What does "Rigol DS1054Z is the best oscilloscope for hobbyists" mean?

[JPortici]

Thanks for the video but I am very sorry that I see a big lag between knob turning and the signal moving vertically accordingly. The lag is similar to the Fluke 123 handheld oscilloscope I used many years ago. As far as I recall, the Yokogawa DL 1540 oscilloscope I used (sweep time: 5nd/div to 50s/div, 56kword (4-channel), 150MHz, display rate of max 60 displays/sec when 4 channels are used)  did not have this problem. From EEVblog #701, the Keysight 3000T also has no lag. What is the reason for the lag? Which specification should I look for if I want a digital scope without the kind of lag shown in this video?

the reason? my bet is "trying to do too many things and excelling at nothing"
in a word, software. scopes with very simillar hardware from other manufacturers (if i remember the teardowns correctly) don't seem to struggle this much in handling the UI

-- the GDS-2000E is a BEAST.
-alot faster
-better FFT
-more memory per channel (yeah if you want to boast 24 Meg on a single channel.. go get a 2 chan siglent. will give you up to 130Meg per channel)
-CAN decoding

it also costs 3 times the 1054Z, all justified. _{(but of course if you PAID the options the price difference would be much smaller for the 100MHz version)}

[JPortici]

In general, how is the GDS-2204E compared with the DS1054Z? Does it have the lag I mentioned in Post 30? From [...] , it looks like there is no lag.

If you look at the screen during the discussion of memory depth in that video (around 2:15), you will see that it was set to 10 kpoints only. I can't read the memory depth in JPortici's video of the DS1054Z lag, but would assume that it was set much higher. I don't have the Rigol scope in front of me, but seem to recall that the UI lag is significantly reduced at lower memory depths. Hence, not an entirely fair comparison.

Nevertheless, the GDS2000E series has the advantage of its Zynq chipset, which allows tight integration of the real-time sample processing in FPGA and the microprocessor-based UI. I have never used one of these scopes, but if GW Instek has made good use of the chipset, would expect them to have an edge over the Rigol in the responsiveness of the UI.

correct! it was at 3 MPts. just changed to 30 kpts and 3 kpts (and to auto, which also reduced the sample rate), no difference.
trust me on this one, i'll upload the video when i'm at work if you're skeptical

[ebastler]

it was at 3 MPts. just changed to 30 kpts and 3 kpts (and to auto, which also reduced the sample rate), no difference.
trust me on this one, i'll upload the video when i'm at work if you're skeptical

No need from my end; I trust your observation, of course!
It must have been in some other context then that the reduced memory depth led to faster operation.

Looking forward to updating the firmware on my own DS1054Z on the weekend and taking it for a spin! ;-)

[StuUK]

Received my 1054Z earlier this week, no regrets at all, so it has a couple of bugs and a bit of lag, so what!!! it's excellent value for the money!!!

[R005T3r]

Not to mention that I want full support from the manufacturers even if the gear costs a little bit more...

But it isn't just a 'little bit more', is it?  Just moving from $400 to $2000 gets nowhere.  Essentially low end scopes of dubious heritage.  By the time you get to NIST traceable, you're up in the multiple 10s of thousands of dollars.

We wouldn't be talking about an entry level 100 MHz scope but a full blown 1GHz+ scope with a datacenter full of sample memory.

Maybe something like this:
http://www.keysight.com/en/pdx-x202179-pn-DSOX3104T/oscilloscope-1-ghz-4-analog-channels?nid=-32541.1150356&cc=US&lc=eng

Oddly enough, this $14,500 scope has less sample memory than the DS1054Z and its still only an 8 bit device with a 3 digit DMM.  My grandmother had a 3 digit DMM!

Even the $17k DSOX4104A has less memory and an 8 bit ADC with a 3 digit DMM
http://www.keysight.com/en/pdx-x201951-pn-DSOX4104A/oscilloscope-1-ghz-4-analog-channels?pm=spc&nid=-32539.1150543&cc=US&lc=eng

And the $200k Infiniium Z series is only 8 bits (12 bits with averaging - this is making some assumptions about the nature of the signal):
http://www.keysight.com/en/pcx-x205212/infiniium-z-series-oscilloscopes?nid=-32529.0&cc=US&lc=eng

This entire scope selection game is pretty much nonsense!  You can spend a metric buttload on high end scopes and all you bought was a bit of bandwidth and a name!

Regarding the scope, to tell the truth, I also found stupid the DMM and waveform generator addons. I like having things separated.
If you don't know what you are doing with high end instruments and you blow them up, well, it's your fault not the producer... The thing is: if there's a bug, and Both the instruments have that bug, how much time will pass in order to fix it? Also, there will be other firmware resales because they have "half solved" the problem? This is another aspect to consider, and frankly I want things like instruments working as expected. Compatibility with accessories and accessories also play an important role: who knows what deserves the future...
I'm not sure that you will get just a bit of bandwidth more and a name.... There must be other things that may justify the price such as they have made extensive work to test the instruments before the release on market, can't just be a name in my opinion, anyway I still have to make research on that, so I could also be wrong.

[ProBang2]

Received my 1054Z earlier this week, no regrets at all, so it has a couple of bugs and a bit of lag, so what!!! it's excellent value for the money!!!

The last software version is implemented?
Should be the version "00.04.04 SP1".

(Press: [Utility] => System => System Info)

If not, then I would  highly recommend to update.
(You get rid of the -IMO- last major bug, the RMS crossover channels error.)

Aside of this:

Congratulations for your new tool. Have a lot fun with it.   
(Perhaps, sometimes you will be even a little impressed what this little device makes possible
- as it happens sometimes to me...)

[Fungus]

Only YOU can determine if the lag would be troublesome to YOU!

No, it's troublesome to everybody.

EVERYBODY would like their DSOs to respond like an old analog 'scope (which is just a potentiometer connected directly to the vertical amplifier - zero lag).

The question is whether or not it's $1000 of annoyance.

I don't know about other people but I really don't think I spend much of my time adjusting the vertical position. Normally it's when I turn channels on/off, which isn't often.

[Fungus]

Which specification should I look for if I want a digital scope without the kind of lag shown in this video?

That won't be written in the specification.

If that's the only thing you're seeing after starting various threads and many many pages of discussion then you're stuck. You literally have no choice. You have to pay $2000+ for a GDS2000 with options (or maybe something even more expensive that has brand-prestige to impress your friends).

You need to admit to yourself that you've already made your choice. This really isn't an engineering decision. It's pointless discussing features, bandwidth, etc. in here.

[JPortici]

it was at 3 MPts. just changed to 30 kpts and 3 kpts (and to auto, which also reduced the sample rate), no difference.
trust me on this one, i'll upload the video when i'm at work if you're skeptical

No need from my end; I trust your observation, of course!
It must have been in some other context then that the reduced memory depth led to faster operation.

Looking forward to updating the firmware on my own DS1054Z on the weekend and taking it for a spin! ;-)

better do it anyway

in this one i found the only combination where it is almost on part with a tek tps2000
as a bonus, the horizontal knob. used with the zoom mode for the most fun of all (accelleration goes from 1 pixel to quarter the trace to full trace in three undefined steps!)

EVERYBODY would like their DSOs to respond like an old analog 'scope (which is just a potentiometer connected directly to the vertical amplifier - zero lag).

The question is whether or not it's $1000 of annoyance.

I don't know about other people but I really don't think I spend much of my time adjusting the vertical position. Normally it's when I turn channels on/off, which isn't often.

100% agree.
i actually move the traces up and down, left and right a bit, when i'm setting up the signal to analyze (and almost invariably i switch off get to the analog scope*)
to me, what i have is too much. moving, wait 1-2 seconds to see if i moved too much and then repeat. you saw it on video, that's my scope.
somebody elses is not like that? is my scope defective? please provide a video proof. i'll be more than glad to have a good reason to return/demand a new one

tonight i might make a third video, moving the traces when the scope is stopped. i actually didn't try that after the update

*hey! what if i connect the vertical output of the analog scope to the rigol? so i could set up things in a lag-free environment and then look at the automated measurements

[CatalinaWOW]

OP.  Regardless of which oscilloscope you buy, regardless of how much money you spend, you will need to understand the impact of your instrument on the research you are doing.  All instruments will have noise, scale errors, offset errors, timing jitter and any number of other imperfections.  You will need to analyze the impact of these errors on your specific conclusions.  The results of that analysis will tell you the minimum instrument capability that you need.  No one here knows what your research will pivot on, only you can do that analysis.

The impact of brand name?  Your analysis should convince any serious researcher of the validity of your results.  But not all grant committees or other funding sources are composed of serious researchers, so there may be some benefit to using a name brand.  This is a double edge sword, because some of those same funding sources will make decisions based on how carefully you are spending their money.

One important consideration that is often overlooked is peer support.  All oscilloscopes are complex instruments and it can be difficult to determine how to do something that the instrument is capable of performing.  What brands/models are used at the facility you will be working at?  Following the local crowd has merit.  Does the brand or model you are considering have good manufacture support, or good support in on line forums?  The Rigol, because of it's low price is widely used and widely supported on-line.  Even the broken record folks who repeat the same complaint over and over provide value.

[Fungus]

wait 1-2 seconds to see if i moved too much and then repeat. you saw it on video, that's my scope. somebody elses is not like that?

I think I figured out why different people are seeing different things. New thread here:

https://www.eevblog.com/forum/testgear/the-ds1054z-vertical-response-discussion-thread/

[eeguy]

OP.  Regardless of which oscilloscope you buy, regardless of how much money you spend, you will need to understand the impact of your instrument on the research you are doing.  All instruments will have noise, scale errors, offset errors, timing jitter and any number of other imperfections.  You will need to analyze the impact of these errors on your specific conclusions.  The results of that analysis will tell you the minimum instrument capability that you need.  No one here knows what your research will pivot on, only you can do that analysis.

The impact of brand name?  Your analysis should convince any serious researcher of the validity of your results.  But not all grant committees or other funding sources are composed of serious researchers, so there may be some benefit to using a name brand.  This is a double edge sword, because some of those same funding sources will make decisions based on how carefully you are spending their money.

One important consideration that is often overlooked is peer support.  All oscilloscopes are complex instruments and it can be difficult to determine how to do something that the instrument is capable of performing.  What brands/models are used at the facility you will be working at?  Following the local crowd has merit.  Does the brand or model you are considering have good manufacture support, or good support in on line forums?  The Rigol, because of it's low price is widely used and widely supported on-line.  Even the broken record folks who repeat the same complaint over and over provide value.

Thanks for the suggestions.

I think most likely I will go for the GDS2204E (if it is better and more recent than the 2204A) or one of the Keysight 2000, 3000T X-Series. Do these two brands have good support both from the manufacturer and from online forums? I asked GW Instek the differences between 2204E and 2204A via their online chat. No reply. Somebody from Keysight called me. He said that he would send me the educational price list two days ago but there has been no pricing information yet.

[nctnico]

You can look at GW Instek's website and see the GDS-2204E has more and better features so yes, it is newer.

[Fungus]

It seems like the vertical response is purely a software issue.

Oh how we'll laugh if the next Rigol update makes it move smoothly. 

BWAHAHAHAHAHAHAHAH!

[StuUK]

Received my 1054Z earlier this week, no regrets at all, so it has a couple of bugs and a bit of lag, so what!!! it's excellent value for the money!!!

The last software version is implemented?
Should be the version "00.04.04 SP1".

(Press: [Utility] => System => System Info)

If not, then I would  highly recommend to update.
(You get rid of the -IMO- last major bug, the RMS crossover channels error.)

Aside of this:

Congratulations for your new tool. Have a lot fun with it.   
(Perhaps, sometimes you will be even a little impressed what this little device makes possible
- as it happens sometimes to me...)

it is the latest and will be doing the upgrade/hack later once I'm happy I'll not cock it up

[Fungus]

it is the latest and will be doing the upgrade/hack later once I'm happy I'll not cock it up

No hurry. All the options are unlocked on a new 'scope for a free trial so you can play with them. The only thing you don't have is 100MHz yet.

Make sure it all works properly before you hack it.

PS: The hack is easy, no need to worry. Don't unlock the 500uV/div option though - it's not supported in hardware and will annoy you if you select it accidentally.

[julian1]

It means it's good bang for the buck and still leaves a budget for the other bits of kit  useful for a generalist/hobbiest. Namely - LCR meter, Signal Generator, bench-meter, dual/tripple power-supply, rework station, iron, components, lighting, books etc.

[StuUK]

it is the latest and will be doing the upgrade/hack later once I'm happy I'll not cock it up

No hurry. All the options are unlocked on a new 'scope for a free trial so you can play with them. The only thing you don't have is 100MHz yet.

Make sure it all works properly before you hack it.

PS: The hack is easy, no need to worry. Don't unlock the 500uV/div option though - it's not supported in hardware and will annoy you if you select it accidentally.

Thanks I saw that in this thread https://www.eevblog.com/forum/beginners/rigol-1054z-73964/msg1023342/#msg1023342
DSER not DSFR

[rstofer]

Only YOU can determine if the lag would be troublesome to YOU!

No, it's troublesome to everybody.

EVERYBODY would like their DSOs to respond like an old analog 'scope (which is just a potentiometer connected directly to the vertical amplifier - zero lag).

The question is whether or not it's $1000 of annoyance.

I don't know about other people but I really don't think I spend much of my time adjusting the vertical position. Normally it's when I turn channels on/off, which isn't often.

This is where I come out.  Yes, there is lag and I never had lag on my Tek 485 but I never had measurements either.  Toss in decoding and it seems like a good tradeoff.  It takes a little longer to get the display the way I want it but I get a whole lot more information.

[David Hess]

If it had that serious a bug it would have been found by now by the community.

The lack of AC precharge because Rigol was too cheap to implement proper ground coupling is a serious bug by design.  Or maybe they did not know any better.

This weeks firmware update has fixed everything that matters. Can we just enjoy our 'scopes now?

Oh, good.  How did they implement AC precharge in firmware?

On the other hand, it is only a $400 DSO and it has 4 channels so if one is blown out, it is not a significant problem.

[Fungus]

On the other hand, it is only a $400 DSO and it has 4 channels so if one is blown out, it is not a significant problem.

The list of forum members with blown channels is a long one, right? There's threads and threads of people complaining. 

[EEVblog]

If it had that serious a bug it would have been found by now by the community.

The lack of AC precharge because Rigol was too cheap to implement proper ground coupling is a serious bug by design.  Or maybe they did not know any better.

First I've heard of it. But I obviously don't read every thread and message on here.

[David Hess]

If it had that serious a bug it would have been found by now by the community.

The lack of AC precharge because Rigol was too cheap to implement proper ground coupling is a serious bug by design.  Or maybe they did not know any better.

First I've heard of it. But I obviously don't read every thread and message on here.

Your own reverse engineering thread revealed no ground input coupling circuit and what Rigol's ground coupling setting actually does has been discussed in a couple of other threads.

Other oscilloscopes allow the use of ground input coupling to precharge the input AC coupling capacitor to prevent damage to the front end circuits; they do *not* just disconnect the input BNC and short the input to the buffer to ground.  This is not only a consideration for the JFET or MOSFET input buffer because a failure of the input protection diodes will almost always be just as bad.  This is not a consideration at low input voltages but it is much easier to exceeded the peak input voltage if the AC coupling capacitor has already been charged to the opposite polarity.