Does that mean that there are some valid nonces (~25%) that the chips just can't find, or can the S/W shift the starting nonce values to scan that space too?  First. Every core has 10 nonce bits hard coded, so 25% of the nonces won't be found. |
|
|
|
s |= rehash(op->midstate, op->m7, op->ntime, op->nbits, pn+0x02800000) ? pn + 0x02800000 : 0;
s |= rehash(op->midstate, op->m7, op->ntime, op->nbits, pn+0x02C00000) ? pn + 0x02C00000 : 0;
s |= rehash(op->midstate, op->m7, op->ntime, op->nbits, pn+0x00400000) ? pn + 0x00400000 : 0;
Ah, yes, I misunderstood what you were talking about. I've never seen those used either. Probably some superfluous code in the first iteration. In my code, I only check the other 3. Chainminer doesn't check them either. |
|
|
|
|
Note that only 756/1024 of the total nonce space is checked, the remaining part is just not implemented on the chip. Also, some chips perform worse than others. They could have some manufacturing defect, affecting part of the cores. Raising the frequency too far also increases the error rate. Unstable power regulator could have the same effect... and then there's always communication and software errors.
|
|
|
|
|
USB on the S-HASH board is just for troubleshooting and configuration. Main operation is through ethernet, directly from board to pool (and your cable/ADSL gateway of course).
|
|
|
|
But ... is the reason: a portion of the chip doesn't work properly?
i.e. the chip has some design flaw in it that means it only gets ... a bit over half of that 5GHash?
... so that Gen 2 chips would just be the fully working version of the Gen 1 chip?
Anyone got any info on that? Each chip has 756 hashing cores. I've done some tests, and seen that in most chips, nearly all of them return results. The fact that they don't get 5GH/sec is probably due to max clock being lower than expected. Though, I will also add another question: Who's idea on the single USB 'thing' was it, to have it NOT return results, until it completes the nonce range?
Does nobody learn for the mistakes of others or have any experience in doing this stuff? It's a little bit more work to return nonces as soon as they are found. The chainminer software does it, though. |
|
|
|
Cannot wait to get my boards! Is there a software option for overclocking?
No, the power regulator doesn't have a voltage adjust input. It is designed to be set using a feedback with 2 resistors, just like on the H-CARD. We thought about adding some extra circuitry around it, but we weren't sure that wouldn't interfere with the feedback loop. |
|
|
|
Oh, and have you been in touch with LUKE-JR about support on BFGMiner or do we not need this?
I thought I read somewhere these have stratum support so could potentially go straight to my selected pool. Is that correct?
The boards have their own mining software integrated in the controller chip. You set up the pool through the web page, and it will connect directly to stratum server. This means you can turn off your PC at night to save power, if you want, while the boards keep on hashing. |
|
|
|
Thanks cscape for the information. I'll be modding probably at least 150 of these boards between my setup and friends being I have a nice soldering station and hot air station so I'll try a variety of these resitance values. Do you have an idea of what 2.2k ohms and 2.7k ohms equates to approximate voltage values. It will be interesting to see how these boards perform. 2.2k results in about 0.75V. I haven't tried 2.7k yet, but I did try 3.3k, and that was too much. oh yeah did you see my question in the above post about the potential error I highlighted in orange about the R02F resistance values.
Yes, the resistor is 10k, but people are measuring it in-circuit and getting distorted values. |
|
|
|
|
Wattage isn't important, but you should pick a resistor with low temperature coefficient, to reduce the effects of a warming board causing the voltage to drift up.
Also, as well as 1.5k you should get 2.2k and maybe 2.7k.
|
|
|
|
Next time take a paint stripper to remove a damaged chip:)
If anybody decides to use a paint stripper, first cover the undamaged area around the chip with some aluminium foil to protect it from the heat. First start out slow, so the board has a chance to heat up, and then increase the heat to take it off quickly. |
|
|
|
I will probably start off by increasing the global minimum difficulty to 2 soon.
And my brief venture into the world of Eligius comes to an abrupt end. Thanks for the warning, at least. Why would the minimum difficulty make a difference ? |
|
|
|
so there is no problem in using internal fast if it works any better? Is the speed range the same 53-57?
No, there's no problem. Not sure about the speed range. I would just experiment with different values and see what works best. |
|
|
|
What does it really change using "Internal Fast" instead of "Internal Divided Slow" ?
The chips have 3 oscillator options: external, internal slow and internal fast. In addition, there's a /2 divider that can be switched on. Bitfury recommended the slow internal oscillator if it was fast enough. The /2 divider makes the clock slightly less jittery, and also allows smaller steps in clock speed. The H-cards have an option for external clock, but it's not used on the m-board. |
|
|
|
The + needs to be in the center, and the - on the outside ring. This is the most common configuration, but of course there are some exceptions, so watch out. |
|
|
|
|
Well, it was made months ago, but there were no chips for sale for large scale production. Even now, the 16 bitfury chips cost 50x as much as the ARM, so it's not such a big deal.
After the chips, the most expensive component is the power regulator module, and you'll need something like that for better overclocking of the h-cards.
|
|
|
|
Just a quick question, when you say 12v dc supply, does that mean I can power using a pc psu? and if some what connectors would I use? just bought 2 from technobit.... Thanks  The latest version has a barrel jack for power instead of the 2 pin molex in the picture. Nicest solution is a compact brick with a barrel jack. You need about 4A if you don't go wild with overclocking. A PC PSU will work too, but needs an cable adapter. |
|
|
|
Is there anyone who can confirm/deny that C05I and C04I on the M-card are supposed to be soldered together on one side?
They aren't supposed to touch each other, but it's not a problem because they are wired in parallel on the board. |
|
|
|
|
Can you post some pictures of the board ?
|
|
|
|
Each board is included with his own adjustable external clock.
So, there's one clock for all the chips ? |
|
|
|
|
Show Posts
