EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: BMOE on May 02, 2016, 12:30:17 pm
-
Hi.
I have got a task from my boss to chose 4 oscilloscopes with a budget of 2500usd for all of them. I know that it's a tough price and my first question was if it was 2500usd each, but it wasn't.
Prefferably two of them can be a bit better and the two others a bit worse :)
Do you have any suggestions of best bang for the buck?
Regards
/Matias
-
"Best bang Per Buck" is easy: The Rigol DS1054Z
But you'd have $900 left over - we can't have that, can we?
If we split the difference two ways you can buy two DS1054Zs and two others for $850 each. There really isn't much in the $850 price range that's a whole lot better than the DS1054Z.
Maybe better to get three DS1054Zs and spend $1300 on something else. With $1300 you can go up to the next level in bandwidth and screen size but you won't get many bells or whistles.
Maybe it would be good to have an idea of what they'll be used for.
-
The $1300 ballpark gets you a 4 channel GW Instek GDS2000E series. Perhaps you could make a package deal with a test equipment dealer. However the risk of buyer 1 good and 3 bad oscilloscopes is that the 3 bad ones will be left in a corner and people fight for the good oscilloscope.
-
The $1300 ballpark gets you a 4 channel GW Instek GDS2000E series. Perhaps you could make a package deal with a test equipment dealer. However the risk of buyer 1 good and 3 bad oscilloscopes is that the 3 bad ones will be left in a corner and people fight for the good oscilloscope.
I wouldn't describe the DS1054Z as "bad", it does a fine job of being a 4-channel general purpose oscilloscope.
But yeah, there's a danger of one person monopolizing the more expensive scope and causing interpersonal tensions in the office.
-
Maybe you can split it between "different" 'scopes, not better/worse.
eg. Two Rigol DS1054Zs and two R&S HMO1002s.
A (hacked) DS1054Z will have more channels and bandwidth than the R&S, the R&S will be smaller and cuter than the Rigol.
Some people will prefer one, some will prefer the other - less overall fighting! :popcorn:
-
I wouldn't describe the DS1054Z as "bad", it does a fine job of being a 4-channel general purpose oscilloscope.
Depending on your purpose it can be terrible or terrific. The devil is in the details ;)
-
Hi.
I have got a task from my boss to chose 4 oscilloscopes with a budget of 2500usd for all of them. I know that it's a tough price and my first question was if it was 2500usd each, but it wasn't.
Prefferably two of them can be a bit better and the two others a bit worse :)
Do you have any suggestions of best bang for the buck?
Regards
/Matias
Welcome to the forum BMOE
For accurate recommendations we need more info to help with your selection.
4 general bench DSO's?
BW?
# channels?
General description of the type of duties they will need to fulfill.
-
We have products which use very high frequency stuff, like USB3, 10Gbe, PCIe gen3 and so on, and of cource we can not use these DSOs for that so yes, it would be general purpose, measuring DC-DC ripple, rise/fall times and so on. Most of the times we only need one channel, but atleast one DSO with 4ch would be nice.
-
it would be general purpose, measuring DC-DC ripple, rise/fall times and so on.
There's two whole cans of worms, right there... :popcorn:
-
We have products which use very high frequency stuff, like USB3, 10Gbe, PCIe gen3 and so on, and of cource we can not use these DSOs for that so yes, it would be general purpose, measuring DC-DC ripple, rise/fall times and so on. Most of the times we only need one channel, but atleast one DSO with 4ch would be nice.
Surely your boss would've given you some guidelines.....or are specs up to you? :-//
Without a better idea of what you'd like I wonder if he's setting you up for a fall. :scared:
IMO for lab/repair shop general purpose use BW should be 100+ MHz.
Would you need things like Decode, Function Gen or MSO ?
If you do and HV work a 400V input rated DSO could also be worth considering.
-
Hum...
1 "refurbished" analog Tektronix
3 Rigol
No?
-
1 "refurbished" analog Tektronix
3 Rigol
Seriously? How about
0 analog boat anchors that have been obsolete for almost two decades now and are pretty much unsupportable, plus lack even the most basic features found in any DSO like storage or measurements?
4 Rigols
Rigol scopes aren't exactly stellar but even the cheap-ass DS1054z is a much more capable scope than any old analog banger.
-
4x 2 channel GDS-2072E GW instek. There is a discount right now on them. http://www.tequipment.net/Instek/GDS-2072E/Digital-Oscilloscopes/?v=0 (http://www.tequipment.net/Instek/GDS-2072E/Digital-Oscilloscopes/?v=0) Various decodiong options already there so no need for hacking (unlikely that company will bother with this crap) or buying options to get a decent functionality. Or add a bit of money and get 4 x 4CH ones. Where are you located BTW, USA?
DS1054z are popular because of hacking bandwidth and decode options but it is not a very good scope IMO, especially for a company.
-
Hum...
1 "refurbished" analog Tektronix
3 Rigol
No?
I'll go with "No".
0 analog boat anchors that have been obsolete for almost two decades now and are pretty much unsupportable, plus lack even the most basic features found in any DSO like storage or measurements?
4 Rigols
Sounds like a better option to me.
Me? I'd get one DS1054Z for evaluation. Check it out, try it for a week on your signals, see if it's up to the job (it most likely is).
Then take it from there. You might end up just buying 4 of them, you might end up with 3 of those and one more expensive one. The point is you haven't committed all your budget to a single decision without trying something first.
DS1054z are popular because of hacking bandwidth and decode options but it is not a very good scope IMO, especially for a company.
In what way are GW instek any better?
-
In what way are GW instek any better?
In almost every way. Faster response. 1mpts FFT (FFT in 1000Z is a joke), more waveforms per second. Better true sensitivity. Decoding already included. No crapxon in the PSU, japanese caps instead.
-
In what way are GW instek any better?
In almost every way.
Every way, except... less bandwidth, less channels, less memory, no trigger out, no graduated display, no fine horizontal/vertical settings ... and all the other ways listed here: https://www.eevblog.com/forum/testgear/gw-instek-gds-2000e-released/ (https://www.eevblog.com/forum/testgear/gw-instek-gds-2000e-released/)
Obviously it's going to be better at some things - it costs more even with a special sale price, but it's not miraculously better. I'm more interested in why you claim the DS1054z is a bad scope "for a company". What does a company have to do with it?
(And why wouldn't the R&S scopes score higher than GW-I in all your criteria for a similar price?)
-
In what way are GW instek any better?
In almost every way. Faster response. 1mpts FFT (FFT in 1000Z is a joke), more waveforms per second. Better true sensitivity. Decoding already included. No crapxon in the PSU, japanese caps instead.
And lets not forget: higher samplerate and no bugs! IMHO you should get a scope which is bug free (at least without the ridiculous bugs the DS1000Z series has) in a professional environment otherwise people will end up chasing ghosts especially if they aren't daily oscilloscope users. Measurement results should be reliable otherwise a scope will cost more than it's own purchase price quickly. For a company a tool must make money!
@Fungus: the GDS2000E series has intensity grading. You can actually adjust that in several ways by setting the intensity and fade-out time!
-
@Fungus: the GDS2000E series has intensity grading. You can actually adjust that in several ways by setting the intensity and fade-out time!
This kind of intensity grading...well well..
Then also old Owon SDS7102 have.
Up to this time I have think variable fade out time persistence and intensity gradient display is bit different things...but always learn new things.
-
@Fungus: the GDS2000E series has intensity grading. You can actually adjust that in several ways by setting the intensity and fade-out time!
This kind of intensity grading...well well..
Then also old Owon SDS7102 have.
Up to this time I have think variable fade out time persistence and intensity gradient display is bit different things
From a functional point of view intensity grading and persistence are very closely related. Both keep multiple past acquisitions on screen for a while.
-
The DS1054Z has intensity grading or am I missing some aspect of it that's being referred to?
Regarding FFT, perhaps I'm in a minority, but practically speaking I find the FFT feature on pretty much any scope I've used is of limited value. Nice to look at, and occasionally useful, but generally I find it remains unused: I think it's mostly a rather overrated feature for real life use.
Happy to be educated on why I should be using the FFT on a scope instead of a spec an.
-
Happy to be educated on why I should be using the FFT on a scope instead of a spec an.
FFT on a scope will work from DC so it is usable for low frequency measurements where a typical spectrum analyser starts from 9kHz or more. I have used this to measure the frequency response of communications related gear. I have also used FFT to look at the spectrum of part of a signal in order to understand which part of a signal causes an EMC problem. In this case time domain and frequency domain where very much intertwined.
-
The DS1054Z has intensity grading or am I missing some aspect of it that's being referred to?
It does, although a bit limited. Especially the infinite persistence mode can be useful to spot glitches and judge the presence of jitter and its maximum magnitude, although the limited grading will limit its value to assess which are the most common occuring values or shapes, which can be much easier in scopes with 256 levels, and, especially, in those which use different colors to create a sort of "heat map".
Regarding FFT, perhaps I'm in a minority, but practically speaking I find the FFT feature on pretty much any scope I've used is of limited value. Nice to look at, and occasionally useful, but generally I find it remains unused: I think it's mostly a rather overrated feature for real life use.
The FFT in these Chinese scopes is mostly a "check-box" thing. But look at the FFT available in higher end scopes and it can indeed be useful. My other DSO is a venerable LeCroy 9400 which I must recap, and it has an awesome FFT. True, calculating it can take even 30 seconds (yikes!) but you can get really detail information from it. It shouldn't be difficult for Rigol to replicate that, given the speed offered by the processors in these modern DSOs (the 9400 has a Motorola 68000) but, well, it seems they won't bother.
The 9400 lets you choose the number of points, wether you want the modulus, real or imaginary part, etc. Yep, the Physics background it quite apparent ;)
Although an 8 bit converter has a very limited resolution, a good FFT can help you spot distortion problems, for instance.
-
Happy to be educated on why I should be using the FFT on a scope instead of a spec an.
FFT on a scope will work from DC so it is usable for low frequency measurements where a typical spectrum analyser starts from 9kHz or more. I have used this to measure the frequency response of communications related gear. I have also used FFT to look at the spectrum of part of a signal in order to understand which part of a signal causes an EMC problem. In this case time domain and frequency domain where very much intertwined.
I guess I don't do much down at audio to need to worry about that ;-) but equally the first thing that comes to mind is to use a sound card solution which would have far greater resolution and dynamic range although granted not as convenient. I guess that's my point really: the FFT offerings on scopes are typically low in terms of resolution and dynamic range, and a scope's noise floor is very high. For EMC, I can't imagine ever using a scope instead of an SA myself, for the reasons I just mentioned. The only time I've used the FFT function on a scope recently for real work is on the NCO peripheral in a PIC, but even then I ended up using an SA.
-
Regarding FFT, perhaps I'm in a minority, but practically speaking I find the FFT feature on pretty much any scope I've used is of limited value. Nice to look at, and occasionally useful, but generally I find it remains unused: I think it's mostly a rather overrated feature for real life use.
I guess this is because you've never used a scope with a proper FFT implementation, and it's true that especially on many entry-level scopes FFT pretty much sucks.
Happy to be educated on why I should be using the FFT on a scope instead of a spec an.
Aside from getting a spectrum of very low frequency signals (most SA's start at 9kHz or even in the MHz range), FFT can be useful for analyzing pulsed and or frequency hopping signals which a conventional swept SA often has problem capturing because of the band sweep and the low dwell time (which a modern SA gets around by using - FFT ;) ).
A good(!) scope is also more flexible in terms of frequency resolutions, which on most SAs are usually limited to fixed settings at between 10Hz and 5MHz while FFT on the scope can go below 1Hz or way beyond 5MHz (which helps analyzing wide bandwidth signals).
The main reason however is that FFT is part of the scope, i.e. you don't need another device to look at the frequency domain. And not everyone has a spectrum analyzer.
-
Every way, except... less bandwidth
You need to hack 1054Z first (usually no go for a company), otherwise instek has more bandwidth.
less memory
Well, unless you hack 1054Z, they are on par with single channel used. If using multiple, GDS-2000E beats it as it has 10Mpts per channel. Also that memory in Rigol is not so usefull at all. Sample rate in GW instek also does not drop when multiple channels used.
I'm more interested in why you claim the DS1054z is a bad scope "for a company". What does a company have to do with it?
Unless you hack it, there are no goodies because of which people buy it. If compared with not hacked DS-1054Z, you can buy instek GDS-1054B scope as well. It will be certainly better than not hacked DS-1054Z with the same price.
-
The FFT in these Chinese scopes is mostly a "check-box" thing. But look at the FFT available in higher end scopes and it can indeed be useful. My other DSO is a venerable LeCroy 9400 which I must recap, and it has an awesome FFT. True, calculating it can take even 30 seconds (yikes!) but you can get really detail information from it. It shouldn't be difficult for Rigol to replicate that, given the speed offered by the processors in these modern DSOs (the 9400 has a Motorola 68000) but, well, it seems they won't bother.
Rigol has already made a massive improvement in the DS1000Z FFT via firmware update since the 'scope was first released:
https://www.youtube.com/watch?v=dD-3ni7tYM4 (https://www.youtube.com/watch?v=dD-3ni7tYM4)
I'm not saying it's all perfect now but Rigol obviously does care.
-
Rigol has already made a massive improvement in the DS1000Z FFT via firmware update since the 'scope was first released:
Yep, they upgraded it from a completely shameful joke to a bit less of a joke but still a joke.
-
@Fungus: the GDS2000E series has intensity grading. You can actually adjust that in several ways by setting the intensity and fade-out time!
This kind of intensity grading...well well..
Then also old Owon SDS7102 have.
Up to this time I have think variable fade out time persistence and intensity gradient display is bit different things
From a functional point of view intensity grading and persistence are very closely related. Both keep multiple past acquisitions on screen for a while.
It could be true for slow waveform/s, but when several traces are captured in a display refresh time, then you will see an even block of waveforms, regardless of the persistence settings. Which, essentially means that most of the time you won't have any differences in shadings on an Owon XDS.
I am looking for a second scope, and I am thinking between the XDS 3102A, Siglent SDS 2104X and the GW Instek GDS-2104E. In this regard, the Owon loses points.
-
@Fungus: the GDS2000E series has intensity grading. You can actually adjust that in several ways by setting the intensity and fade-out time!
This kind of intensity grading...well well..
Then also old Owon SDS7102 have.
GDS-2000E has 256 color gradients. It is intensity grading, different kind of it but you can tweak it more than the flavor of intensity grading in Rigol scopes.
here is an old comment from a real owner:
Well, my brand new GW-Instek GDS2202E scope arrived yesterday and it is having a lasting impression on me. I'm very glad I bought it.My Rigol DS1104Z has been moved to the back burner.
I just want to clarify something here regarding Rigols Ultra-Vision whose concept it is to make their digital scopes screens look and feel and behave like the old crt scope screens, that is increased intensity on modulated waveforms where the 'knots' appear. I must say that Rigol has done a very good job here, but wait...why make a digi scope screen behavior the same as a CRT screen?Whats so fascinating about CRT screen behavior? CRT scopes are gone and they are in the past now.
Cheers
commie
-
Rigol has already made a massive improvement in the DS1000Z FFT via firmware update since the 'scope was first released:
I'm not saying it's all perfect now but Rigol obviously does care.
Don't get me wrong, I'm not a practitioner of that Rigol bashing thing. I've purchased a DS1074Z and, despite its many limitations I am very happy with it. I have even recommended it to some friends.
Now. There is a problem with Chinese product development (or I have had really bad luck with the Chinese products I've tried).
I haven't seen a single product with that "spark" you perceive in that old Lecroy. It's quite clear that the development team for the 9400 did really strive to create something unprecedented. it's crazy to achieve that level of performance with that processor. I am sure the development process was amazing. You can see similar hints in many American, European or Japanese electronic products.
The Chinese products I've tried so far look more like a routine thing. Someone there says: let's build an oscilloscope. So they look at what a customer is supposed to expect, and they make it. They make it very cheap, and that's all. I acknowledge that cutting the right corners in order to lower costs is a real art. But that's not what I see in these products.
The other day was speaking with a colleague who was evaluating some Chinese made network infrastructure equipment. The documentation was a real joke, and it's supposed to be the top tier of Chinese carrier class equipment. Just a list of fields in user interface windows with explanations such as: "RSSI: this field shows the RSSI".
I don't want to be derogatory, but I think that China has a long way to go before they will build products perceived as great. Maybe one day they will acquire that "mojo" and we will wear Rigol t-shirts with pride. But it will take time.
Maybe they have a serious cultural problem. After all their regime doesn't exactly encourage free and independent thinking, and that has a lot to do with creating great products.
I think they are making progress, though. It seems that representatives from Siglent, Rigol and Owon at least are reading this forum. And, as you say, they are beginning to pay attention to the feedback of their customers, which is a huge step.
I hope they won't take my comments as offensive. I try to be constructive pointing out what they lack now. And, who knows, maybe in some years I end up purchasing a new Chinese oscilloscope and wearing a t-shirt with their logo! :)
-
The Chinese products I've tried so far look more like a routine thing. Someone there says: let's build an oscilloscope. So they look at what a customer is supposed to expect, and they make it. They make it very cheap, and that's all.
It is an expected philosophy and working method there. I work with chinese software developers and they are very much result oriented. Just work and make the job done! In fact, they are fast. Insanely. However, of course, because of that they make huge amount of bugs and totally ridiculous code. So, at the end, you spend much more time to fix their code and bugs, than originally needed to implement that feature. So, all in all it is a nightmare.
However, because of that, you will have feature filled and cheap oscilloscopes. As long as they update the firmware frequently, it could be an acceptable trade off. You just need to be aware of that.
So it's not that these engineers are stupid or under educated. They are pretty smart, indeed. They just give a shit on your west working culture, like there is a people with guns behind every developer and kill of them who works precisely and who takes care about optimization.
-
The Chinese products I've tried so far look more like a routine thing. Someone there says: let's build an oscilloscope. So they look at what a customer is supposed to expect, and they make it. They make it very cheap, and that's all. I acknowledge that cutting the right corners in order to lower costs is a real art. But that's not what I see in these products.
Maybe they have a serious cultural problem. After all their regime doesn't exactly encourage free and independent thinking, and that has a lot to do with creating great products.
That's a good reason why I left China. The lowest level of R&D is clone, then simply implement, then refine, finally develop concept. Most Chinese companies are still in the clone or implement phase, while some advanced ones like Rigol are in the refiners' league. There is only a few Chinese big companies can do conceptual or leading edge fundamental science development, while they are usually subsidiaries of western companies.
One of the reasons is, the majority of consumers in China don't give a shit to originality or somewhat similar things. Most consumers just want the cheapest, and give absolute no respect to intellectual property. Therefore, small but innovative companies are guaranteed to be cloned by big companies and can not compete in the market. That drives companies less innovative. Chinese government tried to enforce copyright recently more rigorously by shutting down some pirate video game websites and online music websites, and got bitched by the people on various Chinese BBS and forums a lot.
It's getting slightly better, especially from the root of the problem -- respecting to intellectual property from the common folks. However, there is still a long way to go. Even now, some of my friends call me traitor to the western world because I pay money to buy legit copies of western software and hardware. Many people, even the young ones, still have the red China thinking pattern that paying money to the western is betraying the country.
I can certainly relate to most of your comments. ^^
I've identified a gap in the market for a mildly specialised DSO but trying to get acceptance that there's market demand for it let alone dedication of resources for its development has been very hard work.
Maybe I'm just not good at selling an idea to a manufacturer. |O
But we'll keep bashing away at it and hopefully we'll have something to share on this later. ;) :popcorn:
-
Hi.
I have got a task from my boss to chose 4 oscilloscopes with a budget of 2500usd for all of them. I know that it's a tough price and my first question was if it was 2500usd each, but it wasn't.
Prefferably two of them can be a bit better and the two others a bit worse :)
Do you have any suggestions of best bang for the buck?
Regards
/Matias
How about tell us where you are? I think if you are in the US you will have different choices than if you are in the Congo jungle.
If in the US I would contact TE Equipment and ask them. Like someone said, you may be getting set up. Let someone else make the decision for you.
-
@Fungus: the GDS2000E series has intensity grading. You can actually adjust that in several ways by setting the intensity and fade-out time!
This kind of intensity grading...well well..
Then also old Owon SDS7102 have.
Up to this time I have think variable fade out time persistence and intensity gradient display is bit different things
From a functional point of view intensity grading and persistence are very closely related. Both keep multiple past acquisitions on screen for a while.
In main principle intensity grading works so that there is more intensity when more past acquisitions are overlayed and collide.
Persistence is just counting time how old every one acquisition is. But I have not seen never this kind of persistence what count fade off time for every single acquisitions. .
More I have seen persistence what only count how old one display frame is where may be one or thousends of acquisitions at once. How you get statistical density information to eye with this. Of course persistence function is important and more nice if there can use also fast fade off times and adjustment steps are small.
If there is advanced persistence what count age of every single acquisition it may give more advanced result, but still collision based intensity gradation is very different least in theory.
Both keep multiple past acquisitions on screen for a while.
Intensity gradation display alone keep only one display refresh cycle. Intensity come from how much collisions inside in this one frame and just this "collision intensity gradation" (or how to name it) is one important information for user. Usually, afaik, this is what we talk when we talk about DPO intensity gradient display.
If then add persistence, it keep "old screen(S)" as long as user have selected. And in this fade off time there may also be intensity adjusted fade off with less or more intensity steps. Also persistence (with or without _persistence intensity gradation_) is important specially for fast single rare events.
They are not alternatives to each other. They both are needed but persistence fade off intensity gradation is nearly only for nice image and for fooling peoples to think there is real "collision based" intensity gradation.
-
@Fungus: the GDS2000E series has intensity grading. You can actually adjust that in several ways by setting the intensity and fade-out time!
This kind of intensity grading...well well..
Then also old Owon SDS7102 have.
Up to this time I have think variable fade out time persistence and intensity gradient display is bit different things
From a functional point of view intensity grading and persistence are very closely related. Both keep multiple past acquisitions on screen for a while.
In main principle intensity grading works so that there is more intensity when more past acquisitions are overlayed and collide.
Persistence is just counting time how old every one acquisition is. But I have not seen never this kind of persistence what count fade off time for every single acquisitions. .
Tektronix (at least the TDS500/600/700 series), Lecroy Wavesurfer 3000 and GWInstek work this way. It makes sense too. Unless you want infinite persistence older traces are just cluttering the screen so fading old acquisitions away at a selectable rate keeps 'rare' events long enough on the screen to see but removes them before they get in the way.
-
@Fungus: the GDS2000E series has intensity grading. You can actually adjust that in several ways by setting the intensity and fade-out time!
This kind of intensity grading...well well..
Then also old Owon SDS7102 have.
Up to this time I have think variable fade out time persistence and intensity gradient display is bit different things
From a functional point of view intensity grading and persistence are very closely related. Both keep multiple past acquisitions on screen for a while.
In main principle intensity grading works so that there is more intensity when more past acquisitions are overlayed and collide.
Persistence is just counting time how old every one acquisition is. But I have not seen never this kind of persistence what count fade off time for every single acquisitions. .
Tektronix (at least the TDS500/600/700 series), Lecroy Wavesurfer 3000 and GWInstek work this way. It makes sense too. Unless you want infinite persistence older traces are just cluttering the screen so fading old acquisitions away at a selectable rate keeps 'rare' events long enough on the screen to see but removes them before they get in the way.
Yep, R&S does the same. The old traces fading out gradually - I thought that every brands do the same. :wtf:
But, all in all, is that true, that the GW does not have intensity grading which summarized and calculated all the data in one screen refresh period? So in GW intensity grading === persistence?
-
Tektronix (at least the TDS500/600/700 series), Lecroy Wavesurfer 3000 and GWInstek work this way. It makes sense too. Unless you want infinite persistence older traces are just cluttering the screen so fading old acquisitions away at a selectable rate keeps 'rare' events long enough on the screen to see but removes them before they get in the way.
Yep, R&S does the same. The old traces fading out gradually - I thought that every brands do the same. :wtf:
But, all in all, is that true, that the GW does not have intensity grading which summarized and calculated all the data in one screen refresh period? So in GW intensity grading === persistence?
Intensity grading is always a form of persistence. That is how the old analog CRT tubes work as well. The fosfor is hit by the electron beam and gradually fades back to black. You can't summarise all data in one screen refresh period because the data written close to the end of the screen refresh cycle will be on screen shorter than data written at the beginning of the screen refresh cycle. Also a screen refresh cycle is way to short for people to notice something on the screen so you'll need to display the data for a much longer time. All in all you'll need a form of gradually fading persistence mode to do intensity grading.
-
why make a digi scope screen behavior the same as a CRT screen?Whats so fascinating about CRT screen behavior? CRT scopes are gone and they are in the past now.
It is not. No one want full emulation of old analog phosphor. (of course it is quite easy to add also this to intensity gradation DPO but I do not know anybody who want this "fade in" - "fade out". This "fade in" is most weak point in analog scope. Even with most fast Tektronix micro channel MCP CRT)
What we want is analog tube good thing, intensity gradation (because it give more information from signal) BUT without analog tube phosphor intensity "wakeup" delay.
We get best things from analog scope and best things from digital scope together. It is more than old Digital DSO and much more than old analog.
Best things we get when we have two things.
1. Intensity gradation from waveform datapoints image map overlay collisions.
2. Variable user settable persistence time. If it is with adjustable fade-out intensity gradation speed it is more nice. Best if there is adjustable "keep" time and adjustable "fade out" time. (Or something like in audio. Attack/Sustain-Deacay/release(fade out --- but in oscilloscope we want this attack slope is as fast as ever possible - but in analog scope it is slow)
(If improve more this persistence then it include also age counting for every single acquisition in one TFT image of course it is super. But, this is bit hard and do not give so much advantage for user exept perhapsonly bit more nice image - nearly cosmetics)
-
Tektronix (at least the TDS500/600/700 series), Lecroy Wavesurfer 3000 and GWInstek work this way. It makes sense too. Unless you want infinite persistence older traces are just cluttering the screen so fading old acquisitions away at a selectable rate keeps 'rare' events long enough on the screen to see but removes them before they get in the way.
Yep, R&S does the same. The old traces fading out gradually - I thought that every brands do the same. :wtf:
But, all in all, is that true, that the GW does not have intensity grading which summarized and calculated all the data in one screen refresh period? So in GW intensity grading === persistence?
Intensity grading is always a form of persistence. That is how the old analog CRT tubes work as well. The fosfor is hit by the electron beam and gradually fades back to black. You can't summarise all data in one screen refresh period because the data written close to the end of the screen refresh cycle will be on screen shorter than data written at the beginning of the screen refresh cycle. Also a screen refresh cycle is way to short for people to notice something on the screen so you'll need to display the data for a much longer time. All in all you'll need a form of gradually fading persistence mode to do intensity grading.
Here you are some examples just to clarify what I mean. In the below examples persistence was completely off.
In this case you can see 600 pts data, but accumulated in time, captured in stop mode:
(https://www.eevblog.com/forum/testgear/help-to-chose-4-oscilloscopes-for-2500usd-togeather/?action=dlattach;attach=222185;image)
In this case you can see 500 kpts data, accumulated vertically, this is a single acquisition:
(https://www.eevblog.com/forum/testgear/help-to-chose-4-oscilloscopes-for-2500usd-togeather/?action=dlattach;attach=222187;image)
This is very useful not only in the special cases above, but every time. This is a single acquisition as well, hence it totally independent of waveform update rate. Just one acquisition, data captured, crushed vertically and shown:
(https://www.eevblog.com/forum/testgear/help-to-chose-4-oscilloscopes-for-2500usd-togeather/?action=dlattach;attach=222189;image)
This is an opposite, only 600 pts of data, but accumulated in time. In stop mode.
(https://www.eevblog.com/forum/testgear/help-to-chose-4-oscilloscopes-for-2500usd-togeather/?action=dlattach;attach=222191;image)
Neither of the above samples are monochromatic and show useful information.
This is 600 pts of data, in single acquisition. The best thing, this is also not monochromatic :P
(https://www.eevblog.com/forum/testgear/help-to-chose-4-oscilloscopes-for-2500usd-togeather/?action=dlattach;attach=222193;image)
I do think, that GW does the above in a way, without any persistence settings. (Or maybe they just renamed the intensity grading to persistence?)
-
Here you are some examples just to clarify what I mean. In the below examples persistence was completely off.
In this case you can see 600 pts data, but accumulated in time, captured in stop mode:
(https://www.eevblog.com/forum/testgear/help-to-chose-4-oscilloscopes-for-2500usd-togeather/?action=dlattach;attach=222185;image)
Looks nice but how can we know how/why/when the image was captured....that information is not in the screenshot. :--
-
Looks nice but how can we know how/why/when the image was captured....that information is not in the screenshot. :--
Well, you can almost read it from the traces ;) It is 5 Vpp, 2 Mhz, FM, freq dev: 400 kHz, FM freq: 100 Hz (but that does not matter too much). As it based on the waveform update rates mainly, you should get it much nicer on any newest Siglent or Rigol. This is about 7300 waveforms/s, clearly not the highest value available nowadays, yet pretty good.
-
Looks nice but how can we know how/why/when the image was captured....that information is not in the screenshot. :--
Well, you can almost read it from the traces ;) It is 5 Vpp, 2 Mhz, FM, freq dev: 400 kHz, FM freq: 100 Hz (but that does not matter too much). As it based on the waveform update rates mainly, you should get it much nicer on any newest Siglent or Rigol. This is about 7300 waveforms/s, clearly not the highest value available nowadays, yet pretty good.
Here you are some examples just to clarify what I mean. In the below examples persistence was completely off.
In this case you can see 600 pts data, but accumulated in time, captured in stop mode:
I meant Stop, not measurements, sure there's much info that can be read from a waveform but neither Stop (shows Auto trig and Run) or Memory depth are shown.
When building documentation of a project, screenshot information is a valuable record and needs be complete as possible.
-
Here you are some examples just to clarify what I mean. In the below examples persistence was completely off.
In this case you can see 600 pts data, but accumulated in time, captured in stop mode:
I meant Stop, not measurements, sure there's much info that can be read from a waveform but neither Stop (shows Auto trig and Run) or Memory depth are shown.
Pardon? It is written "Stop" mode. Which essentially means: I pushed the Stop button and took a screenshot. So it is exactly which you can see in Run, it just freezes the screen (but it does not matter at all, I could take a screenshot in Run mode as well, the result is all the same. What is important here: it is not a single acquisition). The memory depth is 600 pts, as noted, but you can also calculate that from the sample rate 500 MS/s and from the Time base.
Are you just nitpicking here, or what is your point? (I am sure you just don't want to abuse my tech writing skills :box:)
-
Here you are some examples just to clarify what I mean. In the below examples persistence was completely off.
In this case you can see 600 pts data, but accumulated in time, captured in stop mode:
I meant Stop, not measurements, sure there's much info that can be read from a waveform but neither Stop (shows Auto trig and Run) or Memory depth are shown.
Pardon? It is written "Stop" mode. Which essentially means: I pushed the Stop button and took a screenshot. So it is exactly which you can see in Run, it just freezes the screen (but it does not matter at all, I could take a screenshot in Run mode as well, the result is all the same. What is important here: it is not a single acquisition). The memory depth is 600 pts, as noted, but you can also calculate that from the sample rate 500 MS/s and from the Time base.
Are you just nitpicking here, or what is your point? (I am sure you just don't want to abuse my tech writing skills :box:)
Nitpicking ...maybe, having a go at you...NO, looking hard at the R&S UI....yes.
My point is there is very limited info of scope parameters displayed on the screenshot when captured.
I'd expect more. Period.
-
I think you guys have chased away the OP :-//
-
Yep, I have a feeling that choosing any scope below 3000 Euros is like choosing a car without doors, or without lamps, or without seats. You could have everything essentially in different scopes, but you could not have all in one device.
Is that an impossible mission to implement a proper intensity gradation in new scopes nowadays? Nope, Siglent and Rigol have succeeded. But they fail in other important aspects. Building temperature controlled, high quality fans with rubber mounts and put heatsinks raises the prices, okay, but how much? 20 euros? Is that worth to skip that? Why skip basic functionality, like measurement statistics. It almost costs nothing. Too small numbers of FFT points, ohhh, c'mon.
(https://i.imgflip.com/13mg9i.jpg)
-
Yep, I have a feeling that choosing any scope below 3000 Euros is like choosing a car without doors, or without lamps, or without seats. You could have everything essentially in different scopes, but you could not have all in one device.
It's not quite that bad, but yeah; anybody who ever bought a new car can tell you it wasn't perfect and when they went back to the dealer to complain the dealer just shrugged and said "they're all like that".
PS: You think 3000 Euro scopes or Mercedes Benz are perfect? They might make more effort, but...
-
For those who say there in no real intensity grading in GDS-2000E
(https://www.eevblog.com/forum/testgear/help-to-chose-4-oscilloscopes-for-2500usd-togeather/?action=dlattach;attach=222614;image)
-
For those who say there in no real intensity grading in GDS-2000E
Nice! Is that with persistence off?
-
For those who say there in no real intensity grading in GDS-2000E
Nice! Is that with persistence off?
You can see on the picture that it is 16ms. But this setting does not matter, intensity grading works regardless unless you switch persistence off completely. In my Rigol DS-2072, there is no persistence off setting at all, only MIN. And when you set anything higher than MIN, intensity grading stops working. As of the visual appearance, it looks nicer for the eye on Rigol but it is because there is some overall display smoothing going on which is not present in GW instek. As of the feel of scope overal, Rigol DS-2000 feels almost like a piece of crap compared to the GW instek standing nearby. Buttons so much nicer, menu so much easier to use and instant fast, controls not cramped together.
-
PS: You think 3000 Euro scopes or Mercedes Benz are perfect? They might make more effort, but...
Not commenting about MB but for scopes 3k€ is pretty much roughly where it changes from 'mostly a toy' to 'a reliable, dependable and well supported professional tool'.
-
For those who say there in no real intensity grading in GDS-2000E
(https://www.eevblog.com/forum/testgear/help-to-chose-4-oscilloscopes-for-2500usd-togeather/?action=dlattach;attach=222614;image)
This is an amplitude-modulated sine wave, right? So why is the curve with the lowest amplitude so much brighter than the others, and specifically brighter than the one with the highest amplitude?
It seems that, on the GDS-2000E, the brightness of a dot is a measure of how long ago the dot was written to the display (consistent with the term "visual persistence oscilloscope" which GW Instek chose). In contrast, what I look for in an intensity graded display is that the brightness indicates how often a dot is re-written. In the latter case, the minimum and maximum modulation curve should be of equal brightness in your example, I think.
I believe this difference is what others have also pointed out earlier in this thread, when they questioned whether GW-Instek has "real" intensity grading. I would also argue that GW-Instek simulates persistence, but not intensity grading (which enables discrimination of rare events from regular signals).
-
This is an amplitude-modulated sine wave, right? So why is the curve with the lowest amplitude so much brighter than the others, and specifically brighter than the one with the highest amplitude?
It seems that, on the GDS-2000E, the brightness of a dot is a measure of how long ago the dot was written to the display (consistent with the term "visual persistence oscilloscope" which GW Instek chose). In contrast, what I look for in an intensity graded display is that the brightness indicates how often a dot is re-written. In the latter case, the minimum and maximum modulation curve should be of equal brightness in your example, I think.
I believe this difference is what others have also pointed out earlier in this thread, when they questioned whether GW-Instek has "real" intensity grading. I would also argue that GW-Instek simulates persistence, but not intensity grading (which enables discrimination of rare events from regular signals).
That setting is 16ms persistance, read during 1 frame at 60 fps. As how waveform looks like, it is about the same on Rigol DS-2000. What people don't get is that there is no intensity grading without persistence on Rigol too :palm:, actually you cannot turn off persistence on Rigol.
how long ago the dot was written to the display
In that case this waveform would be in one monotone color with no color grading at all.
In the latter case, the minimum and maximum modulation curve should be of equal brightness in your example, I think.
No, because during the same time period minimum modulation trace takes less screen area than max modulation trace = max modulation trace hits the same dot on the screen for less time that min modulation trace.
-
:palm:
(https://www.eevblog.com/forum/testgear/help-to-chose-4-oscilloscopes-for-2500usd-togeather/?action=dlattach;attach=222837;image)
-
I think with so much passion on persistance, someone should start a new topic on it. Really no answer to OP who has been scared away.
-
Seen side by side, I don't think there can be much argument over who's doing it properly:
(https://www.eevblog.com/forum/testgear/help-to-chose-4-oscilloscopes-for-2500usd-togeather/?action=dlattach;attach=222847;image)
Can anybody do the DS1054Z? I believe there's not much difference between that and the DS2000 but it would be nice to see a pic.
-
In my place people always choose the 200MHz crappy Tek DPO2k and 350MHz Tek MDO3k, while there is a 2GHz Tek DPO5k, and a 1GHz Keysight MSOX3k, and nobody but me uses them.
I would not use the Tektronix DPO5k either unless I had to. When I evaluated one a couple years ago, the human factors engineering was poor, the graphical interface had lots of bugs, and it crashed a lot.
-
1 "refurbished" analog Tektronix
3 Rigol
Seriously? How about
0 analog boat anchors that have been obsolete for almost two decades now and are pretty much unsupportable, plus lack even the most basic features found in any DSO like storage or measurements?
4 Rigols
Rigol scopes aren't exactly stellar but even the cheap-ass DS1054z is a much more capable scope than any old analog banger.
Are there any features in even hacked DS1054Zs which can make up for low bandwidth? For some measurements, that is all that counts. I could understand buying a pair of DS1054Zs along with a couple of used high bandwidth analog oscilloscopes.