probe is necessary.

[nctnico]

A DSO and analogue scope with the same bandwidth should indeed show the same signal.

You can find a good old 350MHz analogue scope for EUR280 (Tek485). An equivalent DSO is significantly more expensive.

No, you can buy a 500MHz DSO (which works fine but needs some cosmetic fixing) for less than that and the DSO will be much newer. Been there, done that.

[GK]

A DSO and analogue scope with the same bandwidth should indeed show the same signal.

If the DSO has a high enough sample rate. On my bench at work sits a Lecroy LT344, which was a kinda hot thing when it was bought brand new back in 2003 (I need to start pestering for an upgrade, but that is a different topic). Far from the pinnacle of DSO technology, but a good and well optioned performer in its price bracket. Its analogue bandwidth is 500MHz but I sure as hell wouldn't shift my equal-bandwidth 7904 off my home bench if I managed to pick up an LT344 or equivalent for a good price. The LT344 has a sampling rate of 500MSa. In real-time sampling mode that's a whopping 5 sampling points at 1nS/div. Suffice to say it's RTS waveform "fidelity" (for want of a better term) at the highest sweep rates hardly compares to that of my 7904. Yes there is ETS/RIS but they are workarounds with other drawbacks/limitations. It's takes a pretty high-end DSO to have a RTS performance that won't potentially leave one wanting over the high frequency waveform reproduction capabilities of a 7904, let alone a 1GHz-bandwidth 7104. You should see the stupid prices test equipment re-sellers down here still try to flog in-house-calibrated secondhand DSOs like the LT344 for.

[tggzzz]

A DSO and analogue scope with the same bandwidth should indeed show the same signal.

If the DSO has a high enough sample rate. On my bench at work sits a Lecroy LT344, which was a kinda hot thing when it was bought brand new back in 2003 (I need to start pestering for an upgrade, but that is a different topic). Far from the pinnacle of DSO technology, but a good and well optioned performer in its price bracket. Its analogue bandwidth is 500MHz but I sure as hell wouldn't shift my equal-bandwidth 7904 off my home bench if I managed to pick up an LT344 or equivalent for a good price. The LT344 has a sampling rate of 500MSa. In real-time sampling mode that's a whopping 5 sampling points at 1nS/div. Suffice to say it's RTS waveform "fidelity" (for want of a better term) at the highest sweep rates hardly compares to that of my 7904. Yes there is ETS/RIS but they are workarounds with other drawbacks/limitations. It's takes a pretty high-end DSO to have a RTS performance that won't potentially leave one wanting over the high frequency waveform reproduction capabilities of a 7904, let alone a 1GHz-bandwidth 7104. You should see the stupid prices test equipment re-sellers down here still try to flog in-house-calibrated secondhand DSOs like the LT344 for.

I'm not going to argue with any of that! But some will, I suspect.

[System Error Message]

so other than the occasional low priced 2nd hand bargains i supposed analog scopes arent much use other than a bargain for getting something of high frequency.

Now that i have my DSO it is very decent and being able to freeze/store and than take measurements with the cursor including other tools it has i think makes it far better than an analog scope however i think the only issue with the DSO is sampling rate and aliasing which analog scopes dont do. Being able to connect to PC for further analysis is also a good thing because that 800x600 resolution doesnt compare to 1080p or 4K. Not sure if one would argue that CROs have the resolution of a photon . The features on a DSO are quite impressive.

So im guessing when comparing an ASO do DSO of equal frequency the only use of an ASO is measuring signals where sampling is an issue?

So other than the bandwidth what is the limiting factor of an ASO in picking up signals? Is it that it has to be processed to be shown on the screen making some sort of arbitary sampling rate or does the signal get converted into photons directly in order to give a theoretically infinite sampling?

[Wuerstchenhund]

On my bench at work sits a Lecroy LT344, which was a kinda hot thing when it was bought brand new back in 2003

No, the WaveRunner LT344 wasn't a hot thing in 2003. Not even kinda of 

The WaveRunner LT344 came out in 1998, and in 2003 it had already been succeeded by two generations (WR2LT, WR6000). Most of the original WaveRunner LT scopes were bought for non-EE tasks like physics or mechanical engineering where the low sample rate wasn't an issue, and IIRC production of that series stopped at the end of 2001 or early 2002 as it didn't sell becasue the successor (WR2LT) offered much better specs (sample rates from 1GSa/s to 4GSa/s) and more performance for the same price. The WR2LT came out in 2001 btw.

In 2003, the LT344's successor WR2LT was already in its last production cycle as the WR2LT's successor WaveRunner 6000 came out the same year. And the WR6000 (a Windows2000 based X-Stream scope with sample rates  of 2.5Gsa to 5GSa/s) was a huge step up from the older WR2LT.

If you really bought a LT344 in 2003 then you simply bough an obsolete out-of-production scope.

It's takes a pretty high-end DSO to have a RTS performance that won't potentially leave one wanting over the high frequency waveform reproduction capabilities of a 7904, let alone a 1GHz-bandwidth 7104.

No, not really. Not even back in 2003. I agree that the LT344 pales in comparison to the Tek 7900 in terms of real time performance, but as I said the LT344 is pretty old technology (1998 vintage), and even back in 1998 it wasn't exactly the epithome of digital scopes, not even from LeCroy (the WRLT was pretty much LeCroy's new mid-range series back then, they also had their LT Series of high-end scopes with up to 1GHz BW and 8GSa/s sample rate). However, a WR2LT already offered much better RTS performance than the old WRLT, and in 2003 the actual a hot thing would have been a WaveRunner 6000 (and the 7904 wouldn't look exactly great against a 1GHz WR6100).

And then there's the thing that 1GHz is pretty much the upper BW limit for analog scopes, while even back in 2003 you could get fast real-time scopes with 6GHz and sample rates of 20GSa/s.

And it didn't take that long until DSOs even exceeded the waveform rates of analog scopes.

You should see the stupid prices test equipment re-sellers down here still try to flog in-house-calibrated secondhand DSOs like the LT344 for.

Yes, but that's pretty much true for all obsolete pieces of test gear. Just have a look on ebay at the sometimes mind-blowing asking prices for gear that these days can easily replaced by newer and cheaper gear. But used T&M brokers live in their own bubble, one where competition and technical progress (which regularly brings us better performance at lower prices) doesn't exist and where old crap gains in value like good wine. They also don't seem to mind that they're sitting on their stock that is pretty much depeciating its already low value every day it just sits there. The rules of economics have already sorted out many of them, and in the days of ebay many have realized that these days it's better to sell stuff fast and cheap with little profit margins than hang onto it for years to wait for that one buyer who really needs a certain piece of kit and who's willing to pay hyperinflated prices.

[GK]

On my bench at work sits a Lecroy LT344, which was a kinda hot thing when it was bought brand new back in 2003

No, the WaveRunner LT344 wasn't a hot thing in 2003.

The WaveRunner LT344 came out in 1998, and in 2003 it had already been succeeded by two generations (WR2LT, WR6000). Most of the original WaveRunner LT scopes were bought for non-EE tasks like physics or mechanical engineering where the low sample rate wasn't an issue, and IIRC production of that series stopped at the end of 2001 or early 2002 as it didn't sell becasue the successor (WR2LT) offered much better specs (sample rates from 1GSa/s to 4GSa/s) and more performance for the same price. The WR2LT came out in 2001 btw.

In 2003, the LT344's successor WR2LT was already in its last production cycle as the WR2LT's successor WaveRunner 6000 came out the same year. And the WR6000 (a Windows2000 based X-Stream scope with sample rates  of 2.5Gsa to 5GSa/s) was a huge step up from the older WR2LT.

If you really bought a LT344 in 2003 then you simply bough an obsolete out-of-production scope.

I wrote "a kinda hot thing". And you appear to have missed my following sentence:

"Far from the pinnacle of DSO technology, but a good and well optioned performer in its price bracket".

It's takes a pretty high-end DSO to have a RTS performance that won't potentially leave one wanting over the high frequency waveform reproduction capabilities of a 7904, let alone a 1GHz-bandwidth 7104.

No, not really. Not even back in 2003. I agree that the LT344 pales in comparison to the Tek 7900 in terms of real time performance, but as I said the LT344 is pretty old technology (1998 vintage), and even back in 1998 it wasn't exactly the epithome of digital scopes, not even from LeCroy (the WRLT was pretty much LeCroy's new mid-range series back then, they also had their LT Series of high-end scopes with up to 1GHz BW and 8GSa/s sample rate). However, a WR2LT already offered much better RTS performance than the old WRLT, and in 2003 the actual a hot thing would have been a WaveRunner 6000 (and the 7904 wouldn't look exactly great against a 1GHz WR6100).

And then there's the thing that 1GHz is pretty much the upper BW limit for analog scopes, while even back in 2003 you could get fast real-time scopes with 6GHz and sample rates of 20GSa/s.

And it didn't take that long until DSOs even exceeded the waveform rates of analog scopes.

Any DSO with a 1GHz analogue bandwidth and an 8GSa/s sample rate is pretty damn "high-end" from a general hobbyist/home lab owners perspective (which was the context of the discussion), be it a 13 year old design or a current production model. 

But this is taking the discussion off track. I simply used the LT344 vs the 7904 to raise a technical contention with tggzzz's assertion that "A DSO and analogue scope with the same bandwidth should indeed show the same signal" as these are "same bandwidth" scopes that I handily have first hand experience with. It's nice that you "agree that the LT344 pales in comparison to the Tek 7900 in terms of real time performance" but I don't know why the contribution in addition to that is necessary.

And yes we really did buy a LT344 "brand new" in 2003. If it was already out of production for a year or two as you claim then it would have been stock that the distributor was keen to shift, which is likely why I recall the favorable price vs performance standpoint. "Obsolete" or not it satisfied our requirements.

You should see the stupid prices test equipment re-sellers down here still try to flog in-house-calibrated secondhand DSOs like the LT344 for.

Yes, but that's pretty much true for all obsolete pieces of test gear. Just have a look on ebay at the sometimes mind-blowing asking prices for gear that these days can easily replaced by newer and cheaper gear. But used T&M brokers live in their own bubble, one where competition and technical progress (which regularly brings us better performance at lower prices) doesn't exist and where old crap gains in value like good wine. They also don't seem to mind that they're sitting on their stock that is pretty much depeciating its already low value every day it just sits there. The rules of economics have already sorted out many of them, and in the days of ebay many have realized that these days it's better to sell stuff fast and cheap with little profit margins than hang onto it for years to wait for that one buyer who really needs a certain piece of kit and who's willing to pay hyperinflated prices.

And your point is? I haven't seen any sellers down here trying to flog 7904's for $5000 plus. I am also not aware of any affordable (compared to what I payed for my 7904 and the wrecks to serve as a life-time supply of spare unobtanium) current production 1GHzBW/8MSa/s entry-level DSOs.   

[Martin.M]

a 7904 (empty frame without plugins) is good for 100$ / €.
If you need spare parts it is a good idea to ask in the Tek collectors communitys if they have, like the "TekScopes" at yahoo or my wellenkino.de
We collect not only scopes, also the little spare bags and boxes from oregon.

greetings
Martin

[joeqsmith]

The roll off may not be at all the same.  Long gone are the days of 0.35.... 

Quick Google search "rise time bandwidth modern DSO filters", first hit:

http://www.edn.com/design/test-and-measurement/4340453/Relating-wideband-DSO-rise-time-to-bandwidth-Lose-the-0-35-

A DSO and analogue scope with the same bandwidth should indeed show the same signal.

If the DSO has a high enough sample rate.

I'm not going to argue with any of that! But some will, I suspect.

[joeqsmith]

Besides, I wouldn't exactly consider clinging to yesteryear's test gear a sign of intelligence 

There is no sign of intelligence here...  All of my test gear is from yesteryear.  My newest scope is 10 years old.   

But wait.  Don't you own a 7300 and 64xi?   I think you need to send those old relics to me or people are going to start thinking you too are less intelligent!    

[GK]

Well a high enough sample rate is just one requirement; probably (actually not really probably) a great deal more potentially significant than whether or not the rise-time/bandwidth relationship between two scopes with the same specified -3dB bandwidth differs by 0.35/RT to 0.4 or 0.5/RT 



[upsidedownquoting]

The roll off may not be at all the same.  Long gone are the days of 0.35.... 

Quick Google search "rise time bandwidth modern DSO filters", first hit:

http://www.edn.com/design/test-and-measurement/4340453/Relating-wideband-DSO-rise-time-to-bandwidth-Lose-the-0-35-

A DSO and analogue scope with the same bandwidth should indeed show the same signal.

If the DSO has a high enough sample rate.

I'm not going to argue with any of that! But some will, I suspect.

[/upsidedownquoting]

[tggzzz]

A DSO and analogue scope with the same bandwidth should indeed show the same signal.

You can find a good old 350MHz analogue scope for EUR280 (Tek485). An equivalent DSO is significantly more expensive.

No, you can buy a 500MHz DSO (which works fine but needs some cosmetic fixing) for less than that and the DSO will be much newer. Been there, done that.

Which ones, out of curiosity?

[Wuerstchenhund]

I wrote "a kinda hot thing". And you appear to have missed my following sentence:

"Far from the pinnacle of DSO technology, but a good and well optioned performer in its price bracket".

But that's the point, in terms of real-time performance it simply wasn't. Not in 1998, and a lot less in 2003.

Any DSO with a 1GHz analogue bandwidth and an 8GSa/s sample rate is pretty damn "high-end" from a general hobbyist/home lab owners perspective (which was the context of the discussion), be it a 13 year old design or a current production model. 

I gess the various owners of LeCroy LC Series scopes in this group would beg to differ, and these scopes often go for way below the $1k mark (nothing for the broke student, but well into hobbyist territory, which also are pretty much the only people that buy these old scopes today).

But this is taking the discussion off track. I simply used the LT344 vs the 7904 to raise a technical contention with tggzzz's assertion that "A DSO and analogue scope with the same bandwidth should indeed show the same signal" as these are "same bandwidth" scopes that I handily have first hand experience with.

But they aren't the same BW scopes. The LT344 doesn't have 500MHz real-time bandwidth, that's merely the analog bandwidth of the front end. A DSO's usable BW however is determined by not just the FE but also the acquisition system BW, and for the LT344 in real-time mode the latter is <250MHz.

And yes we really did buy a LT344 "brand new" in 2003. If it was already out of production for a year or two as you claim then it would have been stock that the distributor was keen to shift, which is likely why I recall the favorable price vs performance standpoint. "Obsolete" or not it satisfied our requirements.

That is well and fine, but again, I was just addressing your "Far from the pinnacle of DSO technology, but a good and well optioned performer in its price bracket" statement. In terms of real-time performance for a 500MHz DSO, the LT344 was sub-par already the day it came out. Which makes a comparison as real-time scope against a 500MHz+ analog scope rather pointless.