A newbie needs some help with my first DSO

[Fungus]

Some people expect that...

I think MOST people would expect that. That's why Dave was so surprised.

It is not without reason that the default setting for all scopes that I know across brands is “auto” as far as memory management is concerned.
And in this setting you can't look “behind” the screen and I bet 99% of the time you don't need to, otherwise it wouldn't be a default setting.

 

Here's a Rigol in "auto" mode. Visible/hidden memory is shown in a special indicator at the top of the window.

As you can see, the memory being captured goes far beyond the visible screen, both before and after. I'm guessing about seven screens either way in this example.

[Someone]

OTOH, there is no such setting on Keysight DSOs. You always get full memory for the last acquisition.

Except not all Keysight scopes do that, which has been tested and found to not occur in your magic corner case. This is all brand and model specific.

It cannot be simplified down to every/all/universal. Your claims of "industry standard" are a joke when it's not the default even on the scopes that support it.

[Someone]

I am planning to test and experiment with the effectiveness of basic R/C de-bounce filters on various switches.

Sounds like a perfect example of where the Siglent way is going to be a pain in the ass. You'll be forever zooming in to look at detail then forgetting to zoom out again to prepare the 'scope for the next capture.

How many times do you want to post this crap? You're factually incorrect.

So why keep pushing a beginner, who asked for some clarity on the best way to use their scope, away from the way that scope is designed to be used?

Actually the question was:

The link to Dave's video was actually very helpful for me in making sense of what memory depth is. I confess that i was assuming it would work more like the Keysight, which is part of why I was having trouble understanding - I hit "stop," but there was no data past the end of the screen. I think I am now understanding that, for the Siglent and a few others, what is captured is only what fits on the screen at a given moment - thus, one must zoom out, capture, then zoom in to see detail. Right? Maybe?

So there you have the proof you are so friggin' wrong in your thinking it isn't even funny, right in your face: people expect a DSO to capture beyond the screen because they find it useful. It is simple as that, like it or not. And yet you keep on going trying to push your limited view onto everyone. FFS 

The best way is what gets people to results in a way which suits them best. Imagine somebody coming along insisting you stand behind your desk while you prefer to sit on a chair behind your desk.

You're taking the forced solution you pushed as a "win". I see that as a failure to educate the OP in the diversity of ways they could achieve the same result. Adding confusion to their mind instead of helping them understand the underlying concepts, which is where the question started.

There is no stopping and starting again. If you want to capture data around the trigger/magnified window then of course you have to set that before starting. With either method. So your communication has as I keep saying only added confusion and noise, no answered their honest questions.

Post truth world.

[awakephd]

FWIW, I would hate for anyone to get too excited about what I might or might not have expected. I was expecting the waveform capture to start at the trigger point, so discovering that the scope captures all sorts of pre-trigger data was a big surprise. A very welcome, big surprise!

[Someone]

Here's a Rigol in "auto" mode. Visible/hidden memory is shown in a special indicator at the top of the window.

Here is a different Rigol model that in Auto mode keeps the memory depth equal to the visible window:

So, it is model specific and people need to learn how their specific unit works. Industry standard my foot.

[Fungus]

FWIW, I would hate for anyone to get too excited about what I might or might not have expected. I was expecting the waveform capture to start at the trigger point, so discovering that the scope captures all sorts of pre-trigger data was a big surprise. A very welcome, big surprise!

The trigger point is in the center of the buffer.

[tautech]

FWIW, I would hate for anyone to get too excited about what I might or might not have expected. I was expecting the waveform capture to start at the trigger point, so discovering that the scope captures all sorts of pre-trigger data was a big surprise. A very welcome, big surprise!

The trigger point is in the center of the buffer.

And where if you place H Pos on the first graticule ?

[Fungus]

So, it is model specific and people need to learn how their specific unit works. Industry standard my foot.

I don't have mine any more but that's not how I remember my DS1054Z working.

Anybody still have theirs...?

The Siglent refuses to capture outside the screen even when you set the memory depth manually to something much larger than the display.

[Someone]

So, it is model specific and people need to learn how their specific unit works. Industry standard my foot.

The Siglent refuses to capture outside the screen even when you set the memory depth manually to something much larger than the display.

So, lots of other scopes do this too. You just set the window to your preferred length and position then leave the zoom window open on the bit you want to see.

At this point you're arguing about the difference between the presented Rigol example with its large boarders, vs the Siglent which takes up around 1/3 of the screen.

So it's back to you taking preference to not see any content of the long capture unless you stop, vs others happy to see some detail of the long capture. But instead of just saying that its some convoluted mess about how CANT DO THING.

Why get emotional about it? Add personification to a scope? "The Siglent refuses" 

Should I bomb any mention of a Rigol scope with your example of a scope that cant keep the memory depth to only the visible screen? It cant do some things other scoopes can, but its not some major issues that greatly limits the use of the scope.

[2N3055]

So, it is model specific and people need to learn how their specific unit works. Industry standard my foot.

I don't have mine any more but that's not how I remember my DS1054Z working.

Anybody still have theirs...?

The Siglent refuses to capture outside the screen even when you set the memory depth manually to something much larger than the display.

Why are you claiming things that are not true?
Where did you get this information?

If you have Siglent scope that has manual memory control, it captures it same as R&S or any other scope that has it. You get trigger, and it captures exact amount you set it for.

But that beside the point, claiming that it is a good thing that scope always captures fixed amount of data instead of only what you need is simply laughable.

Old digital scopes had manual fixed memory length and you always had to twiddle manually and remember and think what you set and what is optimal.
Always long memory decreases number of triggers/s because it always captures lots of data. It slows down a scope, measurements, everything.
But OTOH, short memory forces scope to drop sampling rate.
And it frustrated people to no end. Manual memory mode is one of the main reason why people hated early DSO. If you had it on short memory it would alias all the time. If you set it to long memory it was SLOOW... Only few triggers per second... Wonder why...hmmm...

So after some time, scopes started coming with Auto memory management mode.
That one would automatically set memory length as you go through time base. Shorter memory on short time bases, longer as you go to longer timebases to keep sampling rate up (to avoid aliasing). As scope would get to max memory, then sample rate would eventually be dropped.
And everybody liked it because they were fed up with constantly twiddling memory settings.

This Auto mode exist on Siglent and Rigol. And Tek. And R&S and Micsig.
And manual mode works exactly the same on those too. Some Siglents have, some don't have manual memory mode. Nobody uses it anyways. It is a pain for actual use..

Difference (that confuses people) is that some scopes, when automatically set memory length, will have only few fixed memory lengths, and some will have an exact length for each timebase individually.

For instance Keysight 3000T. It has Auto memory mode. People keep repeating "it captures outside the screen!".  It actually does not....
It uses a a trick when you stop the scope.
Take for an example a burst of data happening every 5 seconds.
You set it in Run mode, Normal trigger mode. It waits for signal.
Burst comes in. Capture.
You press Stop.
If you change timebase now, there is nothing outside the screen. WHAT?
BBButt Dave said there is....
Well. Dave is (partially?) wrong. It is not his fault (well he could have researched better before going on record..) but he was tricked.

See, MSOX3000T (and all other Megazooms IV) captures only what is on the screen in Run mode. But when you press Stop, it quickly reconfigures acquisition engine, and then initiates another trigger and captures ±200µ around the trigger. Unless you press Single in which case it will capture ±400µ around the trigger. That deletes last capture (one that you were looking into that prompted you to press Stop). If you are at 20µs/div or shorter. Unless you are at slower timebase in which case it doesn't. Or maybe. Or it does something else.

So it is a crapshoot. You might get something, or not, or maybe. If you do, it is going to be some random time. And only at fast timebases, 20µs/div and faster. To a serious person, that is not something to rely on. And the capture on the screen is not the one that you were looking at when you pressed Stop. That one was deleted and you get a different new one. Unless it times out (because signal is rare) in which case you get only what was on screen. So basically it is a lottery..

Only sure thing that will guarantee you will capture x amount of time around the trigger is to set the scope to do explicitly. With setting pretrigger time and post trigger time. Simplest, quickest and most intuitive way is to set it that scope shows the interval in question on the screen. It is simple. Like on analog scope. If it is not on the screen, you are not looking at it and you cannot see it.
It works on any and every scope including old CRT ones. And on digital scope, if you Stop it at that moment, you can than magnify portion of detail, by using timebase, Zoom, multiple windows or whatever scope in question supports and whatever is useful to you.


Everything else is mental acrobatics, or some special (forced?) corner case.

[Electro Fan]

2N3055, that ^ is a helpful post - all of it, but especially the discussion of how the Megazooms IV operate.  Thx

[Someone]

See, MSOX3000T (and all other Megazooms IV) captures only what is on the screen in Run mode.

That was also covered in the mega thread:
https://www.eevblog.com/forum/testgear/oscilloscope-zoom-out-quirk/msg3059524/#msg3059524
By you! Same posters present in that conversation, but come back here with lies misinformation.

[newbrain]

Only sure thing that will guarantee you will capture x amount of time around the trigger is to set the scope to do explicitly.

I recently sold my Rigol DS1054Z and got me a Siglent SDS804X-HD.
The first time I tried to "zoom out" (more precisely, increase the timebase) on a stopped capture I thought "Oh, nothing outside the screen, I'll set a larger memory".
Did that, and, as we know, no dice.
I RTFMed. Always on History mode - mind blown.

Literally 10 minutes after getting the scope, I adapted to the different WoW; now I would really reluctantly go back to the old one, if I had to renounce history and zoom mode versatility.
But to each their own - at least until facts are correctly explained.

[Fungus]

The Siglent refuses to capture outside the screen even when you set the memory depth manually to something much larger than the display.

Why are you claiming things that are not true?
Where did you get this information?

[2N3055]

The Siglent refuses to capture outside the screen even when you set the memory depth manually to something much larger than the display.

Why are you claiming things that are not true?
Where did you get this information?

.....when you set the memory depth manually......

[nctnico]

The Siglent refuses to capture outside the screen even when you set the memory depth manually to something much larger than the display.

In Siglent's defense: the more recent models do support setting the memory length regardless it is more than needed to fill the display. Siglent finally woke up after Dave's video. So you'd have to check the model and firmware version.

[tautech]

The Siglent refuses to capture outside the screen even when you set the memory depth manually to something much larger than the display.

Why are you claiming things that are not true?
Where did you get this information?

 

[Martin72]

Siglent finally woke up after Dave's video.

Yes, exactly, because a forum operator had made a video and a handful of forum users blew the same horn, saying that it was really important to be able to do that, so Siglent woke up and gave the few people what they wanted.
I would think about that again.
Perhaps look into when and with which models this is possible and what these models have in common.
A little hint, this has something to do with a (math) function that did not previously exist in Siglent scopes.

[nctnico]

Siglent finally woke up after Dave's video.

Yes, exactly, because a forum operator had made a video and a handful of forum users blew the same horn, saying that it was really important to be able to do that, so Siglent woke up and gave the few people what they wanted.
I would think about that again.

You may say it caters to a few people but then explain me this: All DSOs from many different brands I have owned and many which I used, supported capturing beyond what is needed to fill the screen (including the Siglent SDS2000 with the initial firmware but except Lecroy). Even my Tektronix 2230 from the 1980's had capturing beyond the screen. You can not convince me that oscilloscope manufacters from all over the world have been putting a feature in their equipment for about 40 years which only very few users find important. It just doesn't make sense.

Also keep in mind that EEVblog only has a very small number of EEs taking part. EEVblog is nowhere near representative where it comes to test equipment users or EEs in general. 9 out of 10 EEs I talk to never heard about EEVblog. Then again, Siglent seems to be quite interested in Dave's videos and sends him their equipment to test regulary. So it is reasonable to assume Siglent takes note of Dave's findings. OTOH test equipment manufacturers which have been around for a couple of decades already have a much better view on what is important to their customers and implement (=spend money on) features accordingly. Take a modern DSO from R&S or Tektronix for example, they all allow to set & use the maximum memory length even though these are new designs. That is no accident for sure!

[KungFuJosh]

Take a modern DSO from R&S or Tektronix for example, they all allow to set & use the maximum memory length even though these are new designs. That is no accident for sure!

As was previously said, SOME scope models from SOME brands do that, including Siglent. I don't see the point in this discussion at all.

[awakephd]

This may not be very exciting to those of you who are more experienced ... but just in case anyone is interested, I am happy to report that I have had my first success in using my new SDS804X "for real."

I am sure this was a trivial example for most - a small custom PC board built around an Atmega Tiny 84a, a DVR8837c motor controller, an IR sensor, and a couple of buttons; I designed this to replace a failed controller board. Having 4 channels and the ability to capture relatively slow events (both in rolling mode and in single-shot mode where I could zoom in to check timing between various signals) allowed me to debug a problem with the DVR8837c (turned out that one of the leads didn't get soldered properly) - and it also alerted me to a strange bug that was completely unexpected, a periodic wake-up from sleep mode that shouldn't be happening - I still am working on what that's about. I was also able to test the debounce filter on the switches (using single-shot mode, then zooming in) to discover that I somehow put a 10uF capacitor where I meant to put a .1uF capacitor ... just a wee bit of difference. (In general I really like using 0603 and 0805 SMD components, but I hate that there are no markings on the capacitors - or at least, none that I am aware of - it wasn't until I saw the l - o - n - g slope of the signal that I realized something was wrong.

Again, none of this compares to the kinds of high-speed engineering some of you all do, but I am really thrilled - this is exactly the sort of thing I had not been able to pin down using my old analog scope, and thus the reason I decided to buy a DSO!

[tautech]

This may not be very exciting to those of you who are more experienced ... but just in case anyone is interested, I am happy to report that I have had my first success in using my new SDS804X "for real."

..........to discover that I somehow put a 10uF capacitor where I meant to put a .1uF capacitor ... just a wee bit of difference. (In general I really like using 0603 and 0805 SMD components, but I hate that there are no markings on the capacitors - or at least, none that I am aware of - it wasn't until I saw the l - o - n - g slope of the signal that I realized something was wrong.

Let ST42 be your next investment. 
https://www.eevblog.com/forum/testgear/design-a-new-precision-lcr-tweezers/

[tggzzz]

This may not be very exciting to those of you who are more experienced ... but just in case anyone is interested, I am happy to report that I have had my first success in using my new SDS804X "for real."

I am sure this was a trivial example for most -

Again, none of this compares to the kinds of high-speed engineering some of you all do, but I am really thrilled - this is exactly the sort of thing I had not been able to pin down using my old analog scope, and thus the reason I decided to buy a DSO!

Good stuff!

It is easy to forget that a scope is merely one tool that helps you get a job done. What matters is that you decide what you need to find out, you know what each tool can and cannot do, you choose an appropriate tool, and use it effectively. Often it is easier to answer a question, and more difficult to find the right question to ask.

It is frequently the case that there is overlap in tools capabilities.

[awakephd]

Let ST42 be your next investment. 
https://www.eevblog.com/forum/testgear/design-a-new-precision-lcr-tweezers/

Thanks for the tip! Ordered and on the way ...