Best entry level oscilloscope for computer engineering?

[nbritton]

I find the hacked Rigol 1054Z very compelling. However, decoding SPI on this scope is limited to 25 MHz. Is there a better bang for your buck scope if your primary interest is computer engineering? Would it be better to get the 1054Z + logic analyzer + function generator, or put the money towards a MSO that can do all of those things? Besides the basics discussed in EEVblog #168, what other test equipment is fundamental for a computer engineering lab? This is just a hobby for me and also I'm living in a small apartment right now. I'm a linux engineer by day, if I can frame this as professional development I think I can budget what is necessary to get the right equipment. One of the things I'd love to do is build my own 1970s era computer from scratch so that I can further my understanding of computers. I'd also like to be able to interface with things like Raspberry Pis and other Linux based SBCs for embedded system design.

[Armxnian]

The decoding functions on the DS1054z are pretty meh, as well as the fft, but the scope is a great value all around. MSOs are nice but expensive. Your first plan consisting of the three separate pieces of test equipment will yield better value. I think Saleae products are still great, but you can find cheaper analyzers.

I think it's fine to buy necessary components like multimeters and scopes, and in your case logic analyzers initially, but I don't see a point in buying everything at once. You can end up with a bunch of stuff that sits around, and not have enough money to buy things you realize you need or want after you get started. It would be wiser to buy things as you require them.

[D3f1ant]

I give  to one of the Saleae logic pro models. I think a logic analyser will probably be more useful to you than a 2 ch analog scope. With logic pro you get some basic analog features too, but very decent signal decoding, and many channels, which is very handy debugging code and digital systems.

[Fungus]

Get a dedicated logic analyzer if that's your thing.

The DS1054Z is only good for very basic logic analysis. The MSO versions aren't much better. You really need a mouse, keyboard and gigabytes of RAM for serious logic analysis (eg. Saleae).

You also need a 'scope to look at signal integrity on your data lines. Saleae and DS1054Z.

PS: Did 1970s computers have SPI?

[nbritton]

Would you guys pick a hacked DS2072A for $839 or a hacked DS1054Z for $399?

The DS2072A is 2-Channel, 300 MHz, 2 Gs/s, 56 Mpts, 500uV, 50 Ohm Input
The DS1054Z is 4-Channel, 100 MHz, 1 Gs/s, 24 Mpts

[dadler]

Would you guys pick a hacked DS2072A for $839 or a hacked DS1054Z for $399?

The DS2072A is 2-Channel, 300 MHz, 2 Gs/s, 56 Mpts, 500uV, 50 Ohm Input
The DS1054Z is 4-Channel, 100 MHz, 1 Gs/s, 24 Mpts

Subtract 6% with the EEVBlog discount @ Tequipment.

I have both of those scopes. I prefer the DS2000 as it has a larger screen, higher sample rate, higher bandwidth, 50ohm input impedance, a really nice jog dial for segmented memory, and more. Personally, I'd go with the DS2000 + a newer model Saleae (one of the FPGA-driven versions).

Edit: The DS2000 also lets you hide the side menus, which gives you more horizontal divisions. This is one of my favorite options on the scope.

[Fungus]

Would you guys pick a hacked DS2072A for $839 or a hacked DS1054Z for $399?

I have both of those scopes. I prefer the DS2000

I should hope so! It costs twice as much, something would be seriously wrong if you preferred the cheap one.

(Although sometimes you *do* need four channels, eg. for decoding SPI    In that case the cheap one is preferable.)

[tggzzz]

However, decoding SPI on this scope is limited to 25 MHz.

There's a strong argument you should not decode SPI on an oscilloscope:

• adding the scope may change the analogue signal
• the scope may interpret (i.e. digitise) the analogue signal in a different way to the "real" reciever
• use analogue tools for analogue signals, and digital tools for digital signals

Thus:

• if you are looking at signal integrity, then use an analogue scope
• if you are looking at bits and bytes, use a logic analyser or log things using software

But "never say never"

[Howardlong]

Get a dedicated logic analyzer if that's your thing.

The DS1054Z is only good for very basic logic analysis. The MSO versions aren't much better. You really need a mouse, keyboard and gigabytes of RAM for serious logic analysis (eg. Saleae).

You also need a 'scope to look at signal integrity on your data lines. Saleae and DS1054Z.

PS: Did 1970s computers have SPI?

It's worth also considering the sample rate of the LA: many USB analysers are quite low sample rates (100MHz is common for example). This may or may not be a problem depending on what you're doing.

Maybe I'm weird, but I rarely use a USB LA, almost always I use an MSO. I used to use my Logicport quite extensively, but once I had an scope with serial decode and triggering, it maybe comes out once a year, and that's for the rare occasions I need to look at parallel busses. The reason might be because the scope's always out on the desk and ready to go with the pods and probes attached, whereas the LA is typically stored in a drawer under the desk.

In the '70s, everything was parallel busses for intra- and inter-board connections. I think the first time I saw I2C was in the early/mid '80s. Back then, 100kbps seemed awesome! I don't remember SPI catching on generally for some time after that, but others may have differing recollections.

[ebastler]

There's a strong argument you should not decode SPI on an oscilloscope:

• adding the scope may change the analogue signal
• the scope may interpret (i.e. digitise) the analogue signal in a different way to the "real" reciever
• use analogue tools for analogue signals, and digital tools for digital signals

Sorry, but I am not buying into those arguments. Number 1 and 2 apply to any digital logic analyser as well, and number 3 is not a technical argument at all. I do agree that a PC-based logic analyser is preferrable for digital signal analysis and decoding, but for different reasons: Larger screen, larger memory.

[tggzzz]

There's a strong argument you should not decode SPI on an oscilloscope:

• adding the scope may change the analogue signal
• the scope may interpret (i.e. digitise) the analogue signal in a different way to the "real" reciever
• use analogue tools for analogue signals, and digital tools for digital signals

Sorry, but I am not buying into those arguments. Number 1 and 2 apply to any digital logic analyser as well, and number 3 is not a technical argument at all. I do agree that a PC-based logic analyser is preferrable for digital signal analysis and decoding, but for different reasons: Larger screen, larger memory.

To repeat my statement that you snipped: 'But "never say never" '

In practice, if you follow 3 then a beginner will make far fewer mistakes of the type  that can and do arise from 1 and 2.

 

[nbritton]

The MSO2072A costs $376 more than the DS2072A, is the build-in logic analyzer worth $376? How does it compare to USB based LAs, such as the Saleae? The 16 channel Saleae with 100 MHz bandwidth costs $599. The cheapest 8 channel Saleae with 25 MHz bandwidth is $219.

[Howardlong]

Perhaps another way of looking at this is that given the choice of an LA and a scope, I'd always take the scope.

Given the choice between a 2Gsa/s 300MHz 2ch scope and a 1Gsa/s 100MHz 4ch scope is a little harder, but overall I think I'd go for the 100MHz 4 ch scope.

But given the choice between a 300MHz 2+16ch MSO and a 100MHz 4ch scope I'd take the 300MHz 2+16ch MSO.

My line of work is embedded mixed signal RF, with about 65% MCU (up to 200MHz clock) and the rest analogue, so I am sure others may have different opinions.

[Fungus]

The MSO2072A costs $376 more than the DS2072A, is the build-in logic analyzer worth $376? How does it compare to USB based LAs, such as the Saleae? The 16 channel Saleae with 100 MHz bandwidth costs $599. The cheapest 8 channel Saleae with 25 MHz bandwidth is $219.

USB is much better because it has bigger screen (your PC!), more memory, better controls for zooming and panning, etc. Twiddling knobs on a 'scope just doesn't work as well as a mouse/keyboard. I'd say get the DSO + a Saleae (or equivalent)


Edit: Unless you don't have a PC on your workbench, obviously...

[marshallh]

I have a Logic Pro 16, great hardware, almost-great software hobbled by crashing every acquisition. The only software that supports the new models is BETA.
Also, though the pro has ADCs on every pin, the sample rate is dog-slow and is barely even sufficient to look at i2c.

[nctnico]

I don't like the USB logic analysers because either the memory is short or the sample rate low. A modern MSO (preferably with protocol with decoding) is much more useful.

[Melt-O-Tronic]

I have a Logic Pro 16, great hardware, almost-great software hobbled by crashing every acquisition. The only software that supports the new models is BETA.
Also, though the pro has ADCs on every pin, the sample rate is dog-slow and is barely even sufficient to look at i2c.

I'm intrigued by this post because I'm struggling to stick a crowbar in my wallet and spring for the Pro 16.  I thought the software was supposed to be far better than the competition.  Maybe not the new (beta) stuff then, huh?  How do you feel about their ability to sort it out?

[jancumps]

I don't like the USB logic analysers because either the memory is short or the sample rate low. A modern MSO (preferably with protocol with decoding) is much more useful.

I find this odd myself , but I prefer a scope to be standalone, and a logic analyzer to be USB type.

Maybe because the different type of manipulations I do with each of them?

[MarkL]

I have a Logic Pro 16, great hardware, almost-great software hobbled by crashing every acquisition. The only software that supports the new models is BETA.
Also, though the pro has ADCs on every pin, the sample rate is dog-slow and is barely even sufficient to look at i2c.

I'm intrigued by this post because I'm struggling to stick a crowbar in my wallet and spring for the Pro 16.  I thought the software was supposed to be far better than the competition.  Maybe not the new (beta) stuff then, huh?  How do you feel about their ability to sort it out?

They're coming up on one year with software still in beta.  My Pro-16 shipped Oct 8 last year and I'm still waiting for released software.

I agree with marshallh that the ADC per channel is a big disappointment due to its slow acquisition rate.  It's no substitute for even the worst scope.  I'm sure it added quite a bit to the cost.

I also have an old Logic 8 which I tend to use more.  It's smaller and less bulky, and it runs with the older released software (1.1.15) which is rock solid.  At the moment, I'd recommend getting one of those, or the old 16-input model if you can find one.  Both have been discontinued.

In my opinion, Saleae has spent way too much time making their UI slick and pretty, and they lost sight of making it functional first.

On the plus side, they continue to be active and responsive to bug reports, so I think they'll get there eventually, but I would wait on a Pro purchase.  I regret jumping in so early.

[nctnico]

I don't like the USB logic analysers because either the memory is short or the sample rate low. A modern MSO (preferably with protocol with decoding) is much more useful.

I find this odd myself , but I prefer a scope to be standalone, and a logic analyzer to be USB type.

Maybe because the different type of manipulations I do with each of them?

The reasoning behind it is that an MSO usually has deeper memory and faster scrolling etc.

[Melt-O-Tronic]

I don't like the USB logic analysers because either the memory is short or the sample rate low. A modern MSO (preferably with protocol with decoding) is much more useful.

I find this odd myself , but I prefer a scope to be standalone, and a logic analyzer to be USB type.

Maybe because the different type of manipulations I do with each of them?

The reasoning behind it is that an MSO usually has deeper memory and faster scrolling etc.

Deeper memory than a PC?   

<snip>

<snip>

<snip>

Thank you both immensely for your thoughts.  I'll direct my resources toward other priorities for now and revisit later.

[nctnico]

I don't like the USB logic analysers because either the memory is short or the sample rate low. A modern MSO (preferably with protocol with decoding) is much more useful.

I find this odd myself , but I prefer a scope to be standalone, and a logic analyzer to be USB type.

Maybe because the different type of manipulations I do with each of them?

The reasoning behind it is that an MSO usually has deeper memory and faster scrolling etc.

Deeper memory than a PC?   

Definitely at higher samplerates.

[Armxnian]

Dave made a video about this:

Also, I personally don't think the hardware in an MSO makes up for the fact that you have to use the LA through the scope interface.

The best scenario is where both the hardware and software are the best, such as in a Windows based scope, but those are out of most hobbyist budgets.

[alank2]

I guess I am misunderstanding something perhaps, I only have the first model Logic 16, not the pro, but their website says 50MS/s for analog and 500MS/s for digital.  Is there some problem with the 500MS/s since everyone keeps saying the ADC is slow?  I've used the 1.2.3 beta software and it is much improved over previous versions at least for me.

[Howardlong]

I find the hacked Rigol 1054Z very compelling. However, decoding SPI on this scope is limited to 25 MHz.

I don't know about the analogue channels, but the MSO1000Z LA channels will decode 50MHz SPI, down to a 250MHz sample rate.

Edit: I just tried it on the analogue channels, and while the signals ain't pretty, it does decode fine on a 50MHz SPI stream.