OMEGA - Logic Analyzer

[Carrington]

More information about how the generator works:

OMEGA has its 16 inputs divided into two ports, first port is Input 1-8 and second port is Input 9-16. Each port has individual output enable, therefore 8 inputs can be turned into outputs. Function Generator can work on one of these ports.

The parrern to generate is entered from a file in text format.
The file would look like something like this:

0us    00000001
1us    00000010
2us    00000100
3us    loop

Each line contains absolute time, from beginning (first line must contain time zero). In second column are pin states, they can be in binary format (always 8 chars), hexadecimal (always 2 chars) or octal (3 chars). Output "T" (as tristate) means that the port is input at this time.
Last line will have "end" or "loop", depending wheather the pattern should be repeated infinitely or performed only once.

The function generator can be started by button in test window or by a trigger (including GO button).


Maximum length of generated pattern is somewhat more than one second or 4096 lines (when delay between lines is not more than 2.56us).

[alm]

Here is a simple, stupid test case.
Teka an i2c bus. 5 volt 400khz.' Old skool'

Please analyser : trigger on a missing acknowledge. Any missing acknowledge.

See if it can do that.
The trigger script is easy :
If scl=1 and falling(sda) then arm
If armed : count 8 falling edges of scl then check sda. If low : trigger.

From the specs it sounds like it should. I don't have access to one of these to actually test, but they claim that you can define a condition using boolean logic and any input state or edge, and that you can string two conditions together. No argument that a 16900 has more triggering features, however.

[tridentsx]

These things are toys. You can frequently pick up a big iron analyser for a few 100$. I got a 16900 for 200$ ... When i get back i will post a teardown of a real analyser. It'l become painfully obvious then that these usb boxes are not worthy of the name analyser.

Have you looked at the Open Logic Sniffer with Demon Core FPGA? It  offers many of the features found in an HP 16500 / 16550 timing logic analyzer. Its open source and I know that the sigrok.org project is working on an improved more portable (as in portable between different vendor fpgas) version of this analyzer.

http://www.mygizmos.org/ols/Logic-Sniffer-FPGA-Spec.pdf

Maybe one day there will be an inexpensive USB 3.0 analyzer that can match the HP 16900 with the added benefit that it would fit in my pocket.

[Bored@Work]

I know that the sigrok.org project is working on an improved more portable (as in portable between different vendor fpgas) version of this analyzer.

I would not count on anything from the sigrok project. They promise a lot and deliver very little. Every other day or so they "add" a device, and claim they will support it at some time in the future.

[free_electron]

free_electron:

1) 400MHz is not the same as 400MSPS.
2) The important thing is the rise time.

The habit does not make the monk. 

Eh for a logic analyser it is. If you claim 400MHz this means i can apply a 400MHz clock and datastream and the analyser should be able to follow. This is basic state mode operation. State mode is the most important mode. In timing mode the analyser uses its own sampling clock and goes even harder so you can study things like setup and hold time violations.

[nctnico]

Carrington is right. Most high end logic analysers have a specification for external clock mode and internal clock mode. The internal clock can be very high but the max. external clock frequency is usually much lower. And the probes may have a much lower bandwidth as well which isn't a problem; even though the bandwidth is low compared to the internal sampling clock (which can be several GHz) you'd still be able to see differences in timing.

[uoficowboy]

Exactly. I am currently looking at an LA, but tend to prefer the Intronix Logicport. I have played around with their software and operation is really nice (using the stored demo signals provided). The Logicport also does CAN bus decoding, which for me is a really nice feature for future projects.

I have an Intronix Logicport and can say that the UI on it is pretty solid. I do not get warm fuzzies looking at the UI for this Omega product.

They also have a better selection of interpreters, which is a big one for me.

On a side note - I think I may be at least partially responsible for there being CAN support on the Logicport - I harassed the company about it a few times and then they asked me for some example data, and then a little while after that there was CAN support! The Logicport is really a perfect device except for the high price and somewhat limited memory.

[Electro Fan]

Exactly. I am currently looking at an LA, but tend to prefer the Intronix Logicport. I have played around with their software and operation is really nice (using the stored demo signals provided). The Logicport also does CAN bus decoding, which for me is a really nice feature for future projects.

I have an Intronix Logicport and can say that the UI on it is pretty solid. I do not get warm fuzzies looking at the UI for this Omega product.

They also have a better selection of interpreters, which is a big one for me.

On a side note - I think I may be at least partially responsible for there being CAN support on the Logicport - I harassed the company about it a few times and then they asked me for some example data, and then a little while after that there was CAN support! The Logicport is really a perfect device except for the high price and somewhat limited memory.

How often and under what circumstances do you find the memory to be limited?  When it is limited, what are the practical implications for your use?  Thx

[uoficowboy]

Exactly. I am currently looking at an LA, but tend to prefer the Intronix Logicport. I have played around with their software and operation is really nice (using the stored demo signals provided). The Logicport also does CAN bus decoding, which for me is a really nice feature for future projects.

I have an Intronix Logicport and can say that the UI on it is pretty solid. I do not get warm fuzzies looking at the UI for this Omega product.

They also have a better selection of interpreters, which is a big one for me.

On a side note - I think I may be at least partially responsible for there being CAN support on the Logicport - I harassed the company about it a few times and then they asked me for some example data, and then a little while after that there was CAN support! The Logicport is really a perfect device except for the high price and somewhat limited memory.

How often and under what circumstances do you find the memory to be limited?  When it is limited, what are the practical implications for your use?  Thx

Well, a couple times I haven't had a good way of triggering on an event (would need it to trigger on the interpreter, which is an uncommon function that it lacks, but most LAs lack it so I'm not too upset). So I do large captures and just try to find my event in the capture. Sometimes there's just too much traffic and it fills up your buffer. It's only been a limitation once or twice, but it has happened.

[biot]

I would not count on anything from the sigrok project. They promise a lot and deliver very little. Every other day or so they "add" a device, and claim they will support it at some time in the future.

This, exactly this, is why we in the sigrok project don't participate more on the EEVblog forums. Life is just too short to waste it arguing with forum warriors. Insults and clueless opinions don't add up to meaningful discussion.

A glance at sigrok.org would have led you to this blog post, less than two months old, announcing a release with finished support for dozens of devices. Since then, we've added three drivers in git, and a fourth one will go in today.

Many people here can benefit from what we do at sigrok, and many have. We've also seen many contributions from EEVblog forum members, and indeed we welcome teardowns, protocol descriptions, and documentation of all sorts -- the sort of thing that a lot of people frequenting the testgear forum are into.

Cynicism is easy, and sarcasm is the lowest form of wit. I don't know why the EEVblog moderators put up with your constant putdowns and negativity, but I find it hard to see the value in it.

[Carrington]

Hello biot!

I hope that the sigrok project still progressing more and more.
As far as I can, which is not much, I've been following the sigrok project.

"Life is just too short to waste it arguing with forum warriors. Insults and clueless opinions don't add up to meaningful discussion."

As in all places: The man and his iniquity. 

You can count on my full support.
Cheers.

[Bored@Work]

This, exactly this, is why we in the sigrok project don't participate more on the EEVblog forums. Life is just too short to waste it arguing with forum warriors. Insults and clueless opinions don't add up to meaningful discussion.

Everyone can go to your website, check the list of devices you have there and compare the list to what you actually deliver. You have whole categories, like "Thermometer", where you haven't delivered support for any listed device. Not even all devices you list in your core category are supported.

And why a logic analyzer project needs to branch out into general data acquisitions another question. Esp. before supporting all promised devices, and when there is an established data acquisition project http://www.comedi.org/.

By the way, thanks for clarifying your communication policy.

[Carrington]

Their PRESTO  programmer is excellent - very fast and all sorts of useful options for serialisation, one-button programming etc. Probably the best device programmer software I've used.
They were also very helpful when I needed support for a fairly new PIC -it was added in a few days.

Hi mikeselectricstuff!
On the outside OMEGA and PRESTO look the same, I wonder if the inside too.
Maybe you can make a mini teardonw here. Thanks.

[Carrington]

Examples of measured waveforms:
http://www.asix.net/tools/dwnld_sigma-omega_examples.htm

OMEGA - documents, software, drivers:
http://www.asix.net/tools/dbg_omega_download.htm

[Electro Fan]

Onethingi always find funny with these kind of logic analyser doohickeys is that they claim absurd capabilities.

400MHz ...  With those stupid idc connectors and flatcables ? Once you hit that kind of speed you need to start looking at a bit more exotic connectors like Mictors.

Not a peep about trigger levels.. Signals at that speed are not ttl anymore... They may even be differential.

Not a peep about triggering capabilities. From what i can tell this is just another streamer....

512 megaBIT 16 channels ... 32 megasamples per channel. Ok not bad.

User interface is poor. See the post of the other ui. Lack of list view, lack of analog display , markers, lots of other things.

A lot of these cheapo thingies are kludged together by people who have never used nor seen a real logic analyser user interface. Real analysers let you create multiple viewports into the same datapool. Each viewport can be different, it can be waveform , bus format, analog , table, bus decod ; and if the machine really has balls : inverse assembly or even sourcecode through an OMF link.

According to its specification this thingie should be capable of hooking up to something relatively simple like a pci bus running at 66MHz . In practice it will be introducing so much loadon the bus that the bus becomes unreliable. Then the dilemma is : you are trying to debug a problem with a piece of equipment that causes problems....

These things are toys. Nothing more. Not until they get more advanced trigger capabilites (like sequential trigger) and get a way better user interface. From a hardware perspective they also need a better probing interface.

Hi free_electron,

Kind of resurrecting an old thread (that was linked to by a newer LA thread)...

I can tell you are a LA Pro's Pro so your insight would be great to have here.... have you had any chance to use the Intronix LogicPort (or their software which is the real thing downloadable for demos)?  Any thoughts you have on the Intronix would be cool to hear.

Thanks, EF

[Carrington]

Free_electron is right. I should have written 400Msa/s instead of 400MHz sample speed, because it can cause confusion, sorry.
The user interface, IMHO is acceptable, I've seen worse.
A toy? Obviously is not a TLA. But it is portable, and it have protocol decoders.
I only miss a variable threshold pod or or something similar.