Skip to main content ↑ ↓

Messages

This section allows you to view all Messages made by this member. Note that you can only see Messages made in areas you currently have access to.

Messages - alexwhittemore

16
Open Bench Logic Sniffer / Re: FIFO wing to increase sample depth?
It would probably be less complex to just integrate everything on one board, with the main FPGA itself a 500k gate controlling some other chips, but I'm thinking that the way you do it is to fill internal memory first, then stream the 16bit input samples straight out to the wing, and basically make the wing a fifo composed of some SRAM chips and an fpga to handle them. The real problem is that then there's not enough bus width to control whether the fifo is reading or writing, and there's no clock line, etc.
17
Open Bench Logic Sniffer / FIFO wing to increase sample depth?
One of the biggest problems with the OLS right now is a shallow sample depth - if your signal changes a lot then RLE doesn't do much to help, and as it is the OLS can only store a few hundred bytes of UART, hardly all that much! What I'd love to see is either a bigger FPGA (the pin-compatible 500k gate part, for example) with more bram for sample storage, but that drives up cost for those who don't need the extra space. My thought for an alternative is a memory wing, either with a dedicated FIFO chip or an FPGA and a bunch of fast external SRAM or something along those lines. You'd only get the 16 buffered channels instead of all 32 on the device, but the tradeoff would be VASTLY increased sample depth.

Has anyone thought about this or implemented any ideas? I'd love to hear them!
18
Client software / Re: Android client
As far as programming a standalone unit, I don't think we'd be trying to port much code. The 'client' in such a case would also need to handle all the hardware functionality, so it'd be totally different than running an application on top of an OS, unless there were a full embedded linux board or something. Which is one reason I think the smartphone port makes more sense - no overhead of handling an interface.
19
Client software / Re: Android client
[quote author="sqkybeaver"]imho there is no reason that the should not be somthing like the dso nano but have the logic sniffer built in.[/quote]

I don't disagree, but I think that there is one key difference. Typically, if you're looking at analog waveforms, you're doing something like debugging a component or subcircuit. Using the waveform involves tweaking analog pieces of the circuit.

On the other hand, when you're using a logic analyzer, what you're debugging is generated by digital components or subcircuits, which are almost always, these days, software generated. It's either FPGA HDL you need to modify, or microcontroller code in order to fix your serial or parallel waveforms. If we still used 7400 chips all the time, that wouldn't be so true, but discrete logic isn't typically a major part of today's workflows.

I think what that means as far as tools go is that it's useful to have a standalone scope - you're not limited by bandwidth of a USB bus, you don't have to have a computer present when all you're using it for is circuit debugging, and so on. Intrinsically, it makes sense to have a scope next to your circuit.

But on the other hand, if you're using a logic analyzer, you're almost certainly modifying software and reloading that into your circuit to try to get your waveforms correct. So you've already got a computer on your bench, and you're trying to match bytes and timing from the LA to what your software SHOULD be generating. In that case, it's far more useful to have the waveform data in the window next to your IDE. For that reason, I don't think that I'd find a bench or even standalone handheld LA especially useful. If it's an extra $30 for the bluetooth hardware to use my phone as a display, that's worth it for the novelty and portability in certain situations, and for that matter, bluetooth could replace the USB cable to a computer in the first place. But I don't know that I'd pay $100+ for a standalone unit INSTEAD of my USB unit, whereas I totally would for a scope.

Additionally, there are key interface differences too: Scopes are used on analog waveforms, and LAs on digital. Changing the parameters on a logic analyzer involve punching in integer values and clicking checkboxes corresponding to boolean values, all easy and comfortable to do on a computer in a software interface.

On the other hand, getting your waveform correct on a scope involves turning knobs for amplitude, time delay, trigger value, and so on. This is far more awkward on a computer, especially for the continuous nobs like cursor that can't be replaced by a 0-100% slider. I will always prefer nice smooth knobs to interact with my analog waveforms, necessitating a nice big standalone product. I think even the DSO Nano/Quad come up short here.


EDIT: Forgot to say, you mentioned integrating something like the OLS INTO something like the nano. I would definitely be in favor of a standalone mixed signal scope. That would be one MAJOR case in which a standalone product makes lots of sense.
20
Client software / Re: Android client
Hey all: pppd/ian: I didn't mean to suggest closed development, not at all! I just think we've already done a nice job hijacking an old and only somewhat related thread to get here :).[s:]I've created a new thread to talk about the mobile client/bluetooth hardware idea here:

viewtopic.php?f=23&t=2916

pppd, I'd love your response to my question there, what is your actual bluetooth hardware setup, does it work, and can I copy it from you on my end so I can start playing with making a BT-based iOS client.[/s:]

As for the idea introduced by sqkybeaver of using a graphic LCD and embedded linux board, I've just started a thread about making a benchtop variant here:

viewtopic.php?f=23&t=2915

Spoiler: I think it'd be cool, and I'd totally be willing to plunk down a few hundred dollars for a nice, community-designed bench logic analyzer, even if a little buggy, but I don't think a true community design has the attention span to make it happen without serous bugs and developmental lags such that it's no longer worth the hundreds of dollars hardware cost.

EDIT: Didn't realize the thread had already been split from the beta testers thread, sorry guys! Benchtop vairant still stands though, I think that deserves its own thread.
21
Open Bench Logic Sniffer / OLS Bluetooth android/iphone variant
This is a continuation of a discussion in the client software beta testers thread here:
viewtopic.php?f=57&t=2913&p=28666#p28666

The idea was 'wouldn't it be cool to have a mobile iphone/android interface to make the OLS standalone from a computer, to some extent?.' This involves some hardware modification and/or addition, thus I think the pursuit of such a change belongs outside the software section, at least for now.

My main question for now is to pppd, who has played with a bluetooth mobile link on the OLS before. What was your hardware setup like? Can you post details so I can try to make it on my end to start developing some client software for my iPhone? I can take a stab at what you did, but I'd love to just copy you if you've actually got working hardware.


EDIT: Nevermind! The forum thread was already split off, continue discussion here:

viewtopic.php?f=57&t=2913&p=28674#p28674

LOCK!
22
Open Bench Logic Sniffer / OLS benchtop variant
This is a continuation of a discussion started here in the client software beta testers thread:

viewtopic.php?f=57&t=2913&p=28666#p28666

The idea was 'should there be a mobile client for iphone and android so the unit is more standalone.' That became 'well, it'd be easier to port the software if we just used a linux board with a graphic lcd'. My feeling is that the later basically constitutes making a bench top unit, which I have a couple feelings on and would like to discuss here.

My thought is that, personally, I would like a nicely-made kit-style open source bench logic analyzer - that'd be pretty darn cool. But the real point of the USB LA is that much of the cost and *complexity of development* is shifted off the LA device itself and into a device most everyone already has near their bench, a full computer. Because of this, the OLS is only a small, $50 assembled board. My thought is that it's already difficult for the community to coordinate development of one good interface and one good firmware/FPGA design, I'd be extremely surprised if we could successfully maintain a full bench instrument.

The other problem is that I think it could pretty easily be kept under $500, but even $500 is a TON for most hobbyists, especially when they can get a professional USB based LA with few to no bugs, or a used HP bench unit for a similar price. The concept of paying that much for a unit as buggy as OLA currently is probably won't be palatable to most.

What do you guys think on the matter?
23
Client software / Re: Client beta testers
I'm interested in trying my hand as well, but I'm in the same boat. Can we take this to a different thread or offline? My email is my username at gmail.
24
Client software / Re: Client beta testers
Eh, at that point you're just making an embedded computer. You might as well turn it into a bench logic analyzer, and then cost is already way above the current $50. Not that it wouldn't be super cool! But I like the bluetooth idea more. Bonus points for electrical isolation.

I'd love to help out, I imagine the biggest missing piece is making a mobile client? I'm not too great at iPhone development, but I know my way around obj-c and could probably do it with hardware available. Do you have a functional Bluetooth wing now?
25
Client software / Re: Client beta testers
I'd be a fan of that. I can't imagine it'd be terribly hard since iPhones have raw UART available. The biggest problem is that they don't run java.
26
Open Bench Logic Sniffer / Re: Xilinx ISE on Ubuntu
It turns out that the problem all along was VirtualBox. The USB passthrough just wasn't behaving correctly, which I found out when I had the same problem with windows. Switching to vmware fusion fixed it.
27
Open Bench Logic Sniffer / What to do with unused channels
In general, yes, ground them. The existence of a pull down resistor on the FPGA input internally Has been discussed, but the input buffer makes it useless. My solution was to take a row of male headers (7) and a male+female header to plug into whatever ground point is available, with all unused test leads tied to the header.
28
Open Bench Logic Sniffer / Re: Xilinx ISE on Ubuntu
That's all interesting to hear! I've never done anything extraordinarily complex so webpack's built in synthesis has always been more than enough, but it's interesting to hear about the options.

And don't get me started about Cadence pricing. My favorite: "we're going to sell you 15 support options, mix and match as you like." "But I only want 5 of design support!" "Well, best throw some others in there too to get your money's worth!"
29
Open Bench Logic Sniffer / Re: Xilinx ISE on Ubuntu
What tools are you using at work? Most pro EDA stuff tends to be RHEL just by nature that it's always been that or UNIX. We run a RHEL (or at least CentOS) server farm since that's what Cadence tools like. It seems to me like most tools with licenses north of a few thousand dollars a seat are primarily linux. But then I've also noticed that FPGA tools are somewhat of a different story, for example Cadence is only kind of sort of just starting to try to break into the tools market.
30
Open Bench Logic Sniffer / Xilinx ISE on Ubuntu
This might be a bit of a longshot, it seems like everyone doing core work is on windows, but has anyone had any luck getting Xilinx ISE Webpack 13.1 to properly install on Linux? Specifically, I'm trying to get it running in a Ubuntu 10.10 VM. Everything went more or less smoothly up until trying to install cable drivers for iMPACT, which basically just fails all over.

I like Ubuntu, but I'm not attached to it per se. Part of the problem is that I can't find anywhere some officially supported linux setup.

( ! ) Fatal error: Uncaught exception 'Elk_Exception' with message 'Please try again. If you come back to this error screen, report the error to an administrator.' in /var/www/dangerousprototypes/forum/sources/database/Db-mysql.class.php on line 696
( ! ) Elk_Exception: Please try again. If you come back to this error screen, report the error to an administrator. in /var/www/dangerousprototypes/forum/sources/database/Db-mysql.class.php on line 696
Call Stack
#TimeMemoryFunctionLocation
10.01522449960session_write_close ( )...(null):0
20.01552581576ElkArte\sources\subs\SessionHandler\DatabaseHandler->write( )...(null):0
30.01552582352Database_MySQL->query( ).../DatabaseHandler.php:119
40.06042721112Database_MySQL->error( ).../Db-mysql.class.php:273