Bitcoin Forum
December 14, 2018, 09:16:32 AM *
News: Latest Bitcoin Core release: 0.17.0 [Torrent].
 
   Home   Help Search Login Register More  
Pages: « 1 ... 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 [56] 57 »
  Print  
Author Topic: [ANN] Bitfury is looking for alpha-testers of first chips! FREE MONEY HERE!  (Read 176121 times)
bitcminter
Newbie
*
Offline Offline

Activity: 2
Merit: 0


View Profile
March 19, 2014, 09:17:01 PM
 #1101

Hi,

Could you help ?
I have an H-board v1.2 and rpi and M-board 2.3.
It's working but after my sd card is crashing. I make another from V1 img.
And now it's not working....

for example when i running the chainminer program:

pi@bitfury /opt/bitfury/chainminer $ sudo ./miner
INIT: 16 chips detected
^Cpi@bitfury /opt/bitfury/chainminer $

i make a new chainminer install and make, but same answer Sad
i have no chip.conf and no best.cnf ... is it problem ?
Mboard 2.3 and newer can't use the v1 image due to the way the slots are connected and activated by chainminer

You might also tryout bfgminer

On the note of sdcard corruption. If you have a spare 8Gb+ flash drive laying around you can follow this guide
http://www.raspberrypi.org/forum/viewtopic.php?f=29&t=44177

Shows you how to move rootfs onto a USB drive that way the only time sdcard is touched is at boot. After that it's all from usb

Thanks for your quick answer....
I think i have enough space on my 4G SD Card

Filesystem      Size  Used Avail Use% Mounted on
rootfs          3.4G  2.0G  1.3G  61% /
/dev/root       3.4G  2.0G  1.3G  61% /
devtmpfs        235M     0  235M   0% /dev
tmpfs            49M  212K   49M   1% /run
tmpfs           5.0M     0  5.0M   0% /run/lock
tmpfs            98M  8.0K   98M   1% /run/shm
/dev/mmcblk0p1   56M   19M   38M  34% /boot

Can is ask a link for v2 img file ? If you can....
1544778992
Hero Member
*
Offline Offline

Posts: 1544778992

View Profile Personal Message (Offline)

Ignore
1544778992
Reply with quote  #2

1544778992
Report to moderator
1544778992
Hero Member
*
Offline Offline

Posts: 1544778992

View Profile Personal Message (Offline)

Ignore
1544778992
Reply with quote  #2

1544778992
Report to moderator
1544778992
Hero Member
*
Offline Offline

Posts: 1544778992

View Profile Personal Message (Offline)

Ignore
1544778992
Reply with quote  #2

1544778992
Report to moderator
100% New Software
PC, Mac, Android, & HTML5 Clients
Krill Rakeback
Low Rake
Bitcoin Poker 3.0
Bad Beat Jackpot
SwC Poker Relaunch
PLAY NOW
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1544778992
Hero Member
*
Offline Offline

Posts: 1544778992

View Profile Personal Message (Offline)

Ignore
1544778992
Reply with quote  #2

1544778992
Report to moderator
n2rovert
Newbie
*
Offline Offline

Activity: 3
Merit: 0


View Profile
March 23, 2014, 02:07:41 AM
 #1102

OK, I have waited with great patience till http://bitbet.us/bet/450/bitfurys-asic-will-work-with-power-1/ went to pending resolution status. It gets more and more interesting.

Right now we had trouble with getting chips "in time" for tests (as we expected to get chips on 30th May and do the test earlier in June). So my friend went to Taiwan to get chips, meanwhile packaging factory tries to push delivery to 13th... We'll work hard to push it back to 10th or 11th as they promised second time. But it depends... More information will be available likely on monday.

So - to speedup testing and provide plausible and trustworthy confirmation of performance I have decided that we will send sample chips to 3-4 persons of this forums with following conditions:

- We will send 10-30 chips (number of chips is that what you could select yourself depending on your custom issues - say if you sure that no problem to get 30 sample chips - we'll send 30). Preliminary specs - each chip would give roughly 5 GH/s performance.
- You will solder AS SOON AS YOU GET chip it using dead-bug style (for later chip make PCBs of course, however if you are eager to run it quicker - dead-bug might be right way) with bypass capacitors underneath and will try to run it, measure power consumption and performance with different voltage modes (sweep from 0.5 V to 0.9 V).
- You will post photos in any case - not depending whether chip works or not, it could be also interesting if you could scrub the top and post photo of its internals (if you could remove AL_RDL layer that would be great).
- You have to understand C programming language and boards like raspberrypi (if not you should team up with programmer) - but that's mandatory, and understand how to deal with quick power-hungry chip (i.e. capacitor bypass selection,  power impedance issues).
- It would be great, but not mandatory, if you have good web of trust to these persons who are running website bitbet.us: http://serajewelks.bitcoin-otc.com/trustgraph.php?source=kakobrekla&dest=mircea_popescu as they do not trust our results much, and it would be nice to give them bulletproof confirmations _IN_TIME_ for bet resolution.
- Reporting during some time how chips behave - whether they all alive and hash good or broke with time passed, and guesses why they break (maybe by sending chip to us so we could unpackage them and scan them to find out failure reason).

Additional requirements:

- Being hacker at least a bit... As we haven't enough time to write documentation, and you'll basically get C-code for raspberry that is capable to communicate, not well-tested. Some half-working miner code. Chip pinout. Maybe verilog code... It's a bonus if you could do debug - in case if chip will not work and you'll find bug - we'll likely just pay additional bonus for that. Don't expect good written documentation of any kind and don't expect that I will be able to spend many hours explaining how to make this job. Likely we'll open separate discussion thread and will discuss issues there.
- You would likely need lab power supply, if you don't have - then prepare 0.6 - 0.9 V power supply with current up to 6 amps, power supply of 1.8 V with current 100-200 mA
- raspberry-pi board or board like that (first software will be designed for it), if other board used - you should know what you do and do it yourself - SPI required.
- maybe you should have to use EXTERNAL clock generator giving 150 - 500 Mhz programmable frequency output.
- it would be great to have at hand 500 - 1000 Mhz oscilloscope, even better to have 10 Ghz oscilloscope to make clock jitter measurements, also it would be even greater to have spectrum analyzer with 4-10 Ghz bandwidth to measure power noise spectral content and debug in case if you have excessive pulsations in power network at certain frequencies by reworking capacitor bypass network to get good performance (however all of that can be skipped if we are lucky enough and have good margins in chip).
- handle chip carefully - as ESD-protection may underperform and may not give 2kV HBM protection for example. This should be met, but we don't know if it is met or not yet.

If candidates will be 3-4 or less - they will get chips, if more - I will collect in PM or in this thread requests to take opportunity with testing, make after 24 hours additional POLL topic, and those who will be elected will get chips. 48 hours after this message I hope everything will be known and settled.

Chips will be sent on 10-11th from Taiwan, if we'll find out how to help our packaging factory to make it happen, but it can be later as well. Anyway it is not only bet issue, but we would like to know about chips performance AS SOON AS POSSIBLE. I think that opportunity to get these hashes earliest would be excellent motivation to execute tests in shortest possible terms.

PS. I've read that someone here is reselling metabank.ru offers with our chips. I would like to clarify - first - neither me nor metabank.ru endorses these resellers. Second - metabank.ru is not taking 'prepayment', its just preorders with 100% refund  just to freeze funds - basically people can just show them bitcoins and keep on their OWN wallets (not moving them). Its mainly purpose is just preliminary evaluation of volumes for next production batch. With other sales, etc - that all will be discussed only AFTER testing. So my advice - don't be crazy and don't buy BS (Edited: this is what I meant - is those crazy sellers for $5k with prepayment and without guarantees, BitCentury seems to be legit and faithful).... But of course you may be lucky and that will happen as advertised - that's your win/loose.

PPS. About further distribution of chips - we have ideas (yet only ideas we'll see how that will happen in reality) - standard instrument "chip futures with real delivery with expiration date MM.YYYY", it would be traded online and actual delivery will be done by tytus (Leszek), and instead of direct contract there will be one single public contract. Then - for small to medium order amounts actual acquisition would be done to multiple locations using intermediaries who will do assembly, certification and will have end-user support for the devices. For bigger quantities it will be possible to get chips directly. I think some of delivery locations will convert "bulk" into smaller amounts for chip deliveries as well. But to grasp overall idea - we'll not multiply headache, but actually spread headache among many many heads, so each will have much less headache to pass and grow to higher volumes. Also there's some technical issues that we're discussing - how to organize it... I would like to deploy fully-featured peer-to-peer real-time exchange as open-source project and invite current web-sites to distribute tickers and handle cash-in/cash-out issues (that's not limited to these futures, but are open for many participating parties as well), but being objected that it's quite complex and not trivial to do in limited time. I hope that if we have enough of luck and chip will work, I'll have enough time to make this actually happen.

SO PLEASE, DON'T BOMBARD ME WITH REQUESTS TO BUY CHIPS, ETC. JUST BE PATIENT.
can you sell a kit that we have to assemble and solder smc's etc. ourselves for a price cut?
adib
Sr. Member
****
Offline Offline

Activity: 369
Merit: 250


View Profile
April 01, 2014, 03:50:27 PM
 #1103



H-Card to USB, testing with one chip and adding one by one to see when it refuses to go... hopefully MCP2210 will be able to scan 16 chips.
CGminer disables it after a minute or so... maybe its my busted usb on the laptop.
BFGminer works on another laptop.

Lets see what happends.

UPDATE 1:
3 chips but can anyone tell me why the 50% frequency drops ocur?

Hashing at about 7 ghz

TIPS - 19JLxDkCfn5x667xCeSgmYNop4WLR3ci27
vs3
Hero Member
*****
Offline Offline

Activity: 622
Merit: 500


View Profile WWW
April 01, 2014, 06:45:04 PM
 #1104



H-Card to USB, testing with one chip and adding one by one to see when it refuses to go... hopefully MCP2210 will be able to scan 16 chips.
CGminer disables it after a minute or so... maybe its my busted usb on the laptop.
BFGminer works on another laptop.

Lets see what happends.

UPDATE 1:
3 chips but can anyone tell me why the 50% frequency drops ocur?

Hashing at about 7 ghz

Seems like it recognizes 1 chip only ... which makes me think that your bfgminer version is from before the multi-chip support was added.

Try the development version of bfgminer (a.k.a. "head" or "trunk") - I think Luke added the multi-chip support after 3.10 was released.
Download from luke.dashjr.org/tmp/code/webisect/webisect.php?dobuild=bfgminer

pinkfloyd11
Newbie
*
Offline Offline

Activity: 56
Merit: 0


View Profile
May 16, 2014, 02:08:27 PM
 #1105

hello
I'm testing my device based on vladimir project.

I read this:

C:\test>nf1_init test 0000144196
nf1_init.exe [command] [setting]
Commands:
test [device serial number] - runs bitfury tests to that device
fixID [ignored] - Tests for NanoFury devices and fixes Product String

test mode specified for device 0.

Testing device 1 : SERIAL=0000144196,  Path=[\\?\hid#vid_04d8&pid_00de#6&2d1942d
0&0&0000#{4d1e55b2-f16f-11cf-88cb-001111000030}]
DEVICE FOUND! Proceeding with testing... Press ESC to cancel testing at any time


testing [16:6:38]....


How long is nf1_init test?
What should happen for understand that device mining correctly?

Thans
vs3
Hero Member
*****
Offline Offline

Activity: 622
Merit: 500


View Profile WWW
May 16, 2014, 10:00:29 PM
 #1106

How long is nf1_init test?
What should happen for understand that device mining correctly?

If device is working fine you should start seeing results almost every second.

If you see no results it likely means that your chip isn't hashing (it could be communication issues or voltages for example).

This is how the output should look like:



pinkfloyd11
Newbie
*
Offline Offline

Activity: 56
Merit: 0


View Profile
May 17, 2014, 05:56:40 PM
 #1107

Dear Vladimir

tanks for image!

Since today I read this:

C:\test1>nf1_init test 0000144196 ->[enter]

test mode specified for device 0.

Testing device 1 : SERIAL=0000144196,  Path=[\\?\hid#vid_04d8&pid_00de#6&2d1942d
0&0&0000#{4d1e55b2-f16f-11cf-88cb-001111000030}]


If I set SPIDEBUGTXRX the communication always go on so I can check signal with scope?

Or I need logic analizer?

Thanks again



pinkfloyd11
Newbie
*
Offline Offline

Activity: 56
Merit: 0


View Profile
May 18, 2014, 09:58:13 AM
 #1108

Hello word

today I'm starting to check electric signal with scope. For now the only one measured is out clock in (DC MODE) :

http://i.imgur.com/9QJhHTdl.png

I note that it don't start from 0volt but from 0.4 volt.

About nf1_init it report this error: "Error in SPITransferSetting [248]

Any suggest is apprecciate


Thanks

pinkfloyd11
Newbie
*
Offline Offline

Activity: 56
Merit: 0


View Profile
May 18, 2014, 02:07:32 PM
 #1109

When I use RS232 I connect both RX and TX pin together (microprocessor side). In this manner I can check if chip like max232 work correcty simply receive echo on terminal program like putty.
Which is the equivalent test in this case?
Thanks
KNK
Hero Member
*****
Online Online

Activity: 685
Merit: 501


View Profile
May 19, 2014, 07:11:41 AM
 #1110

When I use RS232 I connect both RX and TX pin together (microprocessor side). In this manner I can check if chip like max232 work correcty simply receive echo on terminal program like putty.
Which is the equivalent test in this case?
Thanks
Connecting MISO to MOSI

Mega Crypto Polis - www.MegaCryptoPolis.com
BTC tips: 1KNK1akhpethhtcyhKTF2d3PWTQDUWUzHE
pinkfloyd11
Newbie
*
Offline Offline

Activity: 56
Merit: 0


View Profile
May 19, 2014, 07:56:41 AM
 #1111

which is program equivalent to putty?
vs3
Hero Member
*****
Offline Offline

Activity: 622
Merit: 500


View Profile WWW
May 19, 2014, 08:13:49 AM
 #1112

When I use RS232 I connect both RX and TX pin together (microprocessor side). In this manner I can check if chip like max232 work correcty simply receive echo on terminal program like putty.
Which is the equivalent test in this case?
Thanks
Connecting MISO to MOSI

Nope, that's not going to work.

You need a strong (very low impedance) pull down to GND, otherwise the MISO pin will stay up in the air.

You'll basically have two 2k resistors pulling down (R8 and R6), and one (R4) pulling up to 3.3V. So your "0" will be 1.1V.
And that would be an ideal case - if you get the transistor to open. But since it won't open at that low voltage (it needs 1.5V+) ... you won't get anything.

Edit: It WILL work if you connect MOSI/MISO on the MCP2210 pins (before all resistors/transistor).
It will NOT work if you do that on the bitfury side (after resistors/transistor).

vs3
Hero Member
*****
Offline Offline

Activity: 622
Merit: 500


View Profile WWW
May 19, 2014, 08:16:51 AM
 #1113

which is program equivalent to putty?

There is no such.

But you can easily adapt the nf1init to do a test like that for you.


By the way - I suspect something being off with your oscilloscope image - the "0" shouldn't be 0.5V
Either you have some kind of a short somewhere that is causing it, or it might be the oscilloscope itself interfering with the schematic.

pinkfloyd11
Newbie
*
Offline Offline

Activity: 56
Merit: 0


View Profile
May 19, 2014, 10:44:42 AM
 #1114

Hi

http://i.imgur.com/uusRYySl.png

then follow above schematics label if i remove R5 R8 R4 and Q1

and short pin 9 and 13 of MCP2210 I should test ONLY SPI COMMUNICATION BETWEEN PC AND MCP2210.

Question: GP5 is an input? What is it pourpose? I need to remove R7 also?

Thanks


UPDATE1:

At this URL:

http://www.microchip.com/wwwproducts/Devices.aspx?product=MCP2210#developmentTools

under "Documentation and Software" there is SPI TERMINAL that should work after removed the above components and create a short between MOSI and MISO. Correct?
vs3
Hero Member
*****
Offline Offline

Activity: 622
Merit: 500


View Profile WWW
May 20, 2014, 06:44:36 AM
 #1115

Hi



then follow above schematics label if i remove R5 R8 R4 and Q1

and short pin 9 and 13 of MCP2210 I should test ONLY SPI COMMUNICATION BETWEEN PC AND MCP2210.

Question: GP5 is an input? What is it pourpose? I need to remove R7 also?

Thanks


UPDATE1:

At this URL:

http://www.microchip.com/wwwproducts/Devices.aspx?product=MCP2210#developmentTools

under "Documentation and Software" there is SPI TERMINAL that should work after removed the above components and create a short between MOSI and MISO. Correct?

Yup - even if you leave the resistors it should still work if you short the MCP2210 pins. Those resistors would be pretty weak and the MCP can easily overpower them for your purposes.

GP5 is being used as both input and output. That's the SCK override (so that you can send the reset sequence to bitfury chips - you can read some more about how it works on the github wiki page).

As for that terminal program - I've never tried it so I can't really tell for sure. It seems like it should work though.

pinkfloyd11
Newbie
*
Offline Offline

Activity: 56
Merit: 0


View Profile
June 30, 2014, 01:59:59 PM
 #1116

Hello

are many days that http://www.bitfurystrikesback.com/ is offline.

Where I can acquire bitfury II generation chips?

Thanks
pinkfloyd11
Newbie
*
Offline Offline

Activity: 56
Merit: 0


View Profile
July 02, 2014, 05:45:05 PM
 #1117

Hi at all
I notice that when bitfury chip start, there is great hardware error.

After few second it slowly begin to go down.

Do you think that is better prolong soft start time of Vcore?

If not, how i can reduce it?

Thanks
pinkfloyd11
Newbie
*
Offline Offline

Activity: 56
Merit: 0


View Profile
July 03, 2014, 03:01:43 PM
 #1118

Hi at all

I notice that when bitfury rev2 start I had great HW Error.
It slowly go down in a few minutes.

How can I avoid this?

1)Use big capacitor on power supply?
2)Use soft start at power on?

Now that I had end test I need to acquire rev2 chip but main sell site is down.
Can you help me?

Thanks
Tyger
Full Member
***
Offline Offline

Activity: 139
Merit: 100



View Profile
July 04, 2014, 08:57:08 AM
 #1119

The rev2 requires good cooling in order to work properly
Taugeran
Hero Member
*****
Offline Offline

Activity: 658
Merit: 500


CCNA: There i fixed the internet.


View Profile
July 04, 2014, 05:12:40 PM
 #1120

The rev2 requires good cooling in order to work properly

How do you define properly and good cooling?

I have a Rev.2 chip nanofury running at 3.65G with no active cooling and the stock heatsink Tongue

Hand soldered btw

Bitfury HW & Habañero : 1.625Th/s
tips/Donations: 1NoS89H3Mr6U5CmP4VwWzU2318JEMxHL1
Come join Coinbase
Pages: « 1 ... 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 [56] 57 »
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!