What do you guys think about this?
https://www.crowdsupply.com/krtkl/snickerdoodle
That looks like an awful lot of expensive IC's to be at the price point they are promising?
It's gotta be subsidized. Either that, or they expect to sell 500,000 of them so they can get the $75 Zynq down to something less than the selling price. And their $55k goal ain't gonna do it.
Is 28K logic cells a reasonable amount of logic for a FPGA?
Is 28K logic cells a reasonable amount of logic for a FPGA?
Where are you seeing that? The two options have 430K and 1.3M gates, respectively.
Snickerdoodle avoids one fault common in "amateur/hobbyist" boards: it has sufficient well-distributed GND pins on the connectors.
Another traditional weak spot is IO Voltages and PSU voltage sequencing, particularly with different IOBank voltages; the Zynq is sensitive to that. I haven't assessed their design, and would be interested in informed comments.
Is 28K logic cells a reasonable amount of logic for a FPGA?
Where are you seeing that? The two options have 430K and 1.3M gates, respectively.
Look at the specifications of the
Zynq SOC. The quantity of Logic cells (and macro cells) within an FPGA a useful parameter. The total number of logic gates, not so much.
The pricing is about as crazy is Andreas with the Parallella.
They pulled it off, but not so much to a profit.
So it's not impossible at least.
Hello, one of the snickerdoodle creators here! Happy to answer any questions you might have (I'll start with addressing the previous posts)...
-Ryan
What do you guys think about this?
https://www.crowdsupply.com/krtkl/snickerdoodle
That looks like an awful lot of expensive IC's to be at the price point they are promising?
It's gotta be subsidized. Either that, or they expect to sell 500,000 of them so they can get the $75 Zynq down to something less than the selling price. And their $55k goal ain't gonna do it.
Hi Bassman - no subsidy (unfortunately). While I can't go into too much detail on pricing for confidentiality reasons, I will say that the $50-75 price you see online is not the volume/distributor pricing. But suffice it to say that yes, we are relying on people to take a 'leap of faith' so we can get some decent initial volumes (vs. just sitting back and saying: I'll buy it when their community is the size of Raspberry Pi's). Hence the original goal: $55k (1,000 snickerdoodles).
Because each component and it’s upgradable counterpart are made by the same respective manufacturer, we are able to realize combined volume discounts across multiple SKUs - helping us pass savings for both the base and upgraded models along to you.
Hmmmm...
Well, at least they're upfront about 'krtkl'...
It’s pronounced “critical” (kr?t??-k?l)…we just removed the vowels and changed the “c’s” to “k’s!”
Marketing stuff right there.
Haha - reading that again, that's a pretty poorly articulated description of 'how' we're able to offer so many combinations. The bottom line is: design and component selection. It's designed to take interchangeable parts and the each upgradable part comes from the same manufacturer. So aside from dealing with the large # of SKUs, it's really not that big of a deal to load one part vs. another. Hope that's a little clearer.
Snickerdoodle avoids one fault common in "amateur/hobbyist" boards: it has sufficient well-distributed GND pins on the connectors.
Another traditional weak spot is IO Voltages and PSU voltage sequencing, particularly with different IOBank voltages; the Zynq is sensitive to that. I haven't assessed their design, and would be interested in informed comments.
Yes! Thanks for recognizing this tggzzz. In fact it's so common that almost every article about snickerdoodle so far has mentioned 154 I/O pins...when if fact it's 154 I/O across 230 pins...(179 across 270 pins for the 7020).
Snickerdoodle avoids one fault common in "amateur/hobbyist" boards: it has sufficient well-distributed GND pins on the connectors.
Bloody hell, I don't even care if the rest of the design is total shite, I want to give the creators a big hug just for that.
Snickerdoodle avoids one fault common in "amateur/hobbyist" boards: it has sufficient well-distributed GND pins on the connectors.
Bloody hell, I don't even care if the rest of the design is total shite, I want to give the creators a big hug just for that.
That's what we like to hear, cp! Big hug accepted
Now tell everyone you know!
-Ryan
Snickerdoodle avoids one fault common in "amateur/hobbyist" boards: it has sufficient well-distributed GND pins on the connectors.
Another traditional weak spot is IO Voltages and PSU voltage sequencing, particularly with different IOBank voltages; the Zynq is sensitive to that. I haven't assessed their design, and would be interested in informed comments.
Yes! Thanks for recognizing this tggzzz. In fact it's so common that almost every article about snickerdoodle so far has mentioned 154 I/O pins...when if fact it's 154 I/O across 230 pins...(179 across 270 pins for the 7020).
IMNSHO the creators would do well to make that point extremely clear at
every relevant point in their website, including a statement about why GNDs are necessary. It is a significant advantage over other boards, so they should trumpet their (almost) USP! Commentards might then pick up on it.
What do you guys think about this?
https://www.crowdsupply.com/krtkl/snickerdoodle
That looks like an awful lot of expensive IC's to be at the price point they are promising?
It's gotta be subsidized. Either that, or they expect to sell 500,000 of them so they can get the $75 Zynq down to something less than the selling price. And their $55k goal ain't gonna do it.
Hi Bassman - no subsidy (unfortunately). While I can't go into too much detail on pricing for confidentiality reasons, I will say that the $50-75 price you see online is not the volume/distributor pricing. But suffice it to say that yes, we are relying on people to take a 'leap of faith' so we can get some decent initial volumes (vs. just sitting back and saying: I'll buy it when their community is the size of Raspberry Pi's). Hence the original goal: $55k (1,000 snickerdoodles).
I realize that the single-piece $60 price is not the volume price, but even if the FPGA is $25 after you've bought 1,000, and you're buying 10,000 of the Samtec connectors, and the WiFi chip, and then the boards themselves, plus assembly, I don't see how you're breaking even.
Board looks great, but the pricing seems a bit low at only a 1000pcs goal, like some already have mentioned.
If they can deliver it at that price, great. But my Dutch pessimism holds me back from ordering one in a crowdfund with this pricepoint and goals.
Snickerdoodle avoids one fault common in "amateur/hobbyist" boards: it has sufficient well-distributed GND pins on the connectors.
Another traditional weak spot is IO Voltages and PSU voltage sequencing, particularly with different IOBank voltages; the Zynq is sensitive to that. I haven't assessed their design, and would be interested in informed comments.
Yes! Thanks for recognizing this tggzzz. In fact it's so common that almost every article about snickerdoodle so far has mentioned 154 I/O pins...when if fact it's 154 I/O across 230 pins...(179 across 270 pins for the 7020).
IMNSHO the creators would do well to make that point extremely clear at every relevant point in their website, including a statement about why GNDs are necessary. It is a significant advantage over other boards, so they should trumpet their (almost) USP! Commentards might then pick up on it.
Point taken, tg. We'll be sure to emphasize it with this next round of revisions! Thanks for the input.
Board looks great, but the pricing seems a bit low at only a 1000pcs goal, like some already have mentioned.
If they can deliver it at that price, great. But my Dutch pessimism holds me back from ordering one in a crowdfund with this pricepoint and goals.
Hi Crash, there may not be much I can say to change your mind but I can personally assure you that if we hit our goal, you will get a board. There is nothing deceitful or dishonest about what we're doing and we would not have posted the project at this price (or any price/goal) if we knew we could not deliver. Would exceeding our goal by a large amount help? Of course! And that will be key to the longterm success of the platform and community. But we ultimately need a group of risk takers, experimenters, and dreamers - no matter how small - to help make it possible for us to get this stuff into the hands of makers and developers like yourself.
We have been doubted every step of the way working on this thing. I lost count of how many times people told us it couldn't be done or that we should just give up and do something else. Everything you see is the result of three guys with a burning desire to give people better tools for building great things. If you don't believe it's possible or want to add your name to the long list of doubters, I have no problem with that. But if you want to 'roll the dice' to potentially be part of a powerful new movement and join the small (but steadily growing) list of people who desperately want to see this project be successful because they believe in doing things that have never been done before, then we warmly welcome you aboard.
What do you guys think about this?
https://www.crowdsupply.com/krtkl/snickerdoodle
That looks like an awful lot of expensive IC's to be at the price point they are promising?
It's gotta be subsidized. Either that, or they expect to sell 500,000 of them so they can get the $75 Zynq down to something less than the selling price. And their $55k goal ain't gonna do it.
Hi Bassman - no subsidy (unfortunately). While I can't go into too much detail on pricing for confidentiality reasons, I will say that the $50-75 price you see online is not the volume/distributor pricing. But suffice it to say that yes, we are relying on people to take a 'leap of faith' so we can get some decent initial volumes (vs. just sitting back and saying: I'll buy it when their community is the size of Raspberry Pi's). Hence the original goal: $55k (1,000 snickerdoodles).
I realize that the single-piece $60 price is not the volume price, but even if the FPGA is $25 after you've bought 1,000, and you're buying 10,000 of the Samtec connectors, and the WiFi chip, and then the boards themselves, plus assembly, I don't see how you're breaking even.
Bassman, you can see my other response. We've spent two years grinding on this thing - iterating the design, adding/removing things, architecting the hw+sw, negotiating with vendors, working out the supply chain, revising & revising & revising the business model. It didn't make sense to us on the surface at first either. You can take my word (or not) that we can afford to keep making this product in the long term. All we ask is you help us get it off the ground.