[ANN] Bitfury is looking for alpha-testers of first chips! FREE MONEY HERE!

[ssi]

Four more chips ordered, I hope they come quick!


Meanwhile, have these four hashing away on my desk at work:

[intron]

Working on bin-fury, a bitfury binner:



It has a clamp shell socket that can hold an ASIC
and using firmware a sequence of test patterns
and voltages is applied while measuring the response.
A OLED display is used for displaying the results.

For all your 'champ or tramp' tests:)

intron

[ssi]

Working on bin-fury, a bitfury binner:

It has a clamp shell socket that can hold an ASIC
and using firmware a sequence of test patterns
and voltages is applied while measuring the response.
A OLED display is used for displaying the results.

For all your 'champ or tramp' tests:)

intron

Sweet!  I was just thinking about something similar yesterday. 

[kaerf]

Four more chips ordered, I hope they come quick!


Meanwhile, have these four hashing away on my desk at work:

https://pbs.twimg.com/media/BSmoBr-CAAA5lXq.jpg

cool. that seems quite useful.

[MWNinja]

https://pbs.twimg.com/media/BSUpPk0CIAA2YGU.jpg


I lack a level shifter.  Should be here tomorrow.

what you lack in level shifters, you make up for in density - nice work!

how do foresee mounting it?

Honestly I didn't give a lot of thought to mounting when trying to get these prototypes spun quickly   Next rev will have more options.  For this one, I can just 3d print something with slots.

However, what I DID think about:

it supports RPi:



AND Beaglebone:

I'm very happy to see support for the BeagleBone!

[ssi]

I'm very happy to see support for the BeagleBone!

Full disclosure:  the current cgminer is very RPi specific.  Some code hacking is necessary before BB will work!

[KNK]

Full disclosure:  the current cgminer is very RPi specific.  Some code hacking is necessary before BB will work!

I have just posted a thread about this problem - https://bitcointalk.org/index.php?topic=282034.0
Pretty soon it will became a nightmare to find the correct software for your hardware, so i would like to invite all software and hardware developers to comment on what they need as communication between the two

[ultrix]

Working on bin-fury, a bitfury binner:

It has a clamp shell socket that can hold an ASIC
and using firmware a sequence of test patterns
and voltages is applied while measuring the response.
A OLED display is used for displaying the results.

For all your 'champ or tramp' tests:)

intron

I would avoid the tall chinese sockets.  I think I ruined a chip on one.  High ESR due to the size of the loop between the power pins and ground bolt.  The service record on the power supply said it was calibrated in early May.  You can find, albeit expensive, low profile ceramic test sockets that are for low ESR.  The wiring is reminiscent of wire bonding.

[-ck]

I'm very happy to see support for the BeagleBone!

Full disclosure:  the current cgminer is very RPi specific.  Some code hacking is necessary before BB will work!

That's an understatement  Oh and it's not "the current cgminer", it's the forked code based on an older version. I received a bitfury USB stick from 101BTC and I've spent a lot of time trying to get it working on a PC instead of an RPi and it's proving far from trivial... Some of the memory and hardware hacks just don't even exist outside of the RPi and some I haven't remotely figured out what they do.

[needbmw]

My first DIY bitfury solar miner.
The box under panel contains RasPi and custom single chip mining board (its heatsink is sticking out of the box).
The chip's core powered directly from the panel (Uoc=1.15V, Isc=5.5A, Ucore=0.87V), and RasPi powered from the cord.
After the first half hour of mining I observed unusual high hashrate (>3Gh/s) of single bitfury chip.
Just to mention I was unable to get more than 2.9Gh/s from this chip when it was powered from buck converter (0.85-0.9V).
You can follow hashrate graph of my solar miner at http://eligius.st/~wizkid057/newstats/userstats.php/15YLSMUNxNZpWjjkego2fNJw78BgigZiiu

[PWRMAD]

Hi
I am interested in developing the Bitfury “H” board into a standalone (computer driven) miner, I am primarily a hardware engineer and do very little with software so I need some real help. I would like to propose a single chip interface (CY7C65211) as the USB to SPI adapter. This UART is very flexible and from everything I have read in the forums just may work. I am fully aware power brick will be required and I have omitted the “H” board power connections from the drawing. I really am interested in your comments as to if this is even possible and if so what software would recognize it. I am inserting a rough drawing for your review and comment. This drawing is basically direct from the app note.

Thanks and please be kind.     

http://www.krw.net/sch2.jpg

[intron]

Hi
I am interested in developing the Bitfury “H” board into a standalone (computer driven) miner, I am primarily a hardware engineer and do very little with software so I need some real help. I would like to propose a single chip interface (CY7C65211) as the USB to SPI adapter. This UART is very flexible and from everything I have read in the forums just may work. I am fully aware power brick will be required and I have omitted the “H” board power connections from the drawing. I really am interested in your comments as to if this is even possible and if so what software would recognize it. I am inserting a rough drawing for your review and comment. This drawing is basically direct from the app note.

Thanks and please be kind.     

You need a bit of processing power to get the bitfury ASICs
to do useful work. Can this CY7C65211 execute user defined
code or is it just an USB transceiver chip? If the latter is
the case, you could add a processor for that. We use an
ARM Cortex M3, fast and cheap.

Also the ASICs have 1V8 IO signals, so there should be
level shifters somewhere in the SPI path. Coming from the
processor it could be just resistor dividers (SCK and MOSI),
coming from the ASICs it can be a small chip (MISO).

intron

[PWRMAD]

Hi and thanks for your excellent feedback, the chip can be and is configured to operate using 1V8 so no level shifter should be required (one of the reasons I selected it) also it is very configurable with the utility. Just a few of the specs are;

Single-channel configurable SPI interface ❐ Master/slave up to 3 MHz ❐ Data width: 4 bits to 16 bits ❐ 256 bytes for each transmit and receive buffer ❐ Supports Motorola, TI, and National SPI modes
Configuration utility (Windows) to configure the following: ❐ Vendor ID (VID), Product ID (PID), and Product and Manufacturer descriptors ❐ UART/I2C/SPI ❐ CapSense ❐ Charger detection ❐ GPIO

The link for the full spec sheet is here; http://www.krw.net/cy7c65211.pdf

Also please keep in mind that this is simply an interface to allow a computer to provide the processing power not the UART.

Thanks again

[Sitarow]

I'm very happy to see support for the BeagleBone!

Full disclosure:  the current cgminer is very RPi specific.  Some code hacking is necessary before BB will work!

That's an understatement  Oh and it's not "the current cgminer", it's the forked code based on an older version. I received a bitfury USB stick from 101BTC and I've spent a lot of time trying to get it working on a PC instead of an RPi and it's proving far from trivial... Some of the memory and hardware hacks just don't even exist outside of the RPi and some I haven't remotely figured out what they do.

This is exciting, looking forward to your results.

[ssi]

I'm very happy to see support for the BeagleBone!

Full disclosure:  the current cgminer is very RPi specific.  Some code hacking is necessary before BB will work!

That's an understatement  Oh and it's not "the current cgminer", it's the forked code based on an older version. I received a bitfury USB stick from 101BTC and I've spent a lot of time trying to get it working on a PC instead of an RPi and it's proving far from trivial... Some of the memory and hardware hacks just don't even exist outside of the RPi and some I haven't remotely figured out what they do.

This is exciting, looking forward to your results.

I intend to sit down and try to get it ported over to BBB soon; I need to set aside a single chip to place for testing purposes.

[vs3]

I'm very happy to see support for the BeagleBone!

Full disclosure:  the current cgminer is very RPi specific.  Some code hacking is necessary before BB will work!

That's an understatement  Oh and it's not "the current cgminer", it's the forked code based on an older version. I received a bitfury USB stick from 101BTC and I've spent a lot of time trying to get it working on a PC instead of an RPi and it's proving far from trivial... Some of the memory and hardware hacks just don't even exist outside of the RPi and some I haven't remotely figured out what they do.

Con - I've looked at the source code in quite a detail and (I think) I have a pretty good understanding of what's going on. I was also looking at 101BTC's stick earlier but I think the chip they used may not (easily) do some of the required functions - specifically the "reset" sequence. If you want PM me and we can discuss details. I am going to (try to) convert portions of that code next week so that I can run it under Windows with another usb-to-serial chip. Basically the RPi code has a mix of bitfutry-specific stuff plus RPi-specific stuff which I'm going to split into separate files. The bitfury-specific will remain the same and I'll add an option to replace the hardware-specific part with what I need for my chip (mostly how to do "Reset" and "TxRx").

[ssi]

I'm very happy to see support for the BeagleBone!

Full disclosure:  the current cgminer is very RPi specific.  Some code hacking is necessary before BB will work!

That's an understatement  Oh and it's not "the current cgminer", it's the forked code based on an older version. I received a bitfury USB stick from 101BTC and I've spent a lot of time trying to get it working on a PC instead of an RPi and it's proving far from trivial... Some of the memory and hardware hacks just don't even exist outside of the RPi and some I haven't remotely figured out what they do.

Con - I've looked at the source code in quite a detail and (I think) I have a pretty good understanding of what's going on. I was also looking at 101BTC's stick earlier but I think the chip they used may not (easily) do some of the required functions - specifically the "reset" sequence. If you want PM me and we can discuss details. I am going to (try to) convert portions of that code next week so that I can run it under Windows with another usb-to-serial chip. Basically the RPi code has a mix of bitfutry-specific stuff plus RPi-specific stuff which I'm going to split into separate files. The bitfury-specific will remain the same and I'll add an option to replace the hardware-specific part with what I need for my chip (mostly how to do "Reset" and "TxRx").

The standard SPI stuff shouldn't be hard... it's mostly the reset sequence, which is bitbanged using GPIO.

[vs3]

Hi and thanks for your excellent feedback, the chip can be and is configured to operate using 1V8 so no level shifter should be required (one of the reasons I selected it) also it is very configurable with the utility. Just a few of the specs are;

Single-channel configurable SPI interface ❐ Master/slave up to 3 MHz ❐ Data width: 4 bits to 16 bits ❐ 256 bytes for each transmit and receive buffer ❐ Supports Motorola, TI, and National SPI modes
Configuration utility (Windows) to configure the following: ❐ Vendor ID (VID), Product ID (PID), and Product and Manufacturer descriptors ❐ UART/I2C/SPI ❐ CapSense ❐ Charger detection ❐ GPIO

The link for the full spec sheet is here; http://www.krw.net/cy7c65211.pdf

Also please keep in mind that this is simply an interface to allow a computer to provide the processing power not the UART.

Thanks again

PWRMAD - that's exactly the chip I am working with
It should be able to do all that you need for driving BF chips. (with some processing power from the PC/host).

[vs3]

(...) I was also looking at 101BTC's stick earlier but I think the chip they used may not (easily) do some of the required functions - specifically the "reset" sequence (...)

The standard SPI stuff shouldn't be hard... it's mostly the reset sequence, which is bitbanged using GPIO.

Yup - and I didn't see an easy way to switch that chip's IO to bitbanged mode (which is why I initially gave up on it .. and then saw it on 101BTC and reread the documentation and still didn't find an (easy) way ... unless I've overlooked that part of the PDF.. not to mention that this is a waaaaay overlycomplicated chip)

[ultrix]

Power circuit built and tested using load resistors.   12V ATX to 0.84V +- 5mV @90A with 92% efficiency.  Going to run it for a bit collecting data while powered by an AX1200i then will rerun it at 0.835V.  Here's part of the DSO output plotted.