Should I buy a Rigol MSO5000?

[nctnico]

But until then, nctnico and his vendetta against hobbyists buying cheap scopes that offer extreme value for money will always baffle me.

Sorry but you are absolutely wrong here. If you like it, you like it but wait until you need advanced feature like protocol decoding to work. What baffles me is that people buy half baked products and pretend they are equivalent to equipment costing 5 to 10 times more.  Rigol will need years to fix it. The threads about the DS1054Z, DS2000 and DS4000 series are all the proof you need.

If Dave can spot several simple issues right of the bat then there are many more lurking because Dave's testing isn't thourough to begin with (and that is OK because really thourough testing would result in a long winded boring video).

Edit: besides that: the Rigol MSO5000 isn't an entry level or cheap scope to begin with. They want a serious amount of money for it.

[rsjsouza]

Read and weep....Dozens of bugs with basic functions ...
It was officially released March 2017.

Dozens? Where? The known issues from the latest release notes only lists two:
https://scdn.rohde-schwarz.com/ur/pws/dl_downloads/dl_firmware/pdf_3/RTB2000_ReleaseNotes_v02.202.pdf

nctnico, I think his point was that I counted 55 bugs squashed after the first firmware. If anything, it was far from perfect when it was released and there were several basic issues.

The remaining triggering bug is almost not even worth mentioning: it says it happens in "rare cases": how rare it may be debatable, but being on the other side of the fence (support and development) I can tell these types of issues may never be addressed due to resources being diverted towards more pressing bugs or features (or even entirely new platforms).

As I mentioned before, any new platform needs time to mature, either from R&S or Rigol. What counts is the time to react. So far, the ball is in Rigol's court with the MSO5000 and the comparison could be R&S and their metric of 55 bugs fixed in two years (among new features). How many were fixed in the same timeframe on the DS1000Z platform? What about the DS2000? I can tell that, within 4~5 years after market introduction, the DS4000 reached the status of a very good and useable platform - not perfect, but all very important bugs were squashed.

[nctnico]

What counts most is the severity of the bugs. If a bug causes an entire feature not to work then it is much more problematic than missing a button press every now and then.

Edit: And perhaps R&S isn't the best example. According to what I can find on the Google GW Instek announces the GDS2000E series beginning 2015. When I wrote my review (published februari 2016 https://www.eevblog.com/forum/testgear/gw-instek-gds2204e-(200mhz-4-channel-dso)-review/msg855862/#msg855862 ) they had a pretty much mature product and where able to fix the bugs I had found within weeks.

[2N3055]

Read and weep....Dozens of bugs with basic functions ...
It was officially released March 2017.

Dozens? Where? The known issues from the latest release notes only lists two:
https://scdn.rohde-schwarz.com/ur/pws/dl_downloads/dl_firmware/pdf_3/RTB2000_ReleaseNotes_v02.202.pdf

Dozens from release. Two (that I mentioned) are what they officially admit knowing about but didn't fix yet. One of those is from initial release 1.0.0

Other dozens are list of "improvements" (fixed bugs)
On release it had more bugs than MSO5000. Just read release document.

But they are R&S and they and they own their mistakes and fix them eventually. And everybody think it is Ok to be understanding towards them and their bugs are Ok.

In two  years (March 2021) we'll reconvene  here and see if Rigol fixed their errors.  That is fair, that is level field, that  is not xenophobia, sinophobia or whatever crap is this smearing campaign here.

If they didn't I will apologize to you and agree you were right.

In meantime, you have nor right nor reason to categorically state  "they will never fix it", unless you are clairvoyant and can see future..

[nctnico]

I never said they'll never fix it. The fact however is that there seems to be a relation between sales and the frequency of firmware releases by Rigol for a product. The MSO5000 sits in a higher price bracket so they have to rely more on sales to businesses than hobbiests. If sales aren't what they are expecting then bug fixes will be slower or they might even abandon it. Time will tell but be honest: do you want to take that chance? Can you postpone buying a piece of equipment you need now by several years? Or put differently: do you want to buy a product which doesn't do what it is specified to do but you might find out when it is too late? There is a reason why I tested the crap out of the GDS2204E: I wanted to be sure it wouldn't throw me a nasty surprise at the worst possible moment like the Siglent SDS2000 did.

[2N3055]

What counts most is the severity of the bugs. If a bug causes an entire feature not to work then it is much more problematic than missing a button press every now and then.

And perhaps R&S isn't the best example. According to what I can find on the Google GW Instek announces the GDS2000E series beginning 2015. When I wrote my review (published februari 2016 https://www.eevblog.com/forum/testgear/gw-instek-gds2204e-(200mhz-4-channel-dso)-review/msg855862/#msg855862 ) they had a pretty much mature product and where able to fix the bugs I had found within weeks.

Severity of bugs: decodes not working properly and reliably, waveform not shown properly (scaling, offset and artefacts), timebase not working reliably, trigger errors, etc etc.. Just read.. Most of them were serious basic functioning of the scope.
Not "pluses".
But they are fanatastic serious company and they fix things and now, after two years it is getting to be reliable. But it was not reliable as released. It was easily worse or at least similar to MSO5000.
I'm not arrogant enough to postulate that Rigol engineers are stupid and incompetent. If anything, they are probably fantastic, and certainly know their job. But releasing new platform, new ASICS, new everything is hard even for the likes of Keysight.

R&S is actually new benchmark how bad can be something at release, and how it can grow to something great.
So if R&S can have 2 years to make it right so can Rigol..

I wish none of them would release unfinished beta class products. But if western companies are allowed to do that, than so can Chinese..

I would like that A brands would prove me wrong and be actually better at this. But they are not..Sometimes they are even worse, especially considering prices they are charging...

[nctnico]

R&S dropped the ball for sure but they put a very nice user interface together on the RTB2000 and related oscilloscopes. Stuff like being able to drag the split screens for example. They probably bit off way more than they could chew and then management kicked in to shove it onto the market. The RTM3000 got released later and that didn't had so many issues.

[Fungus]

The recommendation to hack it to 350 MHz is strange from a professional point of view, because it voids warranty

The hack can vanish just as easily, leaving no trace.

without getting anywhere close to what it should be.

 

It enables all possible features/upgrades, just like official paid-for options do.

[Fungus]

They probably bit off way more than they could chew and then management kicked in to shove it onto the market.

Just like the Rigol MSO5000, then? It's obvious they rushed it our for Xmas.

Weird how R&S get a free pass when they do this but Rigol doesn't, despite R&S costing much more money

[nctnico]

They probably bit off way more than they could chew and then management kicked in to shove it onto the market.

Just like the Rigol MSO5000, then? It's obvious they rushed it our for Xmas.

Weird how R&S get a free pass when they do this but Rigol doesn't, despite R&S costing much more money

Where does R&S get a free pass?

[2N3055]

I never said they'll never fix it. The fact however is that there seems to be a relation between sales and the frequency of firmware releases by Rigol for a product. The MSO5000 sits in a higher price bracket so they have to rely more on sales to businesses than hobbiests. If sales aren't what they are expecting then bug fixes will be slower or they might even abandon it. Time will tell but be honest: do you want to take that chance? Can you postpone buying a piece of equipment you need now by several years? Or put differently: do you want to buy a product which doesn't do what it is specified to do but you might find out when it is too late? There is a reason why I tested the crap out of the GDS2204E: I wanted to be sure it wouldn't throw me a nasty surprise at the worst possible moment like the Siglent SDS2000 did.

What you went trough with Siglent was bad. They screwed you bad, and you are right to be mad. But even Siglent made a lot of improvement in that area. But I agree with you that people need to be cautious.  If they need instrument they have to heavily depend on, they should test right know if instrument does what they need right now. I don't care about promises if I need it NOW.

But that is applicable to every manufacturer. I didn't ( and I advise anybody not to ) buy R&S RTM3000 because it doesn't have basic search on segmented memory and on  such basic protocols as UART, I2C and SPI... You get 400 Msamples worth of I2C in 5000 segments and than you are supposed to go manually trough all them to find a single packet there..
I would like to meet an idiot who thought that was good idea.

So I got Keysight 3000T that has pathetic memory but at least everything works and is a pleasure to work with.
But it is so stable and bug free not because Keysight is God given.. No, it is good because it is almost 10 years old platform that was being debugged all this time. It's revision history is also full of bugs and it's 27 pages long... Result?  Reliable instrument.
But if I need long sample memory, I have to use my Picoscope.  But that is OK, no single instrument can do it all..

3000T also have stupid things:
. there are no histogram measurements. MSO5000 has it..
. you can count pulses only on analog channels, but not on digital where you need it more.
. hardware decode is great for speed. But you cannot apply and/or change decodes afterwards. So you cant just capture something and fiddle with settings until you setup it all right. You have to keep a running captures until you are happy with setting and than capture one you need. Which sometimes you can't do. In which cases software decodes are better. Cue in Picoscope...
etc.

Nobody is perfect.

[tautech]

I never said they'll never fix it. The fact however is that there seems to be a relation between sales and the frequency of firmware releases by Rigol for a product. The MSO5000 sits in a higher price bracket so they have to rely more on sales to businesses than hobbiests. If sales aren't what they are expecting then bug fixes will be slower or they might even abandon it. Time will tell but be honest: do you want to take that chance? Can you postpone buying a piece of equipment you need now by several years? Or put differently: do you want to buy a product which doesn't do what it is specified to do but you might find out when it is too late? There is a reason why I tested the crap out of the GDS2204E: I wanted to be sure it wouldn't throw me a nasty surprise at the worst possible moment like the Siglent SDS2000 did.

What you went trough with Siglent was bad. They screwed you bad, and you are right to be mad.

All that was bad was no timely fix for issues Nico identified and he wasn't prepared to wait for them to arrive.
I get that frustration ^ and the ongoing mentions by Nico  of his early experience with Siglent has been partly responsible for a significant change in culture.

But even Siglent made a lot of improvement in that area.

Exactly.

Nobody is perfect.

No scope is perfect. We will never see one that is.

[dietert1]

In the area of digital scopes LeCroy is the definition of perfect, because they started building them about 40 years ago, when nobody else except HP had the faintest clue what a DSO could be.
When you think you need a car, but you can't afford a new one, you try to get a used one that fits. The recommendation to buy a bicycle instead is ridiculous. Especially if you already had a bicycle.

OT: Three months ago i happened to recap the power supply of our LeCroy 9354A. It is connected to a PC via GPIB and produces TIFF files with  colored traces on the press of a button. When we bought it 10 years ago it came fully equipped with software options. It does not decode serial data streams though. Those scopes currently sell for about US$ 600. The 1 GHz version would be 9374 or 9384.
Our 9354A has a trickle charger for the clock battery, so we can leave it off for years and it will turn on. It also got quiet fans. And it got two 1 GHz active probes.

Regards. Dieter

[Noy]

The handling/useability of  LeCroy is so uncommon many people doesn't like it.
We have a "old" DPO 7254 and a "new"  SDA820Zi-A beside some smaller Hameg/Tektronix and some everyday / everybench 1054z Rigols (enough to check powerup sequences / switching regulator / Low Speed stuff ) use and nearly everyone tries to measure most Highspeed things with the DPO because the handling is much more intuitive then the LeCroy..

Especially the active probes from LeCroy are worse. This tiny fiddly probes with their lego stuff and the tiny resistors you have to solder to your highspeed lines and these resistors are "unsolderable" they won't hold.. Just sneeze beside them and they are Off again... Crap!!!

[Fungus]

When you think you need a car, but you can't afford a new one, you try to get a used one that fits.

Says who? Many people in that situation choose to buy a smaller car and accept the fact that they don't have leather seats and can't fit a sofa in the back.

Buying an old banger just in case they might go into the removals business one day just leads to higher maintenance bills and a smell of old dust every time they get in it.

[dietert1]

My proposal is considering a scope with active probes, if someone wants to work above 100 or 200 MHz. Our LeCroy AP020 1GHz active probes are as easy to use as a passive probe. They just get a little bit warm. Do we need to discuss the difficulties of probing at 10 GHz and how a LeCroy 10 GHz scope sounds like a vacuum cleaner?
Recapping our LeCroy 9354A happened after the arrival of the Rigol DS2202E and nobody knows whether the Rigol will survive ten years. Even with it's age of more than twenty years, our LeCroy 9354A has its advantages, for example the active probe interfaces. It also got a diode gate made of a 50A bridge rectifier in its protective ground connection, so it has a +/- 1.2 V ground free margin and less noise with small signals. It came to us from another lab and the only reason we got it was it did no longer power on. It doesn't smell at all.
I am just demonstrating how $1000 may get you a 500 MHz DSO with two active probes, nothing more.

Regards, Dieter

[nimish]

But until then, nctnico and his vendetta against hobbyists buying cheap scopes that offer extreme value for money will always baffle me.

Sorry but you are absolutely wrong here. If you like it, you like it but wait until you need advanced feature like protocol decoding to work. What baffles me is that people buy half baked products and pretend they are equivalent to equipment costing 5 to 10 times more.  Rigol will need years to fix it. The threads about the DS1054Z, DS2000 and DS4000 series are all the proof you need.

If Dave can spot several simple issues right of the bat then there are many more lurking because Dave's testing isn't thourough to begin with (and that is OK because really thourough testing would result in a long winded boring video).

Edit: besides that: the Rigol MSO5000 isn't an entry level or cheap scope to begin with. They want a serious amount of money for it.

My dude, the MSO5074 is $999. That's not a lot, it's cheaper than my phone, but it is a stretch for a *hobbyist*. The alternative for a *hobbyist* is _not_ a $5000 Keysight scope, it's a $300 cheapass chinese scope or nothing for anything above 200MHz. For the price, there is *nothing* at the level of the MSO5074 + hacks (which as a hobbyist doesn't matter) + a hack pod or real one ($400).

If you paid $5000 for an MSO5074 there's a bridge in Brooklyn I'd like to sell you. But we're talking about a hacked $1000 scope, where such things like minor bugs are ok, because you saved thousands.

This is what I don't get; you are talking about a whole other world in terms of people who need absolute reliability -- in which case, spend the $3k, $5k for quality support (it'll pay for itself!) vs a Chinesium grade scope. But I am, and many others in the thread are, in the hobbyist world where outlay cost matters far more than a few extra hours of time.

[Fungus]

This is what I don't get; you are talking about a whole other world in terms of people who need absolute reliability -- in which case, spend the $3k, $5k for quality support

My question is: If a 'scope costs that much money then why does it need any support. Surely it should be completely bug free, easy to use, reliable, nice manual, etc.

Where does "support" enter into it? 

[nimish]

This is what I don't get; you are talking about a whole other world in terms of people who need absolute reliability -- in which case, spend the $3k, $5k for quality support

My question is: If a 'scope costs that much money then why does it need any support. Surely it should be completely bug free, easy to use, reliable, nice manual, etc.

Where does "support" enter into it? 

When you run into a situation where you just can't take the time to learn the scope and need/want someone to help with something. 10min of talking to an expert can replace a week of spinning in circles.

If you don't need/want that, I don't see the point in spending the money on ultra high end scope gear, because that's what actually costs real money.

[iMo]

The biggest mistake one can make (when looking for an o'scope or 6+ digits DMM like me) is to join eevblog and start to read the relevant threads.

5-6 years back I was discussing with my local seller a purchase of rigol's DS4014 and rigol's xxxx DMM.

Unfortunately, I discovered the eevblog and started to read the threads.

I've been not able to decide on an o'scope and a DMM.

Even worse, it seems to me, more and more, all o'scopes and DMMs, whatever brand, are boxes full of worms.

 

[TK]

I just received the DS1054Z... it is a nice scope and more responsive than the MSO5000 I tried a couple of weeks ago... In my opinion the MSO5000 is a downgrade in terms of display quality, UI responsiveness. 

I still think that something fundamental changed at Rigol in the last couple of years... from being Rigol Technologies, now the name is Rigol xyz Technologies (I don't remember xyz, it is in the back label of the scope).  Probably the original Rigol owners sold the company and now it is being managed by a private equity company that does only care for the money.

[Martin72]

I think nearly the opposite about, after working with the 5000 for "so long" and everyday with the possibility to compare it with a DS1054Z ( we have 3 in the labs), siglent sds-1104x-e ( one), several lecroy models and two tektronix.

Display is weak, no doubt about it, but in my opinion the UI/ menue structure is the same - same complicated.
The menues of a waverunner ws 3024 was "heaven" against, so straight, so easy...
Beside of, the 5000 got much more powerful hardware, the screen ist full usable, menues can be hidden ( this is what I hate most on the ds1054, the permanent menue left and right) and, even it´s very dim, the benefit of a bigger Display - 7" wasn´t suitable anymore for an mid-40s old man like me....

[Martin72]

By the way:

The menues of a waverunner ws 3024 was "heaven" against, so straight, so easy...

This model of lecroy got a bright, big Display, a hefty look and feel and some typical lecroy features....but it´s awful slow in it´s response and got some hard issues like totally frozen after pressing the find triggerlevel button or the missing of an free chooseable measure gate.
It got "countless" firmware updates so far, last from february this year, and still got these issues.
And it costs triple as a mso 5074....so much to say about a so called A-brand.
Actually, the rigol 5074 leave this one clearly behind in nearly all cases, except look and feel, display and integrated 50 ohm termination.

[dietert1]

For those who don't know LeCroy i prepared an example measurement from a USB2 audio transmission. Then two zooms into two packets. Then two gated frequency measurements in the sync fields of the two packets (0101.. sequence gives 240 MHz). One measurement was taken with the 9354A i mentioned before.
The other one is with an 11 year old WR 64Xi which currently sells at a similar price as the 350 MHz version of the Rigol 5000.
Maybe someone can try to do it with a Rigol 5000.

Regards, Dieter

[Mr Nutts]

For those who don't know LeCroy i prepared an example measurement from a USB2 audio transmission. Then two zooms into two packets. Then two gated frequency measurements in the sync fields of the two packets (0101.. sequence gives 240 MHz). One measurement was taken with the 9354A i mentioned before.
The other one is with an 11 year old WR 64Xi which currently sells at a similar price as the 350 MHz version of the Rigol 5000.
Maybe someone can try to do it with a Rigol 5000.

Regards, Dieter

Be careful, the last guy who talked a lot about Lecroy has been banned   

It's also his fault that I bought a Waveruner LT264 and a LT574 and I'm happy with them   Although I now also have an Agilent DSO8064 which is my main scope mostly because it has so much memory (128M) and can decode I2C

I think I paid $700 for the LT264 and around $1k for the LT574.

Durability should not be an issue, apparently they were manufactured in Japan for Lecroy. I've seen older Lecroy scopes with crt which had a plastic housing and they often look damaged. But these Waveruner LT scopes are solid like bricks

And if something breaks then schematics are available

I learned so much just by using them (I guess the explanations from the guy who talked me into them did help a bit), they have very powerful functions of which some are still beyond my grasp

The only thing I really miss is decoding but that's what I have the Agilent for

The interface of the Lecroy scopes I have is different from the Tektronix scopes we had at college but it's pretty straight forward to use. Sometimes I mix up horizontal and vertical controls but that's it. And if I can operate these Lecroys then pretty much anyone should be able to do it