nightyj
Member
Offline
Activity: 89
Merit: 10
|
|
July 01, 2013, 07:07:00 AM |
|
intron have you tried to increase hash rate with external oscillator or higher voltage or you do not have proper cooling for now.
|
|
|
|
|
|
|
Once a transaction has 6 confirmations, it is extremely unlikely that an attacker without at least 50% of the network's computation power would be able to reverse it.
|
|
|
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
|
|
|
|
cscape
|
|
July 01, 2013, 07:12:47 AM |
|
We haven't tried external oscillator. I don't have a generator that can do 200MHz+ outputs. The internal oscillator is fast enough, but it does look jittery on the scope, so internal setup/hold timing may be improved with stable external oscillator, and a higher clock may be used.
|
Happy with your c-scape product ? Consider a tip: 16X2FWVRz6UzPWsu4WjKBMJatR7UvyKzcy
|
|
|
nightyj
Member
Offline
Activity: 89
Merit: 10
|
|
July 01, 2013, 12:55:51 PM |
|
Feeding external clock with its own dedicated power supply will improove these parameters a lot and will help for more stable work at higher clock speeds if chip allows them. I have prepared one variable for selecting the best frequency and then will use fixed for maximised performance.
|
|
|
|
intron
Sr. Member
Offline
Activity: 427
Merit: 251
- electronics design|embedded software|verilog -
|
|
July 01, 2013, 01:37:19 PM |
|
Feeding external clock with its own dedicated power supply will improove these parameters a lot and will help for more stable work at higher clock speeds if chip allows them. I have prepared one variable for selecting the best frequency and then will use fixed for maximised performance.
And do you feed the clock to multiple chips then? With clock buffers or fanning tree or matched lines? intron
|
|
|
|
cscape
|
|
July 02, 2013, 08:03:42 AM |
|
The disadvantage with an external clock is that there's no practical way of tuning the clock frequency to individual ASIC performance. Given variations in process and circumstances, it may be better to optimize each ASIC's internal clock, rather than using a single external clock for all of them.
I don't know if a respin is planned, but maybe a separate VDD for the oscillator circuit would help improve stability.
|
Happy with your c-scape product ? Consider a tip: 16X2FWVRz6UzPWsu4WjKBMJatR7UvyKzcy
|
|
|
nightyj
Member
Offline
Activity: 89
Merit: 10
|
|
July 02, 2013, 08:14:25 AM |
|
Feeding external clock with its own dedicated power supply will improove these parameters a lot and will help for more stable work at higher clock speeds if chip allows them. I have prepared one variable for selecting the best frequency and then will use fixed for maximised performance.
And do you feed the clock to multiple chips then? With clock buffers or fanning tree or matched lines? intron I am talking about checking performance of one chip with external clock vs integrated, if it is cost effective depending on the hash rate increase buffers for each chip will be nice.
|
|
|
|
OrphanedGland
Member
Offline
Activity: 70
Merit: 10
|
|
July 02, 2013, 11:52:14 AM |
|
Hi All, I just received some BitFury ASIC chips today. Big thanks to BitFury for giving myself and the BitCentury team this opportunity. We are currently waiting for our initial prototype PCBs to be manufactured. I will keep you up to date with progress of the testing, but in the meantime check the following link for the pics: http://bitcentury.io/blog/bitfury-asics-have-arrived. Regards, Nigel / BitCentury
|
|
|
|
runeks
Legendary
Offline
Activity: 980
Merit: 1008
|
|
July 02, 2013, 01:02:54 PM |
|
Cool! But how can it estimate hash rate without any accepted shares? Few minutes later: intron_2 2,149.59 MH/s 174 (100.00%) 0 / 1 / 0 What's the power usage operating at this speed?
|
|
|
|
cscape
|
|
July 02, 2013, 01:21:27 PM |
|
What's the power usage operating at this speed?
The chip is running at 0.84V and 2.5A, with a clock of 230MHz.
|
Happy with your c-scape product ? Consider a tip: 16X2FWVRz6UzPWsu4WjKBMJatR7UvyKzcy
|
|
|
zulunation
|
|
July 02, 2013, 07:49:33 PM |
|
intron You were using 15 ceramic capacitors around the chip and 3 electrolytic. Could you tell the nominals?
|
|
|
|
intron
Sr. Member
Offline
Activity: 427
Merit: 251
- electronics design|embedded software|verilog -
|
|
July 02, 2013, 08:46:43 PM |
|
intron You were using 15 ceramic capacitors around the chip and 3 electrolytic. Could you tell the nominals?
The small 0402 ceramics are 1U/6V3/X5R types. The 'ecaps' are merely space holders for bigger caps to be placed there. I was in a hurry, so I chose a nice, large footprint that could hold many types of capacitors. Just put something there, say a 22U or 47U ceramic or something. intron
|
|
|
|
TheSwede75
|
|
July 02, 2013, 08:50:50 PM |
|
Bitfury is so badass! Now let's see if chip-sales are in the pipeline. I would LOVE to get my hands on some early chips for a DIY Project
|
|
|
|
dwdoc
Legendary
Offline
Activity: 966
Merit: 1000
- - -Caveat Aleo- - -
|
|
July 03, 2013, 06:15:39 AM |
|
Bitfury chip sales are not in the pipeline. Chips are designated for sale only to Metabank and 100TH as far as we know.
|
|
|
|
FlappySocks
|
|
July 03, 2013, 11:23:10 AM |
|
There is at least an intent to put them in the hands of those able to mass produce low cost miners. The danger of selling in an open market is one person will buy them all, and keep them for themselves.
|
|
|
|
kwaaak
|
|
July 05, 2013, 10:58:51 AM |
|
What's the power usage operating at this speed?
The chip is running at 0.84V and 2.5A, with a clock of 230MHz. What kind of cooling is being done at 230MHz and >2GH/s?
|
|
|
|
cscape
|
|
July 05, 2013, 11:11:44 AM |
|
230 MHz and 0.84 V requires quite aggressive cooling. The (small) test board had a small heatsink soldered on the bottom (originally for D2PAK), and a small fan aimed at it. Board temperature was about 50 deg C.
|
Happy with your c-scape product ? Consider a tip: 16X2FWVRz6UzPWsu4WjKBMJatR7UvyKzcy
|
|
|
ultrix
|
|
July 05, 2013, 12:56:19 PM |
|
230 MHz and 0.84 V requires quite aggressive cooling. The (small) test board had a small heatsink soldered on the bottom (originally for D2PAK), and a small fan aimed at it. Board temperature was about 50 deg C.
I should be getting test chips shortly and was planning on using an FPGA as an external clock generator such that the rise/fall/hold/etc can be tuned. Should be able to push > 200mhz single ended off of a virtex 6. I saw posts earlier that showed power consumption waveforms while running off of the internal oscillator. I'd be interested in seeing how the two waveforms line up in order to make better guesses at the initial parameters for said clock generator and potentially what its weaknesses are. As it turns out all of the function generators I have don't go much higher than 200mhz single ended. Shame its not differential.
|
|
|
|
eule
|
|
July 05, 2013, 01:57:08 PM |
|
2 or 3 chip USB board should be possible?
|
|
|
|
|
Andrey
|
|
July 05, 2013, 03:55:14 PM |
|
Bitfury posted that link on Russian forums, so seems pretty legit.
|
|
|
|
|