Use this space to cheer the creator along, and talk to your fellow backers.
Have a question?
@smokeymcpot We are managing this project in our particular way and in the way we stated when we asked for backing. There is a lot of work to do and unfortunately we can't manage huge amounts of feature requests from everyone (which usually happens when there is a large forum for a project like this). If you look through the comments we have added features based on the requests of users, but even this was feature creep that added complexity to our existing design that we had sketched out. We have to keep the board design and review process fluid and lean and we certainly have enough peer review within our team (of highly experienced embedded and EE experts).
The upper tier was priced high enough to cover 2 or 3 of the prototypes per person (they are incredibly expensive in the small numbers that we are producing and to be honest 1000 USD doesn't really cover this), the backers knew what they were getting into when they opted to back this tier and they are all happy, every single person who has backed us (from $1 to $1500) are all people who are passionate about the project and believe in us.
By the way we also put out a request for students and other people who couldn't afford to back anything, we asked them to describe what they do and why they'd like to help and we've also sponsored those spots for them. We think we are doing this in the right way and it is working for us, but of course feel free to start another USB protocol analyzer project on Kickstarter (or elsewhere) that completely fits all of your Open Source criteria.
I understand this is not your average Arduino project. Being an electrical engineer who works with programmable logic design on complex multi-layer boards with small pitch SMT components, I both respect and appreciate the amount of work that you are doing to get it together. The vast majority of people interested in this project probably don't care about the design at the board level let alone have the experience or equipment necessary to build and test a prototype. They just want something that works. The other small minority could have valuable input. It never hurts to have an extra pair of eyes on the design process, which is one of the best things about open source.
You are right about the reason there are no cheap analyzers out there now, but if you are going to develop with a small team and keep the design process closed, why release the hardware files at all? Just sell boards. Having to pay to have access to design conversations is not very "open source".
@smokeymcpot as we originally stated in the project summary "With your investment in our project, we will finalise the design, order the components in bulk (so that automated assembly is viable) and we will have initial prototypes built."
There is a reason there are not any cheap high-speed USB Protocol Analyzers out there (let alone open-source) and this reason is because the hardware development is specialised, expensive and putting it all together is very time consuming. This isn't a run-of-the-mill Arduino project or but a high-speed, SMT four-layer PCB design with various specialist components that need to be designed and produced correctly. This is the task that is underway right now, this is a board that has to be designed carefully and with the input of a few talented engineers (including ones in the highest tier). When the hardware is proven and the board is "brought up" then we'll work on the firmware and software and accept valid software contributions from anywhere. We've got a community (a mailing list and group) that is currently open to backers, this will be extended to the public once we've proven the hardware. I notice that you are not actually a backer, so you'll need to wait a little while longer to see these conversations.
First of all I would like to say I fully support the goals of this project. The technology to cheaply make very useful equipment like this really is now within the grasp of normal people for a lot of applications. I do have a question/comment about the "open source" nature of this project though. The power of open source isn't just having the finished schematics/layouts/code available, but also the collaboration of the community working on the design. It appears that you are developing the system yourselves without having the design available. There are many talented people with experience in relevant fields that would contribute to making the best design if they had access to the design in progress. I may be way off base and just missed the link to the git/svn/cvs repository, but if that is not available, this project is really stretching the idea of open source. Take the lead from Linux kernel. Linus makes the final decisions about what gets included, but anyone can contribute.
@bushing Have you had a look at http://www.elinux.org/BeagleBoard/GSoC/2010_Projects/USBSniffer ?
It is not a FPGA, so will not work at the speed you are designing, but it does work today.
Just to let you know
@Seth Thanks for the comment, it is an interesting point, hopefully I can explain how the current solution is engineered and also the reason why your suggestion isn't that easy for OpenVizsla.
The target USB ports need to be as close as possible to the analysis PHY and because of the high speed signals involved, the PHY has to be strategically placed so that the sniffed signals degrade as little as possible. This is achieved this by placing the USB PHY chip as close to the target ports as we can (and also ensuring that USB traces are not crossed in any way).
The PHY is mounted on the underside of the PCB so that the PHY traces are as short as possible.
All of this means you can't introduce additional electrical connections (such as a USB socket daughterboard) without relocating the PHY onto that same daughterboard (which would be expensive). The PonyPort could accommodate such a board (and is actually designed to do so for test purposes) but it would drastically increase costs if we adopted that configuration for end-users.
This may fall into the realm of overengineering, but coming from using other commercial USB analyzers, one of the biggest issues I've had is with USB connector cycle life. That is, the USB connectors on the analyzer failed only after about a year of use due to the large number of cycles and heavy use it was taking. I always thought it would be a better design to have easily-replaceable USB connectors so if a connector fails due to heavy use, it can be easily replaced (i.e. via a snap-in daughterboard or at he very least, making it easy to desolder (though that could certainly damage the board or other nearby components)). What do you think?
@bushinng - Thanks a lot, can't wait to see the final design!!! Unfortunately i came across your project when the funding was already over so I couldn pledge :/ Anyway thanks again for the efforts you and your crew are putting into this project, keep going!!!!!
@Eugenio - There is a mailing list and community that is currently open to backers, anyone who backed the project (no matter how large or small the contribution) was invited. This will be extended to the public, but right now we are finishing the PCB design and ordering the prototype boards.
Are there any new updates? The blog is stuck un the same post since January the 9th and yet no link to a forum has shown up on www.openvizsla.org. Anyone knows anything?
@bushing - I'm at $250 but will help with Mac OS X implementation for the odd early prototype.
Congratulations on x>70,000 hopefully we can shoot for towards/past 80,000 before the end of the day. This will be a fun and exciting project to help and support in the coming new year. Developing an open sourced anaylzer which will open doors to so many hidden proprietary USB treasures for reverse engineers is no easy feat. So good luck to all supporting this effort, pledge what you can and lets break down that pesky USB barrier. :)
@bushing - Thanks for the clarification. I've upgraded to $150 Funding level as a result. I would have liked to have tried my hand at reflow soldering, but probably best I don't for a 5 layer (?) PCB board.
@Peter Sorry, let me just clarify that for you (the wording is a little ambiguous) all PCBs (populated/assembled or not) that will be sent out will be the final, bug-free versions, only the $1000 backers get access to the initial bleeding-edge development boards.
On the matter of upgrading to the $150 backer level. The $100 level provides 'a final OpenVizsla PCB and kit of surface-mount parts to build their...'. The $150 level provides 'a bare-bones OpenVizsla board from our initial production run". Having a final pcb and a set of parts seems to me to be better then having an 'initial production run board'. I would think that bugs would be addressed in the final pcb which the 'initial production run board' might have some undocumented features. Perhaps it's a symantec thing, but that's how I read it.
Same here, Jacques. I'm in for $150 after reading the hackmii post. :)
I'm glad you post it on hackmii.com, I would miss this otherwise! Can't wait to work with mine!
I don't mean to rush you guys, I'd just like to know if it's to soon to schedule a intern student to work with an openvizsla from the first run in February?
@Peter Yes you could! Thanks for the link.
Could the reflow soldering technique be utilized ?
The January 2011 QST had a article on turning a toaster oven into a reflow soldering oven.
@Bertrand No, there is a CPLD that will be there for the configurable routing of some debug signals. In theory it means we can route JTAG from anywhere to anywhere, same with the serial debug connections etc. The entire thing can be bypassed by jumper wires anyhow.
Do you mean that the FPGA will be replaced by a CPLD XO2 ?
I'm writing my diploma thesis regarding crowdfunding and I'm urgently looking for interviewpartners which financed or funded a project.
Please write me at Facebook or firstname.lastname@example.org if you would have time for a skype interview.
Many thanks in advance,
I officially jumped to the $500 level, awesome to see everyone doing so well! I did my best to petition the EFF but they said they're not a granting organization. I again ask if there are any other supporters here to write them an email. I didn't think to mention the $2000 dollars they got as a donation for someone writing a driver for the Kinect. I'd love to see the EFF logo on the board just because this type of project is something that is right up their alley.
On another note do you guys need any help writing file format converters or anything like that? Something that converts your native capture format into the format that some of the existing software packages might use?
@Christian Funny you should say that, we already have testpoints on the target VBUS line (you can see that here) - http://dl.getdropbox.com/u/20978/tp.png and we have talked about automatic power monitoring, so because of your request we'll add the functionality back into the OpenVizsla prototype, this will mean that there will be current shunt and power monitoring of the target bus directly into the FPGA. If this works out like we hope it will, we'll add it to the final board design.
This might be way out of scope, but would it be possible to measure power draw? Some hardware might behave strangely due to peak power not being supported on some machines but not all, and measuring this would help debug those situations. What's the maximum power draw the USB device you are sniffing can have?
If you don't already have designed into the hardware the ability to inject as well as listen to the data stream, this might be something to consider. Not meaning to add to "feature creep", and not suggesting that things be delayed for the additional software development time, but it's a 'lot' easier to add software features if the hardware will support it (grin). I guess it depends on how far along you are with the hardware design / development.
This user's account has been deleted.
I spoke with my wallet when the project was announced. I even adore the name PonyPort, even though it will probably be misconstrued as "Horse's Ass" by those who can't "get it to work." This *IS* overdue, usbmon be damned, and it *IS* going to change how we look at OSS/OSH and, now, OS$.
However, I have only one request prior to receiving my goodie bag:
NO FEATURE CREEP! Meet the deadline.
@David We can ship within the EU also, we didn't list the additional costs, when we were creating the tiers. So if you are feeling generous you could add ~6 bucks to cover postage, but it isn't mandatory.
If I pledge 100 dollars, will the PCBs/components be shipped to the UK, or is this higher for other countries? And fantastic project, should be some great possibilities opened up with this!
A user called 'DermotS' contacted us earlier via Kickstarter's messaging system, but there seems to be a few issues with that at the moment. DermotS could you please contact us via email at email@example.com - thanks!
@Omer Thanks for the support.
This has been discussed in detail below and bushing's response is promising but just wanted to echo what's been said about the dual row QFN package the L2 comes in. It's really not a hobbyist friendly package and choosing it would seriously hinder the DIY aspect of the project. Other than that, can't wait to receive my PCB and start hacking away. Thanks and good luck guys!
Hopefuly this will be under the $150 USD range. but then again its one of its kind being open source and all. backing $500 for 3 that puts it in that range but that is for now, you guys have to get some profit from the sells when launched looks to me it will be at $200+ to $260 each my opinion. 40K make it last its while. can't wait to see the "custom casing that will make the outside of OpenVizsla look as awesome as the inside." w00t² <-- yes I squared w00t
@Devin sure! We can make a note of this for you.
We are probably adding something else to the tier levels (an additional giveaway), but we have not announced it yet (we are still waiting for confirmation from a supplier).
@bushing I'm considering bumping my backing to the $500 level, did you plan on providing three of the badges at that level? [One for each device?]
Congrats! This is a great idea I can't wait and you got more than DOUBLE the amount! Go go go go go go go go GOD SPEED! Rock it!
This is a great project, I'm not much of a hardware hacker but I can certainly see the value in this device and want it to become a reality! I also sent an email to the EFF in hopes that they might support this project. I think it goes right along with their core mission and if there are any other EFF supporters that agree with me please take the time to send them an email letting them know you support it too.
@vincent - thanks for your question, just wanted to explain my comment about the QFN earlier. We are rapidly prototyping our test/development boards not hand assembling (as would be usual) the boards at this stage. This involves getting 15 boards made (at each design iteration) using pick&place/reflow techniques. These boards are destined for the core team and developers, because of this we have been able to use parts that are not necessarily home reflow/hotair friendly (although all these will be tested and attempted at home anyhow). The L2 dual-row QFN part in question is used for this section of the development cycle, but can be easily substituted for the final board (the L2 is just 2 x L1 dies in a dual-row QFN package) and can easily be added to the layout before final production.
I could back from some country in europe, if so try to register in amazon.co.uk or similar, at least I could outside the US.
Why on earth mrQFN packages. BGA's are waaaay easier to solder than mrQFN . Just flux em up and reflow em. mrQFN on the other hand.. I work on several very large projects that use mrQFN and there are tremendous teething problems. Even large assembly houses look weary at these packages.
Other question : Do you have screenshots of the software driving this thing ? Excellent visualisation of packets is key. Can we expect something like the CATC machines do ? or better ;) ?
But it sounds a cool enough project. The badge is likely to be more personally useful to me than the hardware though.
@Patrick, @Matthew, @Chiark ; I just signed up from the UK and also had a bizzare trans-atlantic relocation experience. Went through the sign-up lieing that I was in Yoder, Wyoming Zip 82244 (mostly number the code zip liked I did, because excrement StarWars i twice not give) and eventually got something that looked like acceptance. Then I came back to the kickstarter page which had given me an error, logged in again ... and it took me to my UK Amazon account.
For sure, there's some bizzareness in the payments set up.
excellent project! hope you guys can handle the order. If you need help soldering, or setup, bring some to the CCC, would be happy to help ;)
@Jeff: Thanks - that worked for me!
You guys should cap the donations at 256 backers. Otherwise, the next stop would be $65536. I'm hoping for the latter :)