Those are crap chips
They should be in a $1 disposal sale
better still in the10 cent disposal bin
Topic: [BFL-Chip Groupbuy] $27,5/chip 50% - BTC,PayPal,SEPA - 420 sold, 0/100 batch #4 - page 12. (Read 41816 times)
August 08, 2013, 03:12:52 AM
August 08, 2013, 03:02:23 AM
Lucko I think you have been locking it 
Yeah pgp urls broke
Yes I also think that is what happened... I was sleeping on(keyboard) my laptop... I need to slow down now... Burning candle on both ends is leaving its mark...Yeah pgp urls broke
If anyone is interested how good samples are...
PROCESSOR 0: 10 engines @ 428 MHz -- MAP: B7E4 1011011111100100
PROCESSOR 3: 7 engines @ 354 MHz -- MAP: F206 1111001000000110
PROCESSOR 4: 12 engines @ 384 MHz -- MAP: F6DE 1111011011011110
PROCESSOR 5: 8 engines @ 342 MHz -- MAP: 85DC 1000010111011100
PROCESSOR 7: 9 engines @ 371 MHz -- MAP: 03FE 0000001111111110
This is with tweaked 1.2.5 firmware.... We are still in process of making ours... This is also the best test board can do. BFL power usage data are off a bit. Could be poor quality chips. But just to be sure we will use 4 chips per power module so it will work for sure...
August 07, 2013, 08:11:07 PM
Lucko I think you have been locking it
Yeah pgp urls broke
I don't think the other thread is locked now. I believe the links are broken because the forum now has issues with shortened links. Yeah pgp urls broke
August 07, 2013, 07:52:22 PM
Lucko I think you have been locking it
Yeah pgp urls broke
Yeah pgp urls broke
August 07, 2013, 03:05:34 PM
Any more news on the 16 chip design?
Yes. It is more or less ready and will be made in about 14 days for first test... Based on what we know now we are sure it will work. So it needs only samples now...
And it has a name COINTAMINATION
I must say I love it!!!EDIT: GB buyers that have chips but not enough to fill full 16 chips board can buy more chips now. Board will be hosted for free(-electricity) till chips arrive and then we will add them to the board.
August 07, 2013, 12:35:23 PM
Bin like recycle bin... Looks like samples are chips that they will not even put in Jala...
They seem to be pretty terrible. I've had to use a knife to scrape the epoxy underfill off the top of the die on some. August 07, 2013, 10:41:50 AM
First we need to figure out how much heat is coming off chips... It looks like BFL specs are off... 3 chips 38W... So not by much but they have low hashrates... Only 8GH... And it is not the board it is chips... Samples are low quality...
So then we will know heatsink needed and then we can make a sticker...
So then we will know heatsink needed and then we can make a sticker...
August 07, 2013, 06:53:35 AM
If anyone's buying chip credits, I still have a ton...
August 07, 2013, 03:06:46 AM
Hi Lucko, you need a COINADO sticker on the cooling fan for the units you fully assemble.
Any more news on the 16 chip design?
I can make the stickers if anybody need high quality ones, I can too print white and metallic on transaprent vinyl all with contour cut - Lucko has my contact details now too (sample chips). High quality printings need a vector based logo work what have to be in one layer (no overlaying vectors). So just contact me!
August 07, 2013, 02:50:42 AM
Hi Lucko, you need a COINADO sticker on the cooling fan for the units you fully assemble.
Any more news on the 16 chip design?
August 07, 2013, 01:21:47 AM
Hi,
I send my 3 samples for the bigger board on Monday with registered mail to you - should receive soon
I send my 3 samples for the bigger board on Monday with registered mail to you - should receive soon
August 07, 2013, 01:06:09 AM
Bin like recycle bin... Looks like samples are chips that they will not even put in Jala...
August 07, 2013, 12:40:01 AM
Yes but it looks like samples are bin chips...
As in substandard shit? I'm not a tech guy. Just trying to interpret the lingo.
August 07, 2013, 12:28:13 AM
Yes but it looks like samples are bin chips...
August 07, 2013, 12:21:42 AM
I think so yes... If I'm not mistaken theoretical max for that chip is 2.5...
If only 10 engine is not do to power supply we are in 200 to 300 $ range for sure... Maybe less with enough orders... But we will need some more time to look at findings to be sure... And test with more chips...
EDIT: It looks like chips are the problem...
If only 10 engine is not do to power supply we are in 200 to 300 $ range for sure... Maybe less with enough orders... But we will need some more time to look at findings to be sure... And test with more chips...
EDIT: It looks like chips are the problem...
Are you referring to the maximum for a 10 engine chip? I thought the typical BFL chips would have more power.
August 07, 2013, 12:05:24 AM
For unknown reason topic https://bitcointalksearch.org/topic/bfl-board-project-cointamination-eu-facility-orders-opened-for-chili-267649 was locked... Or someone with power hate that we took the lead... So for now I'm waiting to get answer why but it looks like they have changed some polices (see warning next to my PGP key signature)
August 06, 2013, 08:11:41 PM
If it's 10 engines running, i will consider that to be a faulty chip.
According to BFL website, a grade D chip will have at least 12 engines running. https://products.butterflylabs.com/homepage-subproducts/65nm-asic-bitcoin-mining-chip.html
But this is a sample chip and i guess we can't complain much.
It still shows how professional are they to send chips that are below grade D....
For my jallyes, I have tried restarting it until i got at least 15 engines running on both chips. It works for me where the engine kinda fluctuate between 13 to 15 engines. Maybe you can try that out.
According to BFL website, a grade D chip will have at least 12 engines running. https://products.butterflylabs.com/homepage-subproducts/65nm-asic-bitcoin-mining-chip.html
But this is a sample chip and i guess we can't complain much.
It still shows how professional are they to send chips that are below grade D....
For my jallyes, I have tried restarting it until i got at least 15 engines running on both chips. It works for me where the engine kinda fluctuate between 13 to 15 engines. Maybe you can try that out.
August 06, 2013, 12:38:14 PM
I think so yes... If I'm not mistaken theoretical max for that chip is 2.5...
If only 10 engine is not do to power supply we are in 200 to 300 $ range for sure... Maybe less with enough orders... But we will need some more time to look at findings to be sure... And test with more chips...
EDIT: It looks like chips are the problem...
If only 10 engine is not do to power supply we are in 200 to 300 $ range for sure... Maybe less with enough orders... But we will need some more time to look at findings to be sure... And test with more chips...
EDIT: It looks like chips are the problem...
August 06, 2013, 11:58:24 AM
2.3GH/s is good enough for a chip though it's only 10 engine running. what are the estimated price for a board to have 8 mounted chips?
Kudos to the first working board
Kudos to the first working board
August 06, 2013, 10:31:41 AM
Just to let you know...
I believe we are the first that have a hashing board
Busy day today...
Got it managed to get the board mining.
This first chip has not all internal engines available, so it is slower than normal. But it seems some of them will re-activate when increasing core-voltage.
I have to experiment with that later...
https://www.youtube.com/watch?v=nZIc03dKLIs
Scope: SPI decoding
Left meter: Chip-current in 100 mV per Ampere
Right meter: Core voltage
Got it managed to get the board mining.
This first chip has not all internal engines available, so it is slower than normal. But it seems some of them will re-activate when increasing core-voltage.
I have to experiment with that later...
https://www.youtube.com/watch?v=nZIc03dKLIs
Scope: SPI decoding
Left meter: Chip-current in 100 mV per Ampere
Right meter: Core voltage
I believe we are the first that have a hashing board
Jump to: