[TriEmbed] Lattice FPGA Development Kit

Sun Aug 20 15:48:52 CDT 2023

First I highly recommend everyone to create their own business.  When you
work for someone else you learn how to do things, when you work for
yourself you start learning the right things to do.  Additionally when you
are in business and need to make income, you learn how the value creation
feedback loop works.  That is all your friends and contacts will offer you
advice, but until you tune *your *feedback loop for *your *business on how
to create value for *your *customers you are not really in business, it is
more of a hobby.

I would say the key to success is to create more value to customers than it
costs them to get it from you.  That is your FPGA board must create value
to customers, it does not need to be thoughtful, or clever, or even well
engineered.  It needs to provide a positive return on value for your
customers.

1.  What is different about your board than other FPGA eval boards from
many FPGA vendors and suppliers?
2.  What value does your board and examples does it have that helps me
solve my problem? Is it designed for video filtering, etc?

I know I created a number of products for myself as an engineer.   When I
finally created a product that solves someone else's problem, I marketed,
licensed it all wrong and in less than 6 months China had taken the design,
firmware and the market.   That is that learning was more valuable than
anything I did previously, and I would not have learned that without
creating business and doing it.

A single product does not make a business!
If you build it they will not come!
If you think that if you build it, that they will come, then put the
product up for sale on a website and say it is out of stock, see how many
inquiries you get as to when it will be back in stock.  This will prove
you are right and if you build it they will come.
If you are not in business, you are not improving your business!
It is not about doing it right, it is about learning and improving!
The process of picking products to make is more important than a product!

So here are a few ideas on how to create value:

*Phy protocol analyzer*
We have all used scope for UART/SPI/I2C/CAN/USB signals.  There are
products like https://www.saleae.com/ that are great.  However what if you
had a product that actually did measurements and reports and told you the
problems.  For example a device that reported UART baud rate and other
parameters and reported rise time, fall time, jitter, etc.  Provide
engineers the information they would normally use a scope for.  Then give
them all the data in a nice report format that you would be proud to show
off.

*Chip Replacement*
Many people restore old hardware (PCs, etc) they are often looking for chip
replacements, like a replacement for an 8086 or 8051, or UART controller.
So what if you took an FPGA and made chip replacements?

*Floppy Drive Replacement*
I have old equipment that uses floppy drives. I looked for a USB flash
replacement for a mechanical floppy drive.  The devices I found all had
major issues. I contacted a friend who restores old PCs and he indicated he
wanted a better floppy drive replacement since he found the same issues.
What if you used FPGA to make such a device, maybe at ESP32 such that files
are available over wifi?

*EEPROM/ROM Replacements*
I have older cars and thought about having an EEPROM replacement where I
could change the tuning of the car without flashing new firmware. Like the
floppy drive the idea was to have a PCB that replaced the memory chip with
USB or WIFI access.  Then on the fly I could change the tuning of the car
computer without reflashing chips.

*Buck Boost Controller*
I find many engineers do not understand buck/boost power supplies.  What if
you made a training system where you use an FPGA as a controller to make a
'soft' buck/boost controller? Heck even better is to make the FPGA used for
analyzing power supply controllers reporting results, like phase margin,
etc. Then give them all the data in a nice report format that you would be
proud to show off.

Are any of these ideas worth creating a product?  I do not know, but they
might be more valuable than a generic FPGA eval board, as they solve a
specific problem out of the box in which you can find the correct customers
to market products to.

Trampas

On Sun, Aug 20, 2023 at 1:45 PM Kevin Schilf via TriEmbed <
triembed at triembed.org> wrote:

> Hi TriEmbed,
>
> This is a shameless attempt at gathering marketing data...  :-)
>
> I am putting together a FPGA development board with a Lattice FPGA.
>
> Why choose a FPGA over a micro you say?  Good question.  FPGA's are more
> expensive and less fuel efficient than their micro cousins but offer:
>
> - Higher throughput
> - The ability to create multiple parallel processing blocks.
> - Numerous I/O standards with more precise timing.
>
> If you don't need one of these three things you are probably better in a
> micro; although, I would never say no to a FPGA solution.  :-)
>
> Possibilities -
>
> HW Interfaces- Trusty GPIO of course up to about 1 Gbps, differential
> standards as such LVDS over 1 Gbps, high speed SerDes (serializer /
> deserializer) with bit rates up to several Gbps (PCIe, Ethernet, JESD, etc.)
>
> Soft Processor - Avoiding the expensive license fees associated with ARM,
> RISC-V is an open source Instruction set processor out of Berkeley.  Not
> only can you program the processor but you can change the opcodes by
> implementing custom instructions in the FPGA's fabric.  Linux is
> available.  This is a very exciting possibility  :-)
>
> IP - baseline RTL (Verilog or VHDL) code to implement certain
> functionality, video processing, motor control, DSP, etc.  This is the
> secret sauce of the FPGA, but I am looking for building blocks that could
> be created in advance and absorbed into more sophisticated solutions to
> ease getting started.
>
> Price - What is a reasonable price point in a professional, research, or
> academic setting?
>
> The key to success if providing a thoughtful piece of hardware WITH
> documentation and sample code that makes it useful out of the box.  There
> are enough PCB paperweights in the world already.  :-)
>
> I look forward to your input.  Thank you for your time.
>
> Kevin S
> _______________________________________________
> Triangle, NC Embedded Interest Group mailing list
>
> To post message: TriEmbed at triembed.org
> List info: http://mail.triembed.org/mailman/listinfo/triembed_triembed.org
> TriEmbed web site: https://TriEmbed.org
> To unsubscribe, click link and send a blank message: mailto:
> unsubscribe-TriEmbed at bitser.net?subject=unsubscribe
> Searchable email archive available at
> https://www.mail-archive.com/triembed@triembed.org/
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.triembed.org/pipermail/triembed_triembed.org/attachments/20230820/f34beb6d/attachment.htm>