|
You can't really blame board manufacturers for trying to make a profit. It's called capitalism.
If you want to have an inexpensive FPGA farm, design your own PCB and have the boards manufactured and assembled in China.
You don't even have to come up with your own bitstream - you can take one of the existing ones...
|
|
|
|
|
Order date: 1/24/12 (1) and 1/27/12 (2)
Received: 4/17/12 (3)
No shipment notification was received.
|
|
|
|
Do you happen to be running your Single with cgminer? If so, can you report the temperature it is reading? Some of the screenshots for BFL's upcoming 'EasyMiner' show temperatures in the 50's and 60's (Celsius) but who knows what the ambient temperatures were. I've heard elsewhere that throttling (front LED blinking) occurs around 75C so I'm interested in knowing how close your Single is to that point.
No, I'm running Ufasoft miner. Is there a BFL-compatible binary of CGMiner available? All I found was old (non-BFL-compatible) binaries and source code files. I don't want to install Visual Studio (which version, anyway?) just to build CGMiner... The Windows binaries have BFL support already compiled in. I'm assuming you are looking for the Windows version since you mention Visual Studio. Cgminer 2.3.2 works for sure; the current 2.3.3 should be fine as well. OK, got Cgminer 2.3.3 now and its temperature display reads 64 degrees Celsius. The room is well into the 90s Fahrenheit, due to three GPU rigs (with a total of 12 GPUs) running as well. I will turn one of them off once I receive more Singles. |
|
|
|
Do you happen to be running your Single with cgminer? If so, can you report the temperature it is reading? Some of the screenshots for BFL's upcoming 'EasyMiner' show temperatures in the 50's and 60's (Celsius) but who knows what the ambient temperatures were. I've heard elsewhere that throttling (front LED blinking) occurs around 75C so I'm interested in knowing how close your Single is to that point.
No, I'm running Ufasoft miner. Is there a BFL-compatible binary of CGMiner available? All I found was old (non-BFL-compatible) binaries and source code files. I don't want to install Visual Studio (which version, anyway?) just to build CGMiner... |
|
|
|
Can confirm receipt of BFL single ordered via paypal on 1/26, delivered and fired up 4/14
Wow. Either they forgot my 1/24 order, or they are combining it with my 1/27 order. I'm hoping it's the latter... |
|
|
|
No shipments this week ?  Maybe they are waiting for the next batch of PCBs from China. Who knows. All I know is that I'm at the front of the line, or almost at the front of the line, for the one I ordered on 1/24, and as soon as I get the shipping confirmation, I'll post it here. Meanwhile, the rev. 3 Single I got last week is mining along happily in my 90 degree "mining office", without the front LED blinking. |
|
|
|
Looks like they've almost made it through the January orders...hopefully things are starting to pick up pace  Some or all Jan. 23rd orders have shipped, but at least SOME Jan. 24th orders (including the one for my second Single) have not shipped yet. Thus, they are about a week shy of shipping all January orders. |
|
|
|
Problem with any of their product is the noise from the fan means you can't keep it in your bedroom. Water cooling could solve that.
Yes, and then you could attach it to your waterbed, gradually warming up the waterbed in the process. Sorry, couldn't resist...  |
|
|
|
While the ztex boards don't need to be cheaper they do need to close that cost gap. The quad @ $800 in bulk (and $1000 individually) would have been impressive. Between the $50-$100 annual power savings and the ~$80/mo opportunity cost* it would have been a solid competitor. At $950 it is hard to make the math work even under most optimistic scenarios. They could close that gap by making the Bitstream faster  Or isn't there some improvement space left? Yes, there is: EldenTyrell's bitstream which squeezes three regular SHA algorithms (not double ones as the Bitcoin mining algorithm requires) into a Spartan6-150. Assuming he ever sells it at a reasonable price. But it also seems to have a higher power draw, and Stefan's 8 Amp VCCINT buck controller is not exactly helping in that department. Elden, if you're reading this: What's up? Stefan, if you're reading this: Have you thought about switching to 10 Amp buck controllers? |
|
|
|
p_shep - thanks for the Rev3 pics!
gigavps - are the Rev 3 (assuming Rev4 is a typo) running any cooler? Can't tell what is what on the mgpumon display...
lol. I have no idea which is which either! I was told by BFL that the temps displayed can vary quite a bit. The most important thing is for the singles to not start flashing their front LED at the ambient temp while mining. I have it running in my 90 degree (est.) mining office and it doesn't seem to throttle. Excellent job, BFL. |
|
|
|
It's that time again. Moved some things around a bit.  11 x Rev.2 and (in the back) 4 x Rev. 3, if I count correctly? |
|
|
|
Actually, they're pretty noisy. Quite a bit louder than I was expecting. They gotta hefty Sunon fan in there.
Probably depends on which revision you have. I just received Rev.3 and it's not noisy at all - yes, you can hear it, but you can have a quiet conversation next to it. |
|
|
|
I received my single today, sender's address handwritten on the box - looks like a mom&pop operation, thus I'd say: Not a chance, Inaba. Being a great guy has nothing to do with it. Not a chance.
Was your Single the Rev3 style (1 fan on top, no fan on the bottom), or the older 2-fan design? My understanding is that BFL has been shipping the Rev3 units already but I'm not aware of any deliveries of these being reported. It's a rev3, it has no fan at the bottom, however it does have several (not very high and not very large, graphics-card style) fanless blue heatsinks at the bottom of the PCB. I did not open it - it's mining along happily as we speak. |
|
|
|
Inaba gets $600 if the single ships in less than 6 weeks. D&T gets $600 if it doesn't.
watching *popcorn* Likewise. Ironically, by reviewing the first Single, Inaba is likely to have caused a flurry of orders, which will almost certainly sink his wager.  I received my single today, sender's address handwritten on the box - looks like a mom&pop operation, thus I'd say: Not a chance, Inaba. Being a great guy has nothing to do with it. Not a chance. |
|
|
|
Interesting, 4 hours after I received a shipping notice, I got an email with the subject "Your parcel is expecting of receiving", and a virus attached)
How come, spammers are hooked directly to PayPal/USPS systems?
No, the sender's address is just forged. I can send you an email that looks like it originated from the White House, Mr. Barack Obama himself. |
|
|
|
Shipped: 04/11/20012
That's 18,000 years from now. I realized that there would be some shipping delay, but 18,000 years - that's just cruel, man. |
|
|
|
|
Nah, I cannot afford funding a Hardcopy device, and even if I barely could, I wouldn't invest all of my money into a Bitcoin miner. So many things can go wrong - difficulty can explode, the exchange rate can plummet, MtGox could be shut down by the Japanese government, etc. etc.
The LargeCoin folks up in Vancouver seem to be doing something like that, however.
Maybe even full-custom ASICs, judging from the low estimated power draw of their 20 GH/s box.
|
|
|
|
makomk achieved 27.7MH/s from CycloneIV 22k part. My quess is that is 220MHz core rolled 8 times. Fully unrolled core fits to 75k Cyclone. That gives two cores on 150k part and propably at 300MHz (28nm vs. 60nm), so 600MH/s may be possible...
110 MHz at 4 clock cycles per hash, actually. The design scales down reasonably well to smaller devices. Thanks for the reference. Unfortunately, the URL to Makomk's code in the message you referenced does not exist, though perhaps his code is reflected by the DE2-115-makomk-mod branch of fpgaminer's code. I just compiled that code with LOOP_LOG2 set to 0 and found that it compiles to 77,724 LEs/Fmax=109.84MHZ with the provided project settings, so probably 75K LEs can be achieved by optimizing for density (though this would reduce Fmax). That's slightly older code. It's probably better than the newer versions with LOOP_LOG2=0 but it gives invalid results if you change it to anything else. So, can you make the latest code available somewhere? I'm trying to instantiate two fully unrolled instances of your "slightly older code" on the Cyclone V GX 7 target architecture - so far, unsuccessfully. Once instance is placed and routed just fine, and achieves 140 MH/s in the conservative "slow" simulation and a whopping 250 MH/s in the optimistic "fast" simulation. |
|
|
|
As far as I understand, loops that are not 100% unrolled incur inefficiencies, as partial results have to be fed back in and there has to be logic (multiplexers) to do that, whereas data in a fully unrolled design just percolates from start to finish.
Interesting. So if you can fit one fully unrolled instance in a device so you can get 1 hash/clock and fit another half unrolled instance that gets 1 hash/2 clocks, the second one holds back the speed of the first one. Would it be possible to clock the first at a speed a little faster than the second one? Or would this give difficulties to combine the 2 parts to have 1 output? Yes, you can, and that would be a fallback strategy for the Cyclone V GX 7 in case one cannot fit two unrolled double-SHAs, however it'll hurt the $ per MH/s number. |
|
|
|
so basically what would be the... rough estimation of performance in MH/s for a chip if you can get the 2 instances fully loaded into it?
Impossible to say at this point, because 1) two instances don't even seem to fit 2) the difference between the slow estimate at 140 MH/s/instance and the fast estimate at 250 MH/s/instance is just too great. The maximum seems to be 500 MH/s for two instances, but that's subject to too many assumptions to be realistic. But then again, maybe wondermine comes up with a better implementation than fpgaminer. |
|
|
|
|
Show Posts
