DS1074Z 4 Channel or DS2072 2 Channel Oscilloscope?

[Williamscullen818]

Is the DS1074Z any less useful for most applications? What I mean is, do you think the slight boost in specs on the DS2072A warrant loosing the 4 channel option? I know in terms of useability, the DS2072A is a better choice, but for instance if I was looking at 2 traces (input and output of something) will the DS2072A give me anything other then useability improvements or is there a serious improvement in performance? I seems like it when you look at the insides of the 2 scopes, seems like there is much more stuff inside the DS2072A, not to mention better quality.

On the other hand, The price point of the DS1074Z is about the same as the price of an option of other more expensive models. If I decide I really needed a 4 channel scope, and logic analyzer won't do the trick I could spend another $600 to get a DS1074Z. It's like purchasing an option for the DS2072A down the road, but you get a whole new scope...

[leppie]

Is the DS1074Z any less useful for most applications? What I mean is, do you think the slight boost in specs on the DS2072A warrant loosing the 4 channel option?

I am in the same 'dilemma', but it seems pretty easy to get an extra 2 channels by using another scope, say a cheap analog one using the the trigger out from the Rigol and the ext trigger in on the analog. Sure you will lose the memory aspect on the extra 2 channels, but not a biggie IMO.

I have an Open Workbench Logic Sniffer and it works well enough. Not realtime, but not really an issue either.

This thread and the slowness I have seen from the DS1000Z has made me choose the DS2000 instead. Now just to save some money for it.

[cidcorp]

Well I do own both a DS2102 and a DS1047-S as well as a Logic 16 from Saleae.  When the DS2000 came out, I really wanted the scope but was disappointed that
there we no options to purchase with 4 channels.  After waiting it out for a while I needed to get a replacement for an aging digital scope I had from GW-Instek,  I went
ahead and bought the DS2102 (with the knowledge that the scope *might* be hacked in the upcoming months like the older 1000 series), I went with the mid range one
just in case the hacks never materialized. I was still regretting that the scope didn't have the 4 channels.

I bought the Logic 16 cause I knew most things wouldn't cut it on the 2 channels scope - and I do like this product. The Logic software is well written and features have been
added over time so it feels like it was a good investment.

The DS1047-S is new, just got it and I have to say it's like the 'mini-me' version of the DS2000, with the 2 extra channels the DS2000 should have got.
Haven't done any decoding with it yet so the conclusion is out for now on that, but I can say it is much slower feeling and I HATE that you cannot hide unused menus
from the right and left of the screen, you lose sooooo much real estate.

As for the options: DS2072 (hack it to 200Mhz) will run you about $839, plus the logic 16 another $299 = $1138 for the two, versus the DS1047Z (hack it to 100Mhz) at $585 or $818 if you want to add the built in 2 Channel 25Mhz Waveform Generator, add the Logic or Logic 16 if you need it down the line and the scopes not cutting it - you'll save the $299 for now.

I figure eventually down the line we'll be seeing a MSO2072 or a MSO2074, cause we've already seen the unpopulated sections on the PCBs from Rigol.  If you go DS1074Z and wait you can sell the DS1074Z and buy up to the Mixed Signal version when it comes out.

OR as someone else joked  , Buy both (and the Logic 16)... works for me.

Chris

[Williamscullen818]

There really should be a 4 channel DS2000 option, I would have paid triple what I'd pay for the DS1074Z for one of those. You would think they would be working on one, I'm sure they know they would sell many of them. Does anyone know if the display in the DS1074Z is driven via an FPGA or from the freescale chip? I'm not so sure it is, cause I only see 1 fpga in the device which is doing acquisition with the ADC, but maybe they are driving the lcd in there as well.

I've got a opportunity to get an Agilent DSOX2004A for $1500 cause of a student discount, only think holding me back is the cost of the options and no hacks right now. The Embedded serial option alone is $500, which is about the cost of a DS1074Z. I could probably strech my budget to get that scope, but no sure if it's worth it. Anyone think I should do that over one off the Rigol scopes? That is a price point question tho.

[Williamscullen818]

Another question about decoding on DS2072A... can the scope use the ref waveform as the clock in spi decode mode? I was thinking I could capture one cycle of the clock then put it on a ref channel, then I can use both 2 channels for data +/-.

[Williamscullen818]

cidcorp,
If you had to do some spi debugging, would you grab your DS1074Z, or would you use your DS2000, even though you would have to switch between mosi and miso to decode the data/verify the waveform? What would be your go to scope for something like that?

I'm not taking about the logic analyzer, I'll be getting one of those anyways I think for on the go with the laptop.

[Mark_O]

I don't understand your point other than you want to be condescending towards me.

Oh, no, you've got it all wrong.  It's not just you... I like to be condescending towards everyone .  J/K 

Actually, what I was trying to do was mock you.  Because you were being ridiculous, and arriving at bogus conclusions.

Regardless of the price of those scopes, you're talking about a $4000 Rigol scope that's isn't recording (at least I think it wasn't recording, maybe it just wasn't displaying it all but still captured it?   I'm unsure, it doesn't really explain it, but that's what I assume, If I'm wrong, then well, I'm wrong) all the serial data coming from an Arduino, yes, an Arduino

(highlights added by me)

And there's the rub.  You were wrong.  Totally and completely wrong.  And the problem from that is evident... (some) others will read that, take it at face value, and promulgate it as gospel.  You don't have to wait for it, or look far.  The follow up comments are right below your own. 

"its less than useless."
"I would throw that scope right out of a window."
"frustrating beyond belief."

And yes, I know these were all conditional comments.  IF, then.  Yet they were all speculative commentary, generated from your own, uninformed and incorrect conclusions.  The Rigol does not drop the actual communications stream.  That's all captured intact, and at speeds far beyond anything an Arduino normally generates.  But it's too slow (pig slow, from that video) to keep up with display of the decoded version.  In fact, from the very slow speed of that I2C bus traffic, I'd say it's embarrassing.  But that's why I tried to point out that this isn't necessarily the way one would spend most of their time examining such comms traffic.  (If it was, you'd be silly to buy anything less than the Agilent.)

People will read and pass on what they see posted here.  So the word will be that the Rigol is a POS for serial data analysis and protocol decoding, that can't even keep up with an Arduino!, for god's sake, and will drop comms data left and right.  And that's what they'll remember, and repeat to others.  "Didn't you hear?  The Rigol is crap...".  When you don't know WTF you're talking about, it's not reasonable to post conclusions, then say, "I think" and "I assume".  That's just irresponsible.   

Which is why I was trying to make fun of your comments, in a light-hearted way.  If you chose to take umbrage, be my guest.

[Mark_O]

I normally like to set up and capture a sequence of traffic, especially when I have a device with a deep memory capacity.  Then go back and step/slide through the exchanges, go backwards if needed, etc.

If I understand this correctly it is the reason that tipped me from the Rigol DS1000Z to the Rigol DS2000.  The ability to fairly quickly and easily navigate during the review process (with the big Nav knob and related features/buttons/controls on the DS2000) is IMO a pretty valuable advantage of the DS2000 vs. the DS1000Z.

I'd definitely agree with that! 

Even better would be if Rigol would add some Search and Marker capabilities to the DS2000.  When you've got the exceptional memory depth capabilities of such units, having a way to find a few "needles in the haystack", and do so quickly, would be phenomenally powerful.  IMO (and others may disagree), this might even be MORE useful than having exceptionally high wfms/sec refresh rates, which only increase the probability of seeing an abnormality in real-time.

They've already got all the selective Event/Trigger Defining capabilities in place.  They just need to let you use them in a post-process, to tag & flag all the "interesting" occurrences in the current sample set, already in memory.  Then let you step forward and back through them.  Wow, the time that could save!

Tektronix has been doing this for years.  Agilent does it.  Hameg does it too, even on their HMO units costing less than the DS4000.  It would be nice to see Rigol catch up.

If the next version of the DS1000Z was a MSO1000Z with 4 analog channels, 8-16 digital channels, and the big Nav knob functionality... now that would be cool.

Sadly, I don't think there's any way they could/would add any different knobs to the 1000Z-series.

[Mark_O]

There really should be a 4 channel DS2000 option,

AMEN to that!

I've got a opportunity to get an Agilent DSOX2004A for $1500 cause of a student discount, only think holding me back is the cost of the options and no hacks right now.

Look around the Forum here a bit more.

Another question about decoding on DS2072A... can the scope use the ref waveform as the clock in spi decode mode? I was thinking I could capture one cycle of the clock then put it on a ref channel, then I can use both 2 channels for data +/-.

Unfortunately, no.  Besides, the whole thing with SPI is that interpretation of the MOSI/MISO transitions is totally dependent on the synchronous phase relationship to the Clock (read on rising/falling edges).  Just having "a" REF Clock isn't going to help you out at all.

[Mark_O]

I had asked, "Which logic analyzer would you recommend, that would display decoded serial bus traffic in real-time...?"

Try a used Agilent 16700 series, such as the 16702B, or a Tektronix TLA type analyzer.

Thanks!  Could you give me a pointer to where I'd find the software for CAN decoding for my Tek TLA704?  Either for the 4 analog channels (TLA-7E2) or 136 digitals (TLA-7M4, or -7N?).  That would be great.

[Williamscullen818]

So I ended up purchasing the DS1074Z, tequipment.net told me I could return it within 30 days, so if I find that some features of the scope warrant a change to the DS2072A I can do that. I found that the 4 channel option was extremely difficult for me to give up compared to the DS2072A.

Anyone think that was a mistake?

[Mark_O]

So I ended up purchasing the DS1074Z, tequipment.net told me I could return it within 30 days, so if I find that some features of the scope warrant a change to the DS2072A I can do that. I found that the 4 channel option was extremely difficult for me to give up compared to the DS2072A.

Anyone think that was a mistake?

I don't.  I think you'll probably be very happy with it.  It wasn't long ago that any 4-chan scope at this price was unbelievable.  Much less one with all the capabilities the 1000Z has, and the deep sample capacity.

Folks that need and can make use of a 4-chan scope will notice the lack of 2 full channels way faster than somewhat reduced performance specs and a few features.  However, those without much need for more than 2-chans (or having gotten by for so long with only 2, can no longer remember them), will be much happier with the extra specs and features of the DS2000.

Luckily, there's an option for both camps.

When evaluating what limitations you might find along the way, refer back to Electro Fan's posted experiences.  See if you share any of them, and whether they turn the tide for you or not.  Some complaints folks have about the 1000Z are shared by the DS2000.  While others are unique to it.  However, neither is perfect (but then, neither is the DS4000!).

I ended up purchasing the DS1074Z, tequipment.net told me

Oh, did you order one of their custom $25 cases for it?  Unless you never move the scope off the bench (always possible), they're a very nice accessory to have.  And they're a perfect fit for the 1000Z.

[Williamscullen818]

No, didn't get a case for it. I won't be taking it out often at all. Thanks for the recommendations everyone, this is my first oscilloscope so hopefully in the future I'll be purchasing something that doesn't require as much compromise.

[Mark_O]

There was a neat video showing the serial decoding on a high end Rigol scope vs an Agilent and Tektronix, and the Rigol was missing data every now and then.

I do agree that that was an interesting video.  Interesting enough that I decided to go back, and take another look.  Initially I had been focusing on the price differential between the 3 scopes.  Which made it seem like a somewhat unfair comparison, to me.  This time I noticed it was Part 1, so I went looking for Part 2, which turned out to be even more informative.  Part 2 did some closeup shots, which made it possible to ramp up to 1080P, and see what modes the various scopes were running in.  All were set for 100us/div, but that was the only thing they had in common.

It's important to remember that these modern DSOs are powerful and complicated devices, and it's easy to overlook important details that have a large impact on their operation.  And hence the results obtained, and the conclusions that one draws can easily be compromised.  Here's how they were configured and operating*:

Code: [Select]

Scope Sample     Samples      Relative
Model Rate     Displayed    "Work"

Agilent         500MSa/s 500K       5x
MSO-X4154A

Tek
MSO3054         100MSa/s 100K       1x

Rigol
MSO4054        2000MSa/s        2800K       28x


So not only are the 3 scopes sampling at radically different rates, they're also capturing significantly different amounts of data**.  Then they have to process all that data down, to create the 600-700 points actually displayed.  And finally that data is processed through a decoder, to generate the protocol stream.  The other 'disadvantage' the Rigol is subject to is that it provides 40% more display area, with 14 horizontal divisions, vs. 10 for the other two.  Something that would certainly come in handy for displaying protocol streams.

In light of the above information, the fact that the Rigol is processing 28 times as much data as the Tek may help explain its "slowness".  And not only does the Agilent have hardware decode that the others lack, it also has another 5.6x advantage over the Rigol, due to the smaller amount of data it's processing for display.   


Lastly, I found several things from the YouTube reviewer rather odd:

1)  he emphasized that the Tek was keeping up with the decoded data, but made no mention that its display of the SCL and SDA analog traces were not collapsing and expanding, between Write and Read modes, as the Agilent and Rigol were.  Basically both traces were nearly worthless.

2)  he stated that the Rigol was not capturing all the data, which was untrue, when it was actually only not displaying all the decoded data.  Those are two quite different things.

3)  after already pointing out how slow the Rigol was, he then went on to comment on how it could be made even slower, by turning an additional decoder channel on?  I fail to see what the point was in that, since he didn't turn additional decoders on either of the other scopes, to see what impact that may have had.

4)  he finished by pointing out how unresponsive the user menus were, while the Rigol was busy processing gigasamples of data.  An implementer has to decide, when there's too much going on to keep up with all of it, whether to give priority to data acquisition, display, or interactive user feedback.  Rigol certainly could have decided to make the menu selection buttery smooth, at the expense of the other functions.  Instead, it appears they've tried to strike a balance.  Which, IMO, is the proper approach.

5)  he concluded that the Rigol was underpowered, and needed a faster processor.  That may be true, but his demo did not validate that claim.

Conclusion:  don't take at face value everything you see on YouTube.

--
*NB:  I was checking this on a small screen, and if I had scrolled down, I would have seen in the Comments section that Martin Zuber already noticed some of the same anomalies, and commented on the sampling rates and extra points the Rigol was processing.

[** One other thing to note is that while the Rigol (and the Tek) continuously capture their full sample set while in Run Mode, the Agilent doesn't necessarily do so.  This is a clever 'trick' they devised, which actually makes a fair amount of sense.  Let's say they had a 4M sample deep buffer.  Rather than capture all that on every cycle, they only capture as much data as is being displayed.  In this case 500K.  Then, when you hit Stop, the scope actually continues running, and captures the remaining 7/8 of it's buffer, so it can be examined afterwards.]

[Mark_O]

Oh, I should probably provide a pointer to video.