PC USB Logic Analyzer - Who Has The Best Software?

[Mechatrommer]

please...
1) suggest what USB LA brand has the best software?
2) what feature in LA hardware and software that is critical, important and a must have for digital debugging?
hope to hear some suggestion.

update: this is usefull info i found for LA comparison
http://sigrok.sourceforge.net/wiki/Logic_Analyzer_Comparison

[Balaur]

Hello Shafri,

 I've never had the opportunity of interacting with USB LAs, but I'm using older Tek Windows-based LAs at work.

There is a lot of very nice features in the GUI of those LAs that I'm sure you can easily find in recent USB offerings, however the outstanding feature that I really appreciate is the sophistication of the trigger system.

You have a FSM-like trigger system with states and transitions that supports if/then clauses, events (signal/buses/words transitions), event counters, etc, that could accommodate fairly complex set-ups to allow the LA to trigger on the exact event you are looking for. Very nice system!

However, for all this to work, you absolutely need hardware support that takes your triggering set-up and acts on it.

Other commonly-used software features are:
- channels naming (obviously)
- making a bus from individual signals
- describing signal states for an easy-to-understand name, color, etc in the waveform display
- easy and fast scrolling - although that one is not happening yet

Important hardware features (in a professional settings, although a hobbyst will be happy with far less):
- huge amount of channels (we are using up to 160 signals)
- low channel-to-channel slew
- per-channel user-defined thresholds for logic levels

Cheers,
Dan

[Mechatrommer]

You have a FSM-like trigger system with states and transitions that supports if/then clauses, events (signal/buses/words transitions), event counters, etc, that could accommodate fairly complex set-ups to allow the LA to trigger on the exact event you are looking for. Very nice system!
However, for all this to work, you absolutely need hardware support that takes your triggering set-up and acts on it.
Other commonly-used software features are:
- channels naming (obviously)
- making a bus from individual signals
- describing signal states for an easy-to-understand name, color, etc in the waveform display

wow, i need a beginner tutorial on this esp on triggering system.

- easy and fast scrolling - although that one is not happening yet

i saw in youtube, the salae logic software, it got ipod/handphone style smooth scrolling.
and another colorful GUI from usbee, quick read the manual, but it prove i'm clueless when it comes to triggering logic. http://www.usbee.com

Important hardware features (in a professional settings, although a hobbyst will be happy with far less):
- huge amount of channels (we are using up to 160 signals)
- low channel-to-channel slew
- per-channel user-defined thresholds for logic levels

yes i'm currently looking for "bang for buck" USB LA. but since this is dependent on PC and USB, so software is equally important part just as the hardware i think. most USB LA can provide decent hobbiest level feature, but the kicker usually the "crappy" software.

thanx Dan for suggestion.

[alm]

The only affordable USB LA that supports this kind of advanced triggering is the OLS, although I'm not sure if the software support for it is finished. See the recent thread by 0xdeadbeef. All others have quite basic triggering I think, although the Intronix LogicPort might be slightly better, and Saleae have promised more advanced triggering sometime in the future.

[Thomas]

You can download and testdrive the LogicPort software: http://www.pctestinstruments.com/downloads.htm
It will run in Demo mode if no hardware is present. Then you can play with all the features and see if it fits your requirements.

[ElektroQuark]

You can try the LeCroy LogicStudio software too. But the hardware is expensive.

[Mechatrommer]

You can try the LeCroy LogicStudio software too. But the hardware is expensive.

yes, i want to have a look on software first so i'll get idea what feature should be available in LA SW and HW.
i installed and tried the logicport, will look more detail into it.

[0xdeadbeef]

The only affordable USB LA that supports this kind of advanced triggering is the OLS, although I'm not sure if the software support for it is finished.

The status quo is still that the FPGA supports advanced triggering, but the GUI doesn't.

[Mechatrommer]

let me link some related topic...
https://www.eevblog.com/forum/index.php?topic=4940.0
https://www.eevblog.com/forum/index.php?topic=5352.msg69793#msg69793

as this thread is meant more on software, i've had my time from last night looking for usb LA. i concluded there's balance, ie [lower spec HW (MS/s, #channel) + ok'ish SW] will cost more (alot more) than [higher spec HW + crappier SW]. lets take example...

(let me rate it 1 to 10 according my own standard, being top USB HW is 500MS/s 30+ channel, large mem and fancy triggering logic (width, pattern, boolean and whatnot), and SW based on my relative judgement on GUI/feature/quick play, pls correct me if i'm wrong):

1) Hantek LA5034 (HW 8, SW 2) = $204 http://www.ebay.com.my/itm/Hantek-LA-5034-34CH-USB-Logic-Analyzer-Oscilloscope-500M-68KB-/280749224492?pt=BI_Analyzers&hash=item415df5322c
2) LogicPort          (HW 8, SW 5) = $389 http://www.pctestinstruments.com/
3) ChronoVu LA    (HW 6, SW 2) = $221 (ah crap!) http://www.ebay.com.my/itm/PC-USB-Logic-Analyzer-100MHz-64Mb-memory-/250684716233?pt=BI_Analyzers&hash=item3a5df934c9
4) USBee              (HW 3, SW 7} = USD694 HW+SW combined price (rip off!) http://usbee.com/usbeezxdigitaltestpod.aspx
5) Saleae Logic       (HW 3, SW 7) = USD149 http://www.saleae.com/logic/
6) Saleae Logic 16  (HW 6, SW 7) = USD299 http://www.saleae.com/logic16/
7) ZeroPlus          (HW 7, SW 4) = € 129,00 http://www.zeroplus.com.tw/logic-analyzer_en/products.php?pdn=3&pdnex=list
8} DigiView 3100   (HW6, SW 5) = USD499 http://www.tech-tools.com/dv_dv3100.htm
9) Scanalogic (HW6, SW 8} € 59 http://www.ikalogic.com/scanalogic2/prices.php
10) Lecroy LogicStudio 16 (HW9, SW9) $990 http://www.lecroy.com/logicstudio/logicstudio.aspx

i wish i have hantek (fastest and cheapest) + usbee sw but sadly its not open SDK, so i cant build one sw for my need, its stated in the hantek link above.. there's 2nd design with supply demo VC VB etc, but i cannot find in the net, does anyone know? but since the sw is not upgradable, maybe i should stay away from hantek? salae logic 16 maybe the next option, with less MS/s capable, but $99 more.

edit: with exception to ols la, which only $49 with better SW, but deadbeef complained about having to do something to pc to make it work. sad, i'm not expert in java or sump thing.
fyi: LeCroy sw crashed, shame!

[Chet T16]

There is a salae sdk and clone hardware can be had for $50. My clone works as both a salae and a usbee.

[amspire]

I have the Logiport software with the Intronix LA1034 hardware.

It is a fairly old design now - my board is copyright 2006.

If you are interested I can do some photos of the internals, but basically the hardware is an Altera Cyclone EP1C4F324C6, FTDI FT245BM USB chip, a 93C46 serial eeprom, and a switching supply.

Because of its age the memory is now a bit small (34 bits x 2048) but with compression (for clock speeds up to 200MHz), it can often store an adequate amount of data. I think at 500MHz internal clock mode, it does not compress.

If I was looking for a new analyser now, I would be looking for more memory.

It only has 4 interpreters - IC2, Synchronous Serial/SPI, RS232, CAN 2.0A-2.0B Bus.

No analog channels.

Input logic threshold programmable between -6V to 6V with 50mV resolution, 100mv accuracy so in theory you can do any logic from 15V CMOS down to sub 1V logic. Maximum input voltage is +/- 40VDC which is pretty good. In practice though, sensitivity degrades with speed, so at 250MHz, the sensitivity is something like 0.8 V, so it will be struggling with fast logic around the 1V mark.
 
Anyway, you can download the software and run it in demo mode. Decently written, but hasn't really changed much over the years.

Richard

[Mechatrommer]

There is a salae sdk and clone hardware can be had for $50. My clone works as both a salae and a usbee.

link or PM please i hope its logic16 since original logic one day will be limiting (24MS/s no internal memory) i guess. how do you download usbee sw? (USBee Suite Pro?) its $299 in the website. a total rip off.

I have the Logiport software with the Intronix LA1034 hardware....

you are free to tear down in another dedicated thread (or in this thread i dont mind), we will enjoy watching . i've tried logicport sw yes it looks a bit old but way better than hantek sw, i think its adequate for some work. i wonder since hantek hw spec = logicport hw spec. maybe hantek cloning it, and maybe the intronix sw will work on hantek? is there SDK available for logicport?

have tried saleae logic sw, its kind of fun. the other sw we need to click a button or press keyboard to zoom in and out, but this saleae is using middle wheel to quickly zoom, ipod style left right scroll, i think this is important for quick navigation GUI and debugging large data.

[alm]

There is a salae sdk and clone hardware can be had for $50. My clone works as both a salae and a usbee.

I dislike those clones since about 90% of the work for those USB bulk transfer logic analyzers is in the software. The hardware is just a Cypress USB SoC in bulk mode which sends all data to the host. It's not like a complex multimeter with some add-on software for data logging.

Wasn't there some issue with new versions of the Saleae software being incompatible with clones?

[DrGeoff]

Have been using the Logicport analyser for a few years now and it has never failed me. Software is very stable and the hardware works just fine. I've really only been using it to double-check timing and states during testing and to decode some protocol streams, but it has been a very handy device to have.

[JuKu]

I have a Zeroplus. Now that I've used it a bit more, I have more respect for the protocol analysis features of it. It is really strong on that, and I didn't expect that seeing what is actually going on in a fairly complex system would help so much in the software development section. For example, i had no trouble following a command from host processor serial terminal to the slave processor SPI port to the peripheral registers and back (several SPI busses, some displayed twice in different formats). I even had a few spare lines left after the exercise:-).

If I would buy now, I wouldn't make a compromise on that feature, I'd go for more channels and as deep memory as I can afford.

[Chet T16]

link or PM please i hope its logic16 since original logic one day will be limiting (24MS/s no internal memory) i guess. how do you download usbee sw? (USBee Suite Pro?) its $299 in the website. a total rip off.

Its only the original salae but for $50 i couldn't say no

http://iteadstudio.com/produce/seleae-logic-usbee-ax-usb-blaster/

[Balaur]

Thank you Shafri for preparing the list. Much appreciated. It motivated me to (slowly) have a look at all the tools for a possible acquisition.

I've just played with the Saleae logic software. I'm somewhat dismayed by their approach, but I'm used to a totally different LA GUI, so sorry for my assumptions.

Quick questions:
- Can you combine arbitrary signals into a bus and display the bus waveform? I know that there is a simple parallel analyzer, but it's not what I expect
- Can you have vertical reticles?
- Are you able to set the trigger instant inside the sampled window between 0% and 100%?
- Without bus waveforms, the trigger set-up makes sense but it's cumbersome to use if you want a particular value on a bus. However, for 8/16 channels, it's not a big issue.

Cheers,
Dan

[0xdeadbeef]

Regarding the number of protocols, the ZeroPlus is unbeatable, especially if you get 30+ plugins for free. There's even an SDK to develop your own plugins. Also the GUI is not that bad after all. It's not always 100% intuitive, but pretty capable. The newest beta version even supports LA and DSO stacking. So you are able to combine the input of multiple LAs and a scope. Unfortunately, it crashes when I connect it to my Owon SDS, but this looks like an interesting feature. Generally speaking, the development of the GUI is pretty active compared to other LAs.

What I don't like is that you need to define two buses (and duplicate inputs) to decode SDI and SDO of an SPI bus. Also the trigger system is very limited. And it's kinda crappy that you need to install each and every protocol separately and the downloads are rather big just because of large Chinese help files.
On the good side, it's very cheap and has a good memory depth, especially I you buy the C16032 and patch it into a C16128. Also the thresholds are adjustable from -6V to +6V and the input can tolerate -30V/30V.
So if you your main interest is to decode protocols and you want support for stuff beyond SPI, there is not much of an alternative.

The OLS is cheaper and has the promise of an advanced triggering system, but it has no housing, no adjustable input threshold and can tolerate only -0.5V to +7V. The GUI is ok, but you can't do very much with it. E.g. you can't define buses, the trigger setup is very cryptic.
Of course, everything including the hardware and FPGA code is open source, so you could change it all on your own, but honestly, there are only a few people who worked on the project, currently the development is more or less on hold since several months. E.g. there are only a few protocol analyzers and only one of them is developed by another person than the one who developed the GUI. The fact that the FPGA code supports an advanced triggering system since early 2011, but there is still not even a beta version of the GUI to actually use this feature says a lot about the speed of development.

[amspire]

I added a teardown for the Logicport logic analyzer here:

https://www.eevblog.com/forum/index.php?topic=5387.msg70141#msg70141

[alm]

- Can you combine arbitrary signals into a bus and display the bus waveform? I know that there is a simple parallel analyzer, but it's not what I expect

Not in the current version.

- Can you have vertical reticles?

Not as far as I know.

- Are you able to set the trigger instant inside the sampled window between 0% and 100%?

Pre-trigger buffer size in preferences gives you some control.

- Without bus waveforms, the trigger set-up makes sense but it's cumbersome to use if you want a particular value on a bus. However, for 8/16 channels, it's not a big issue.

Yes, triggering is currently quite limited: the v1.0 software had more options. They have promised better triggering, but no time line.

[JuKu]

What I don't like is that you need to define two buses (and duplicate inputs) to decode SDI and SDO of an SPI bus. Also the trigger system is very limited.

It is a bit more work to set up, but it gets the job done. With "virtual CS" (based on the activity of clock) you only need to copy one channel (the clock), so even the duplicated screen estate is not a big waste. You can copy channels, so you can get to use all channels anyway. I don't see this as a big minus.

The limited trigger is much more of a concern. But are we asking too much from a $150 instrument? Still, I'd buy another or two for stacking if it could trigger on something like this:
-if a SPI bus master out is 0x34 xx xx (talking to a particular device) and there is no falling edge on int_x in next 1 ms... (device missed timing)
-when a device interrupt goes low but the host does not activate the chip select in x_ms (master missed timing)
-when a peripheral SPI out bus has a pattern "0x82 0x15 0x0?" and "?" has bits 1 or 3 set (error on peripheral)

You get the idea.

[Balaur]

Thank you Alm, much appreciated.

Cheers

[0xdeadbeef]

It is a bit more work to set up, but it gets the job done.

I didn't say that it doesn't work, I just said that I don't like the way it works. SPI is by definitions a 2way protocol, so decoding only one data line violates the protocol. Yes, there is a workaround, but it would be less effort and less waste of screen space if the SPI analyzer would simply decode data in and out.

The limited trigger is much more of a concern. But are we asking too much from a $150 instrument? Still, I'd buy another or two for stacking if it could trigger on something like this:
-if a SPI bus master out is 0x34 xx xx (talking to a particular device) and there is no falling edge on int_x in next 1 ms... (device missed timing)
-when a device interrupt goes low but the host does not activate the chip select in x_ms (master missed timing)
-when a peripheral SPI out bus has a pattern "0x82 0x15 0x0?" and "?" has bits 1 or 3 set (error on peripheral)

IMHO there is no LA available which can do this in realtime. Only the dumb USB LAs could do this in theory, but at this point, there doesn't seem to be a single one with better trigger possibilities than the ZeroPlus. The FPGA based LAs would have to implement the protocol analysis in the FPGA which is not possible at least for 30+ protocols. The only way to do this would be an offline analysis, i.e. capturing a buffer, analyzing in the PC software and pretending a trigger condition if there was one. This is how scopes with SPI pattern trigger ability do this and this is also I guess how high end USB LAs like the LeCroy LogicStudio handle it.
Indeed ZeroPlus offers plugin called Protocol Analyzer Trigger for 39€, which however seems to only mark the fitting sample. If at all, it can only work as offline trigger.
There is also a so called Pulse Width Trigger Module which makes it possible to trigger on pulse width of a single signal. But it works only on one signal, obviously can't be combined with other trigger conditions and is a little clumsy to use. IMHO it makes more sense to use e.g. the OLS as trigger generator if it finally supports advanced triggering.

[alm]

IMHO there is no LA available which can do this in realtime.

Make this no affordable LA. I'm sure Agilent can sell you a box that can do this. Would the OLS come close if they ever finish the software side of the demon core?

Only the dumb USB LAs could do this in theory, but at this point, there doesn't seem to be a single one with better trigger possibilities than the ZeroPlus.

I agree, this is a missed opportunity. The fast CPU of your modern desktop and large amount of buffer memory should allow them to do much more than a cheap FPGA, but none of them take advantage of this. I have complained about this in the past, and was told to expect much more advanced triggering in the future, but I wouldn't hold my breath.

This is how scopes with SPI pattern trigger ability do this and this is also I guess how high end USB LAs like the LeCroy LogicStudio handle it.

Or they just use a bigger FPGA. You don't have to actually load all 30 protocols all the time, you could have a separate bitstream per protocol. You could even get fancy and synthesize a bitstream dedicated to the trigger condition.

Indeed ZeroPlus offers plugin called Protocol Analyzer Trigger for 39€, which however seems to only mark the fitting sample. If at all, it can only work as offline trigger.
There is also a so called Pulse Width Trigger Module which makes it possible to trigger on pulse width of a single signal. But it works only on one signal, obviously can't be combined with other trigger conditions and is a little clumsy to use. IMHO it makes more sense to use e.g. the OLS as trigger generator if it finally supports advanced triggering.

I agree with your assessment that these are just clumsy workarounds.

[0xdeadbeef]

Make this no affordable LA. I'm sure Agilent can sell you a box that can do this.

Well, the 10k€ Agilent desktop LA on my bench at work can't do this, neither can the 14k€ LeCroy Waverunner Scope with the ~1k€ SPI plugin. Well, the latter can trigger on SPI patterns, but of course only one the sample memory. So it actually already triggered and just filters the memory as described above.
So I'm not really convinced that there is any device on the market that can combine protocol and timing trigger in realtime. Of course you could let someone design you a specialized LA or do it yourself, but I doubt that there is an existing solution yet.

Actually, I think it should be pretty easy to add e.g. an SPI decoder to an FPGA LA design, but then you'd need a larger FPGA and to change the protocol, you'd need to download a new FPGA bitstream. Or you'd need a VERY large FPGA to fit in the 30+ protocols of a ZeroPlus LA.
Then again, I guess if you put a somewhat capable microcontroller next to the FPGA and decode the protocols in the microcontroller, it should be possible to combine close to realtime protocol triggering with edge/time triggering. Then again, I'm not aware of any device on the market that could do this out of the box.

Would the OLS come close if they ever finish the software side of the demon core?

No, it would just come close to the edge/time trigger capabilities of an old HP desktop LA (16500/16550):
http://gadgetfactory.net/logicsniffer/uploads/Playground/OLS%20Advanced%20Trigger.pdf
http://www.mygizmos.org/ols/fpga.html

I agree, this is a missed opportunity. The fast CPU of your modern desktop and large amount of buffer memory should allow them to do much more than a cheap FPGA, but none of them take advantage of this. I have complained about this in the past, and was told to expect much more advanced triggering in the future, but I wouldn't hold my breath.

Well, the (non realtime) operating system is always a bit of a problem. Sometimes windows decides to use 100% CPU load for whatever for dozens of seconds and it will be pretty hard to overcome this with a large buffer. Surely the reliability of USB analyzers could be improved a lot by using USB3.0 and larger buffers, but they will never reach the reliability of a real HW trigger.

Or they just use a bigger FPGA. You don't have to actually load all 30 protocols all the time, you could have a separate bitstream per protocol. You could even get fancy and synthesize a bitstream dedicated to the trigger condition.

Sure. Would be completely possible. It's just that you can't place your order on any available product at this point.
At least AFAIK.