Upgrade my DSO... or not?

[IDEngineer]

It's been feeling like time to upgrade my DSO. I have a Rigol DS4024 with all options including 500MHz bandwidth. It's in perfect shape and I figure I can probably sell it to recover some of its value to apply toward something new.

I've mentally allowed a budget of up to ~$4K for a new DSO. I've been reading all the threads, watching the teardown videos, comparing spec sheets, etc. And after all of that... I'm questioning whether it's worth changing.

Objectively, I'm sitting here with a 500MHz four channel DSO with 4GSa/sec sample rate, an 8 bit front end, deep memory, reasonably sized screen, and so forth. It's not cheap to duplicate that setup. Yes, it's a Rigol and not one of the "bigger brands", but frankly it works remarkably well for my applications. The 4024's biggest negative is the basically useless FFT but a standalone spectrum analyzer would address that and I already have an HP 8921A that goes to 1GHz until I get a formal standalone SA. Other than the FFT, the 4024 works pretty well.

On the positive side, the 4000 series is getting a bit older and so it might be nice to freshen my single most important piece of test equipment. I'll openly admit there's an emotional component here, once in a while it's just "nice" to have a new primary scope on the bench.

On the negative side: What problem am I really solving? I don't beat the table every day cursing the 4024 because of some fatal flaw. A 10 bit front end might be nice, but I probably don't NEED it. An even larger screen might be useful, but I'm not squinting today. And is it really worth dropping back to 200-350MHz from 500MHz just to get 10 bits or later firmware?

Finally... I'm just not sure I can duplicate this performance within a ~$4K budget. Four channels of 500MHz at 4GSa/sec starts pushing the price tag pretty hard. Maybe I should just buy a can of "New Scope Smell" and be happy with that.

I know the folks here can relate to this sort of quandry, and I'm wondering what you'd think if you were considering such a decision.

[maginnovision]

For $4000 I'd probably keep the Rigol and save it for a newer scope later down the road. Might even catch a good deal on something. Unless you would consider buying used then there are a lot of options...

[tautech]

Yeah well the new 4ch 500 MHz Siglent is $5,689 however its baby 350 MHz brother is a more digestible $3,569 and there's a BW upgrade option for it........no surprises for guessing what I'm thinking.   

Just unpacked one before it goes to a customer and first impressions are really good.
Really really quiet with a smart fan.
Still getting to grips with it...................

[DaJMasta]

Got the same scope (sans bandwidth upgrade) and a lesser budget, but I don't think there's any upgrade in my future for a couple of years yet.  I think the killer feature I'd want above others is 10 or 12 bit vertical resolution, and I'm not prepared to dish out for a nice current gen scope, so I don't think it's realistic for me until there are some used ones on the market or some are showing up as clearance deals, i.e. a few years down the line.

I think there are definitely better scopes than a Rigol DS4000 series if buying new, but right now the price category is sort of a sidegrade rather than an upgrade for responsiveness, analysis features, and maybe a lower noise floor... and I doubt the resell rate for the DS4000 will cover more than half the new scope's cost... so at least for me, not yet worth it.

Agreed that the FFT implementation is sort of obtuse and not that useful, but this was sort of before manufacturers really started focusing on it, so any of the newer generation scopes generally treat it better.  FWIW, if I were to upgrade now, I'd probably go R&S, the problem is that as nice as their scopes have been since they introduced, the better vertical resolution has only been in the last generation and a half or so.  Keysight's got some good options, but they usually command even more of a price premium, Tek's got some but there seem to be complaints with the UI (from some people and not others?) and the price is still up there.... LeCroy has some decent options, but like the other major manufacturers, the pricetag means the value proposition of a Rigol/Siglent/Used scope just doesn't seem the same to me for home use.

[IDEngineer]

Really really quiet with a smart fan.

OK, forgot to mention silence. The Rigol is a bit louder than I'd like though I haven't been tempted to swap for a quieter fan (if that's possible while keeping the CFM).

That Siglent sounds tempting.

[IDEngineer]

I think the killer feature I'd want above others is 10 or 12 bit vertical resolution....

Agreed, that's why I mentioned 10 bits specifically.

BTW, something I read somewhere (that's real definite, eh?  ) suggested to me that the DS4000 series might yield >8 bit vertical resolution under certain conditions. Know anything about that? I didn't dig deeper at the time, I'd have to go back and find whatever reference that was. Maybe we already have >8 bit scopes under the correct circumstances.

Agreed that the FFT implementation is sort of obtuse and not that useful, but this was sort of before manufacturers really started focusing on it, so any of the newer generation scopes generally treat it better./

Agree again. To me a scope is a scope. FFT is a nice bonus but not really a decision maker. If you need a spectrum analyzer, get a spectrum analyzer. I plan to do so at some point, until then the HP 8921A covers me to 1GHz (though it's an older, analog, monochrome device with none of the fancy computational features we expect these days). At least it has an integrated tracking generator, which I used to wind some ham radio antenna traps for an inverted-V a while back.

FWIW, if I were to upgrade now, I'd probably go R&S... Keysight's got some good options... Tek's got some but there seem to be complaints with the UI... LeCroy has some decent options....

Exactly. There's no clear winner that screams "BUY ME NOW!" and justifies the decision. It's almost like you have to rank the kind of work you plan to do, then favor that aspect of the spec sheets and settle for second-rate in the other departments. There's a thread going right now about serial decoding that makes some brands stand above the others, but the whole time I'm thinking "I can use a logic analyzer for that, why compromise my choice of scope for a digital serial data stream?"

Maybe I'm complaining about an embarrassment of riches. We have a LOT of good choices in scopes these days. It's not like back in the days of analog scopes where they all did pretty much the same thing. These days the feature sets themselves differ quite a bit between brands. In many ways that's a GOOD thing, it's nice to have options.

Have to admit, though, it would be nice to be driving a solid R&S or LeCroy. I've never had a scope of that caliber on my bench. Probably never be able to afford one either!

[IDEngineer]

BTW, something I read somewhere (that's real definite, eh?  ) suggested to me that the DS4000 series might yield >8 bit vertical resolution under certain conditions. Know anything about that? I didn't dig deeper at the time, I'd have to go back and find whatever reference that was. Maybe we already have >8 bit scopes under the correct circumstances.

OK, I couldn't stand it so I went looking and found it. Straight off of Rigol's website, the DS4000 User's Guide says this:

when ≥5 μs/div @ 4 GSa/s (or ≥10 μs/div @ 2 GSa/s): 12 bit resolution

Here's the link to the spec sheet, it's right on the first page (which is actually chapter 17 of the user's manual): http://beyondmeasure.rigoltech.com/acton/attachment/1579/f-0169/0/-/-/-/-/file.pdf

However, a couple of pages later it says this:

Vertical Resolution Analog channel: 8 bits, two channels sample at the same time

[0culus]

Honestly, if a higher end DSO that's got better vertical resolution would make you happier, my recommendation is to save up and buy the one you want.

[DaJMasta]

Really really quiet with a smart fan.

OK, forgot to mention silence. The Rigol is a bit louder than I'd like though I haven't been tempted to swap for a quieter fan (if that's possible while keeping the CFM).

That Siglent sounds tempting.

Never made mine fully silent, but swapped out the original fan for a Scythe (if I remember right) of slightly higher CFM rating, the slapped some VGA memory heatsinks on the ADCs (or was it the acquisition memory management chips.... one has a heatsink, one doesn't in the original design but both get hot), and it's been going great for... 3 years now?  You can't really fit a longer fan because there isn't much clearance, and without modifying the shielding you get some noise from the obstruction of the airflow, but it did make an improvement over the stock fan for me.

The new gen Siglents aren't full 10 bit but they do seem to have a lower noise floor than most 8 bit options, so eres mode is less noisy and can give you good results, and while I haven't used the Rigol 5000/7000 series scope interfaces, I was partial to the Siglent UI layout and color grading, so it's at least a strong contender.  I'm glad vendors are thinking about fan noise now, makes a big difference to me especially with something that's going to be on a lot on the bench.

[EEVblog]

I think the killer feature I'd want above others is 10 or 12 bit vertical resolution....

Agreed, that's why I mentioned 10 bits specifically.

Yes, you wouldn't upgrade a Rigol 4000 unless you were going higher bits or something else exotic like GHz region BW

[EEVblog]

Yeah well the new 4ch 500 MHz Siglent is $5,689 however its baby 350 MHz brother is a more digestible $3,569 and there's a BW upgrade option for it........no surprises for guessing what I'm thinking.   
Just unpacked one before it goes to a customer and first impressions are really good.
Really really quiet with a smart fan.

Mine's not "really really quiet". It's ok, but not that good.
A few bugs so far...

[tautech]

Yeah well the new 4ch 500 MHz Siglent is $5,689 however its baby 350 MHz brother is a more digestible $3,569 and there's a BW upgrade option for it........no surprises for guessing what I'm thinking.   
Just unpacked one before it goes to a customer and first impressions are really good.
Really really quiet with a smart fan.

Mine's not "really really quiet". It's ok, but not that good.
A few bugs so far...

FW version ?
The one I have has 0.8.2R1 installed.

[tggzzz]

Dropping back bandwidth (from 500MHz) feels wrong, unless there is a spec that is more important to you - resolution would be the best case.

If you have money and time available, why don't you consider getting a completely different class of test equipment, e.g. a decent logic analyser, or a spectrum analyser?

That might enhance your capabilities and be more interesting.

[2N3055]

Good question.

You could upgrade to second gen Rigol/Siglent 500MHz scopes.
That would give you color grading, histograms, better decoding and generally better analysis options.
They might have some bugs now, but so had (and still does some)  R&S..
They would be better rounded instruments.

On the other hand, as some suggested "save a bit more and go 10 bit",  I just love when people throw statements like that.
R&S have nice scopes but for 4000USD target you would have to go back to 100MHz.  500MHz RTM3000 with options is more than 10000USD list price + tax.
So no, it's not a bit more saving..

OTOH, as you say yourself, if you treat a scope simply as scope, and don't use much of analysis stuff (advanced math, FFT, decoding etc), you really won't gain anything.
DS4000 has nice big screen, good bandwidth and works just fine as general use scope.
If I were you, I would sit down and put down on a piece of paper what I do (or plan to do) and see if there is other instrument I don't have at all that would benefit me more than getting something I already have and that I won't be using to a full potential. Spectrum analyser, multimeter, signal generator (RF or AWG), VNA ?



Regards,

[EEVblog]

FW version ?
The one I have has 0.8.2R1 installed.

Same

[maginnovision]

Good question.

You could upgrade to second gen Rigol/Siglent 500MHz scopes.
That would give you color grading, histograms, better decoding and generally better analysis options.
They might have some bugs now, but so had (and still does some)  R&S..
They would be better rounded instruments.

On the other hand, as some suggested "save a bit more and go 10 bit",  I just love when people throw statements like that.
R&S have nice scopes but for 4000USD target you would have to go back to 100MHz.  500MHz RTM3000 with options is more than 10000USD list price + tax.
So no, it's not a bit more saving..

OTOH, as you say yourself, if you treat a scope simply as scope, and don't use much of analysis stuff (advanced math, FFT, decoding etc), you really won't gain anything.
DS4000 has nice big screen, good bandwidth and works just fine as general use scope.
If I were you, I would sit down and put down on a piece of paper what I do (or plan to do) and see if there is other instrument I don't have at all that would benefit me more than getting something I already have and that I won't be using to a full potential. Spectrum analyser, multimeter, signal generator (RF or AWG), VNA ?



Regards,

Right now he could get a fully unlocked 1Ghz RTM3K for 10K, but I agree, a new scope likely isn't worth it for him. Even the RTB2K fully unlocked is 3700 and would require he lose 200Mhz bandwidth.

[2N3055]

Right now he could get a fully unlocked 1Ghz RTM3K for 10K, but I agree, a new scope likely isn't worth it for him. Even the RTB2K fully unlocked is 3700 and would require he lose 200Mhz bandwidth.

Well, that would be an upgrade, but at the price that is 250% over his budget, a full 6000 USD more.
That is what I said, that is not "a bit more saving"...

In my country there are many people that work for 500USD a month. 6000USD is somebody's yearly salary.
I'm sorry, that is why "a bit more" statement rubs me the wrong way..

Any way, rants aside, I agree with you, even if it was much less to "upgrade" it is not an upgrade if you don't gain new measurements and usability enhancements that you will actually use..
That is why I shared my experience that most of the time, you get more if you invest money into capabilities that you really miss and not incrementally upgrading what you already have. Unless your existing equipment is obviously limiting to your work, in which case it is obvious that you need to upgrade.

As far as scopes, to go further, if you have SA , FFT migh not be really important. If you have LA, you will decode better with that.  Etc etc..
I have MSOX3000T as fast general purpose scope, supplemented with Pico 3406DMSO for 512 MPoint long mem, and a 16 bit Pico 4262 for low level, low noise, low frequency measurement (less than 8uV noise floor  ). I also have Zeroplus LAP-C 16128.
There is not much I cannot see, measure or decode with that combination in my work.
That is my perfect scope. There is no instrument that does all that in one box, at least not at the prices that are not paid by Department of Defense.

[tv84]

Yeah well the new 4ch 500 MHz Siglent is $5,689 however its baby 350 MHz brother is a more digestible $3,569 and there's a BW upgrade option for it........no surprises for guessing what I'm thinking.   

If I had $4000 this option would be the one I would take, because in this price range it's the only one worth it. It's the only option available ATM.

With more money, you could get R&S, etc.

So, my 2 cents: if a SDS5000X is much better that yours, go for it. If not, keep your money for other things.

[YetAnotherTechie]

2N3055 Can any of those LAs decode in real time?

Thank you

[ebastler]

It's been feeling like time to upgrade my DSO.

Frankly, I have not seen any reason in your posts why you feel it is time to upgrade? The Rigol you have seems to meet your needs just fine. If you have spare money burning a hole into your pocket, I'm sure there are other desirable gadgets? 

[2N3055]

2N3055 Can any of those LAs decode in real time?

Thank you

What do you mean? Firstly, only LAP-C is LA, Pico and Keysight are MSO-s.
Secondly, what is real time? Faster than you can read? We had discussion about that many times.
Any LA and scope must not miss triggers. You capture and you read if you have hundreds of packets, you stop and read next half an hour. User is slowest part of the system. Do you mean connect to live SPI bus and visually look if there is something going on? Yes they can do that, but that has limited use.

[Mechatrommer]

Yes, you wouldn't upgrade a Rigol 4000 unless you were going higher bits or something else exotic like GHz region BW

and GHz region BW is not anything near $4K unless you (the OP) are willing to buy used unit. DS4024 is already $4K, so trying to spend another $4K i dont call it an upgrade

[IDEngineer]

Frankly, I have not seen any reason in your posts why you feel it is time to upgrade?

I agree, which is why I said "What problem am I really solving? I don't beat the table every day cursing the 4024 because of some fatal flaw." Maybe the biggest reason is "The 4000 series is getting a bit older and so it might be nice to freshen my single most important piece of test equipment."

About a year ago I purchased a new laptop. Not because my old one had anything wrong with it, but because I refuse to use anything beyond Win7 for as long as I can hold out, and laptops that either ship with Win7 or have Win7 drivers available from their manufacturer's website are becoming increasingly rare. This scope decision is not identical, but you get the parallel... now that test equipment is so heavily software based it's not just about the raw hardware specs like bandwidth and noise floor and resolution. The software environment is a huge part of what you're buying in scopes these days and I'm wondering if maybe it's time to move forward. Like the laptop/Win7, a purchase decision driven by market forces rather than strictly technical issues.

DS4024 is already $4K, so trying to spend another $4K i dont call it an upgrade

That might be the MSRP, but mine cost $2600 new straight from Rigol (not clearance, an actual brand new unit) with all decode options. They're $3K right now on TEquipment not counting the EEVBlog discount, and I see them for around my same $2600 purchase price frequently. The value of an object is its replacement cost, not (necessarily) the MSRP. So this isn't a $4K scope. It's a $2600 scope. (Yes, the 500MHz upgrade theoretically adds value but anyone can buy that same $2600 scope and use the free key generators to unlock the max bandwidth, so again - if it can be duplicated for $2600, it's worth $2600.)

DS4000 has nice big screen, good bandwidth and works just fine as general use scope... you wouldn't upgrade a Rigol 4000 unless you were going higher bits or something else exotic like GHz region BW...

That's kinda the conclusion I'm reaching. It almost feels like the DS4000 series is something of a sweet spot. As I explain above, you can literally buy a brand new 500MHz 4 channel 4GSa/sec DSO with every option enabled for $2600. I cannot find anything in the market to match that price/performance. Looks like that Rigol is going to be on my bench for quite some time.

Thanks for everyone's comments, I really do appreciate them.

[Mechatrommer]

I agree, which is why I said "What problem am I really solving? I don't beat the table every day cursing the 4024 because of some fatal flaw." Maybe the biggest reason is "The 4000 series is getting a bit older and so it might be nice to freshen my single most important piece of test equipment."

i dont think "upgrade" is the right term. i think it is "replacement". such as replacing a dying scope. but since its not dead, if its me, i will use it to its last breath and drop of blood. once its dead then buying (replacing to) another scope it will be. my DS1052E that came back to me gave up its monitor, maybe because it was looking at its replacement (upgrade) DS1054Z unit thats currently sitting on my main working bench right now. i have no time yet to repair.

About a year ago I purchased a new laptop. Not because my old one had anything wrong with it, but because I refuse to use anything beyond Win7 for as long as I can hold out, and laptops that either ship with Win7 or have Win7 drivers available from their manufacturer's website are becoming increasingly rare. This scope decision is not identical, but you get the parallel...

i feel you, thats why i will hold on to my old pc bought around 10yrs ago as long as it can. when the time comes to its last breath, i hope there is still motherboard that i can install WinXP in it, i'm lucky i still have the original WinXP CD working, i've made several backup for it in case the original CD got damaged. i really hate newer Windows file explorer and search utility, thats why i hesitate to upgrade. not that i havent try them, i did, and i have installation cd/copy for every newer Windows, even bought a Win10 license not so long ago, what a waste, but they are going in the shelf for now. until i have Windows with better or same easeness as WinXP's Explorer and Search, they are backbone to my workflow, young players will not appreciate nor get the chance to experience.

[cruff]

OK, I couldn't stand it so I went looking and found it. Straight off of Rigol's website, the DS4000 User's Guide says this:
when ≥5 μs/div @ 4 GSa/s (or ≥10 μs/div @ 2 GSa/s): 12 bit resolution

...
However, a couple of pages later it says this:

Vertical Resolution Analog channel: 8 bits, two channels sample at the same time

That's because the 12 bits of resolution are produced by the high resolution sampling mode via averaging.  Each input channel still only has 8 bits of ADC.