proto-talk

[jadew]

Hey guys,

This is my project: https://www.indiegogo.com/projects/proto-talk/x/8683663

I have no media editing skills, so that's the best I seem to be able to do in terms of pitch video & demos - if you have any suggestions please let me know.

I'd be happy to hear any comments and answer any questions you may have, so... let the bashing begin.

Thanks,
Razvan

[8086]

So it's a bus pirate?

http://dangerousprototypes.com/bus-pirate-manual/bus-pirate-feature-overview/

[jadew]

So it's a bus pirate?

http://dangerousprototypes.com/bus-pirate-manual/bus-pirate-feature-overview/

Like the bus pirate, but with much beefier specs. Aside from the high baudrates, it also features programmable ouput levels, between 1.2 V and 5 V and very low baudrate error (much better even than dedicated interface chips - at least the ones I worked with).

I never used the bus pirate, but I expect there are lots of differences, especially regarding the performance of the device.

Edit:
Again, I never used it, but I suspect it has similar issues with my previous versions. As I said in the main video, one of the primary reasons I started working on this version was the baudrate error. I believe I was getting more than 10% error between 1 and 200 kbs. The proto-talk has less than 0.2% from 1 bps to 10 Mbps.

[sunnyhighway]

I really like that you are brutally honest in both your video and text. The funders do like that too.

But, you might want to talk/write more about the following:

Use cases:
Address the communities/platforms you are aiming for and what it is useful for. Arduino, Raspberry pi, other mcu's, debugging, testing?
How do you use it? You only mentioned that you don't need to write a single line of code to operate it.

Additional requirements:
Do you need a PC to operate it? I have no clue. You might want to tell more about that.

[jadew]

@sunnyhighway:

Thank you for your suggestions, I will address both issues in the text.

To quickly answer your question: yep, you need a PC as it's operated via a command line interface and shows up as a virtual serial port on the PC.

There are two videos (at the moment) in the gallery, that show how it's being used for 1-Wire and I2C, but they don't make any hint on all the usage scenarios.

Thanks!

Edit: Updated.

[Spikee]

The product seems well engineered and there is probably a large enough market to get to your funding goal.

But what you need is some marketing. Talk about your idea's / product on forums like Eevblog and get feedback before you launch the campaign.
Make a website for your company/product and show case it a bit. A reasonable good example of this is the Pax Instruments T400 campaign.
http://paxinstruments.com/

When those things are done you can contact the people at several blogging websites like hackaday and dangerous prototypes and ask if they are willing to feature you on their website. In the way you are doing it now it will probably be very hard to achieve your funding goal.

[timb]

Looks like an ARM powered Bus Pirate pretty much, yeah. Though, with a few more protocols. (Maybe you'll actually develop firmware at s rate faster than the DP guys. BPv4 hasn't been updated in over a year!)

Here's a suggestion: Add a mode to emulate an STK500v2 so it can be used for programming AVR chips.

Support for Spy-Bi-Wire would be nice too. I haven't seen that on any other device.

Seriously though, you should grab a Bus Pirate and see how they handle things like enabling/disabling pull-ups and the ability to use arbitrary voltages through s pullup input line; the ability to switch on and off the 5v/3v3 regs, stuff like that. They put a lot of thought and many revisions into it, I think you could very well borrow a few features.

Then, maybe rename your product to the "Protocol Ninja".


Sent from my Smartphone

[jadew]

@Spikee

Thanks for the suggestion, I really like how the PAX guys are presenting their campaign.

About the website, it is planned, but I didn't get enough free time to work on it. If it gets successfully funded, I plan to have a section for user made applications and modules (that can be ran from the SD card) as well as a small support forum and a help section. At this point it's kinda hard to dedicate time to that, since it would mean less time to work on the device itself.

My marketing skills are completely lacking (I guess it's obvious by now) and I don't want to get all spamy so I'll probably keep discussions confined to this thread.

I will look into your hackaday suggestion, but I always had a feeling that they wouldn't like getting contacted for the purpose of self-promotion.

Thanks,
Razvan

[sunnyhighway]

When those things are done you can contact the people at several blogging websites like hackaday and dangerous prototypes and ask if they are willing to feature you on their website. In the way you are doing it now it will probably be very hard to achieve your funding goal.

I completely agree with Spikee. The most common mistake made by failed fundings is lack of exposure.
 Before you launch your campaign, make sure you have as much exposure as possible ready to be revealed to the world during the funding campaign.

[jadew]

Looks like an ARM powered Bus Pirate pretty much, yeah. Though, with a few more protocols. (Maybe you'll actually develop firmware at s rate faster than the DP guys. BPv4 hasn't been updated in over a year!)

Here's a suggestion: Add a mode to emulate an STK500v2 so it can be used for programming AVR chips.

Support for Spy-Bi-Wire would be nice too. I haven't seen that on any other device.

Seriously though, you should grab a Bus Pirate and see how they handle things like enabling/disabling pull-ups and the ability to use arbitrary voltages through s pullup input line; the ability to switch on and off the 5v/3v3 regs, stuff like that. They put a lot of thought and many revisions into it, I think you could very well borrow a few features.

Then, maybe rename your product to the "Protocol Ninja".


Sent from my Smartphone

Hey Timb,

I have already wrote code that is implementing the AVRISP2 protocol, it's inclusion is guaranteed.

About the pullups: It already supports several pull-up values for two of the pins that are used for I2C, 1-Wire and DHT (at the moment). The level they pull up the value to is programmable between 1.2 V and 5 V, in 1.3 mV steps. It has a small on-board linear power supply, which is setting both the output levels and the levels of the pull-ups.

About the updates - yeah, I think it will receive updates for a long while.

As for protocol support, I plan to add support for as many protocols as I can. On the campaign page, only the features in the current prototype are listed, but more will come.

Thanks

[all_repair]

Being already a regular here on eevblog forum before the fund raising give a good feeling.    Is it open source?  My concern  is will more potocol engines be available as time goes by, or like most other hardware that become an orphan after 1 or 2 update.  Most of us have a bus pirate I think, you may to consider this in prioritising your offering.

[jadew]

@all_repair

Hey, it's not opensource. The reason for that is because I don't want it to be a one off. I plan to keep improving this device long after the campaign is done and I can't do that if I'm being undercut by cheap clones.

There will be an SDK which will allow the creation of bitbanged protocols or the use of the already implemented ones for higher level communication.

These applications will be able to run from an SD card (note that it comes with an SD card slot).


Sorry if this reply is malformed, I'm on my phone at the moment.

Thanks

[Spikee]

jadew if your want i can help your with the marketing and maybe even with the device.
Your can contact me via email.
I have time tomorrow.

[all_repair]

Supported.  Frankly, I don't really know what I am going to get.   I think it is going to get better from what you said, and may evolve into something I can use.

[zapta]

To quickly answer your question: yep, you need a PC as it's operated via a command line interface and shows up as a virtual serial port on the PC.

By PC do you mean Windows or a Mac will do just the same?

[jadew]

@all_repair
Thanks a lot, I appreciate it.

I hope it will serve you well and you can rest assured that it will get nice firmware updates.

Thank you,
Razvan

[jadew]

To quickly answer your question: yep, you need a PC as it's operated via a command line interface and shows up as a virtual serial port on the PC.

By PC do you mean Windows or a Mac will do just the same?

Should work on any OS, as long as you have a terminal client (eg. PuTTY). It gets exposed as a COM port and on top of that, it ignores the line settings, so you can use any baudrate and any settings for the connection and it will still work right. This is particularly useful on Linux which has inherent issues with non standard baudrates.

Anyway, short answer is Yes

I will make some screenshots when I get home, on all 3 operating systems and post them into the gallery section.

Thanks,
Razvan

[zapta]

Should work on any OS, as long as you have a terminal client (eg. PuTTY). It gets exposed as a COM port and on top of that, it ignores the line settings, so you can use any baudrate and any settings for the connection and it will still work right. This is particularly useful on Linux which has inherent issues with non standard baudrates.

Anyway, short answer is Yes

Thanks, you may want to convey the cross platform feature on project page. It will give you extra points ;-)

I also presume that scripting and automation is possible (having the host script talking via the com port).

[jadew]

@zapta

I'm gonna add the screenshots to this post from each individual OS.

I think all 3 programs are available both for linux and mac, while for windows only PuTTY is available. There are obviously other choices in serial terminal clients, but this is what I had installed.

Razvan

[timb]

CoolTerm is a good choice on OS X.

If you want to go Terminal, no need for PuTTY under OS X or Linux, screen will do the job just fine-- and it's built-in!


Sent from my Smartphone

[jadew]

@Timb,

I didn't have screen installed on Mint (couldn't get it installed either - the repository paths have changed or something and didn't have the time to get it fixed).

Razvan

[jadew]

I just added support for smart cards, protocol ISO 7816-3, T=0. I'll get back to this and work on T=1 when other, more important things are done.

A short demo video can be found in the gallery.

Edit: If you're wondering what's with the 10 private comments (it says 10 in there, but you can't see any comment), they're spam.

From the first day the campaign got posted, I started receiving various offers to get many shares and likes or $1 contributions + nice comments, usually for a very low price, like $5.

I didn't buy into any of that and I'm going to delete them now.

Edit 2: I finally calculated the baudrate error for SPI & I2C. It's less than 0.18 for SPI (with one exception at 426 bps, which becomes 427 bps, so 0.23% error).

For I2C, it's less than 0.14%.

[jadew]

Hey guys,

It appears that the project will fail, but I'd still like to thank you and let you know that I appreciate the support you offered.

For the record I did contact Hackaday as suggested, but they didn't write back - I wonder why, I thought it would be right up their alley.

Anyway, given the impending failure, I'm considering where to go from here with this project and I'd appreciate some suggestions. Judging by the low number of visits (500 total) and the high conversion rate (5 backers), I'd say the reason for its failure was the lack of traffic.

At this point I'm ready to move on to other stuff, but before I completely give up, I figured I should ask and see if I'm missing anything.

1) Do you think this project is worth pursuing further? Why?
2) Any ideas on what I could do better/different to get this off the ground?


Thanks

[all_repair]

I think you are not ready, and with that, you should still be able to reach your target amount if you have enough promotion.  But I am biased being a supporter.   I have a bus pirate, and people shall not mind to get another if yours can be another serial tool but cover what bus-pirate is not already covering.  The biggest value I wanted was the continuous software features you were going to put in. 

[marshallh]

Your problem is that you are being too honest/modest with your marketing.
Even if you list in the spec sheet that the device cures cancer, no one will read it.

An ambitious tagline helps - "Prototalk - Better than bus pirate" would be an easy start.
Then make 3 bullet points of how its better.

I don't see anything wrong with the product itself, it just failed to gain enough momentum (which is a fickle beast).

edit: "About me" shouldn't be the first section, rather it should be in the middle or near the end. When I'm reading I want to see quickly if it's useful or not, I don't care who made it yet.