Had that problem on this specific PSU?
Turn out the problem with that one was the price. Its being discussed in page 4 and 5. RichBC eventually went with 2x 5v~ to get a range of 9v-11v. Since the one you are interested in work well, but down to 10V. So its pretty good, if you get it cheap. Others were discussing prices per unit at 230$ and 125$. The link you have is quite cheaper. So look like a good PSU. |
|
|
|
I see, gotcha. A problem someone encountered was that when he went to the lowest range, the PSU went from supporting something like 500W to 100W because the amperage limit also went down with the Volt. It doesn't seem to say what it will support while at 10V. |
|
|
|
How do you find out if you can adjust the voltage? This thing says "Primary Output Voltage (VDC) 12v". If you really want to fiddle, go ahead, you might want to have a look at which PSU the other user went with for doing this. I think he needed to get 2 5V~ PSU to do 10V~ to keep efficiency and stuff. Hmm. But it would be a shame if like other user you try to go to 11.5v and then bam it drops and you can't get any effective downvolting from it. I guess in the worse case this could be the unefficient unit and if you make sure the other unit is v1.91 then you'll have the PSU to undervolt it. |
|
|
|
Excerpt from the S5 manual:
"Notes: Input voltage should not be less than 12.00V, since it is based on serial power solution and there is no DC/DC inside the miner. Higher input voltage will cause higher mining efficiency."
What the balls? The only possible thought on why they would do that is an attempt to keep heat down. Other than that it sounds like a terrible idea in terms of reliability.
And they also had: "When better power efficiency is needed in the future due to higher network difficulties, you may want to buy some special PSUs 9V DC with more than 10A output, which will allow you to have a 0.2J/GH mining efficiency, but at lower hashing speed." So err. Yeah, don't ask. |
|
|
|
Bitcoin mining viruses is actually very hard to be profitable. Since it uses a lot of compute resources, users would feel that their computer begins to slow down hence do a reinstall or scan for viruses. There are mining viruses which only starts when a mouse movement is not detected within certain period of the time. Although this can be more effective, the sound of the fan whirring up can give it away and even if the user doesn't notice, the earnings would drop tremendously.
Thats kind of more or less the case, you can just have the program run in the background and low priority. I can't even tell that i'm mining with my GPU and CPU performance wise. I can play any game, and still mine at full speed, the program just scale back and let other processes go first. |
|
|
|
Yes. So that mean that yours will only be confirmed if there is no transaction at all. In other word, when bitcoin is death.
Naw it can get confirmed too, but it can take a long time or can happen the next block if you're lucky. And no it doesnt stay forever, after a while the transaction get rejected and the fund is returned. Thank you fof correcting me. Yes. If it is not transferable it woll be returned after a while. Transactions on the network would only be rejected if there is another transaction spending the same input getting confirmed. A correct term would be dropping out of the mempool. Depending on your client's policy, the client may continue to rebroadcast the transaction till it gets into a block and thus it would never drop out. It would be included the block once the priority is high enough, with the size and amount being constant, only the coin age is changing. Hence, the long wait for confirmations. of course, you can easily ask a miner to accept it. Eligius implements cpfp and it means that you can spend the transaction with higher fee and get it confirmed. That much is nice, i'm not familiar with Eligius but that could be a nice feature. And for the matter of wallet, i know that my Multi HD will let it return to the wallet if it get dropped. From the TX mempool indeed. Adding a fee later could be nice but, imo there should be a fee to the fee, so that lowballing the fee and then if it doesn't get confirmed you raise it later. Kind of favor spam or something. |
|
|
|
Is it possible to mine litecoin with this?
No. Its a SHA256d ASIC usb Stick. It doesn't do Scrypt. |
|
|
|
It is hard to say. There isn't a full logs of the country which miners are located in. The pools are merely based in China and people may be mining from other countries. KNCMiner also have a big farm in Sweden. From what I see, it is China having the most numbers of publicly revealed mining farms. Can this affect the network in any way?
Most of the hashrate is in china. If china were to someone block the miners from connecting to the network, that could be disastrous since we would be losing over half of the hashrate They could possibly route the pool via tor and it would be harder to block. To be honest, most miners have setup at least one backup pools and a portion of the miners aren't based in China. Possibility of losing the hashrate is still there however. I guess but even though China's pool may not be 100% China, it doesnt mean all China's miner point at a officially Chinese pool. Regardless, China probably have a similar amount of hashrate than all the northern countries combined. And it doesn't make sense for Chinese's pool, nodes and network to try to cut off. They'd have to basically turn everything they have into an alt-internet and then... then could not do anything outside of it with their BTC. They would have the total network... on basically a home lan, not connected to the world = useless BTC. I very much the pools and miners will all decide to collectively self destruct their BTC portion. |
|
|
|
it's that true, i have waiting halving coming because i believe after halving bitcoin price will increase
Just don't hold your breath too much, see what happened with the LTC halving? Sweet fuckall. xD Since its been and still tied to BTC, maybe that's why, and since BTC is well, BTC, maybe it really will have the believed effect but anyways, upward pressure might just mean +10% over 1 year, its not like the BTC price will double for miner to keep up. |
|
|
|
You'd need the virus to attack ASIC now. And to be fair, there aren't many miners that doesn't keep an eye on their throughput. I would definitively notice if something more than a few % would go away...
And its not very easy to hack and infect and propagate such a virus that would work on Linux.
Basically i'd have to say yes it would be more powerful, but no i don't see it happening.
Well Bitcoin viruses can definitely attack Linux. Yeah? Well when a virus that propagate through Windows, manage to infect Linux, bypass the router, inject into an arm processor linux, then i guess we can start getting worried, but there's no skynet virus. |
|
|
|
Was it the hash boards or controller causing the issues? Also, at this point I'm not so much going for overt efficiency gains as much as trying to drop the temp. If I can bring it down to 11.8 or so, maybe even 11.5, I could cool it off a bit.
Its the boards, the way the chips are made its a pain to even check at the chip level or find out which chip would be causing a problem, like in RichBC's situation. Basically even though the v1.91 take the downvolt well, they all react different and it take only one chip to make the whole board not downclock very well. The controller is probably pretty standard, it can even be interchanged from certain miner to another if you just flash the SSD to the proper firmware. |
|
|
|
Hash board V1.3 also.
From a low-level sense, there's no reason the chip voltage absolutely couldn't be controlled from the software except by intentional design. Most likely to prevent twits from cranking it up and blowing out their miners. What about a hardware modification? It's a switching supply of sorts to step it down to chip voltage, there's got to be a voltage reference point somewhere that can be modified, eh?
I really don't know the details. Apparently its because they switched to a "string design", the voltage must be controlled at the feed/PSU level. But again only the 1.91v seem to support it without dropping chips. Maybe RichBC can elaborate on that, but i'm not the guy for that. |
|
|
|
That's why I really want to get the voltage down on that damn supply. The 600W version of that supply has a trim pot and people have dropped it to about 11.6V. My controller board says V1.3 (assuming that's the string that reads something S5 blah FPGA something. The only version number I could find anywhere. On the control board next to where the one hash board connector and the control board power connectors are. The PSU also reads about 35W unloaded.
Can the voltage be modified like my U3's, through editing the config? Saw this in the "monitor" section.
/usr/bin/cgminer --bitmain-dev /dev/bitmain-asic --bitmain-options 115200:32:8:7:200:0782:0725 --bitmain-checkn2diff --bitmain-hwerror --version-file /usr/bin/compile_time --queue 8192 --api-listen --default-config /config/cgminer.conf
one of those "--bitmain-options" params has to be voltage, probably 0782 or 0725.
You can set the option but the hardware doesn't use it. Since it has some string design or some such it doesn't have voltage control. You can see the board version here on the PCB not the controller; http://i1079.photobucket.com/albums/w501/bond007taz/20150129_071133_zpsce3ee8cd.jpgS5_SPOWER_HashBoard_V1.3 is what it read but its quite possible you have a very early batch S5. |
|
|
|
New to this thread and TL:DR on the rest of it. GITS referred me here and told me to bring some data since I just picked up an S5 and am doing a project for the school with probably two of these things. All power measurements are made using a Kill-A-Watt meter behind an HP DPS-800GBA power supply, for some odd reason putting out a loaded 12.65V (12.7V unloaded). Hash measurements are from the management screen on the unit's web interface "GH/S(avg)" box after roughly 10 minutes. Temperatures are in a room where ambient temp was high, around 28C, using the stock (incredibly noisy) fan. Balls, won't let me actually embed the damn thing...  Ouch that is bad efficiency. But most of that is because of the PSU itself so it's hard to say how much of it is due to the voltage. Its interesting still. What PCB version do you have? Its written in top right of the PCB under the bitmain logo. Anyways basically the interesting point is 400-425, youre able to clock it just a bit higher and stay stable with your over voltage. But you get very bad efficient across the board. The highest i could get was 1315, and then it crash rapidly, while you crash a bit later at high volt. I guess it make sense. Ultimately you normally would draw 580w at 350 and you're at 700, ouch. Also your efficiency go up and down mine didn't it was a bit better in the account of the the fan draw being the same and actual efficiency being the same. |
|
|
|
Yes. So that mean that yours will only be confirmed if there is no transaction at all. In other word, when bitcoin is death.
Naw it can get confirmed too, but it can take a long time or can happen the next block if you're lucky. And no it doesnt stay forever, after a while the transaction get rejected and the fund is returned. |
|
|
|
OK thanks guys. So it's like a pool of transactions, the ones with the highest fees are put of higher priority, and my transaction will only go in when all the others above are processed?
It's not that cut and dry but mostly. Some pool or miner may decide to pick up only under average/free transaction just to help the block chain. But yeah generally. Its not a line, its not certain, but it should be good. For instance i did a 0.00001 by accident instead of 0.0001 it took days to get confirmed. 0.0001 odds are 1 block and confirmed. But as always, its never 100%. |
|
|
|
Can this affect the network in any way?
China. Well it give more power to the Chinese. It cause centralization of the hashrate. If they get even more hash% they could easily mount a 51% attack. Which can succeed with lower chance with less than 51% however. But for now we're good. You can get a good idea by looking at this graph; https://blockchain.info/pools |
|
|
|
several things:
1. Don't forget to "jump" the PSU. use paper clip (cut it or use a larger one) to get a U-shaped structure, then connect fourth and fifth connector (fourth is green, fifth is balck) in 24 pin connector (on the clip side) with this paper clip. tThe best is to insert the 'legs' of the paper clip inside the connector without damaging it-should be OK, but proceed cautiously. Some people tape it afterwards or use isolated wire instead of a paperclip, because obviously you don't want to short it by accident.
2. Insert the connector that has 8 pins (Not 6+2) into the receptacle on corsair that says 6+2 PCIe. It has to go only in one orientation (it will be obvious).
3. attach two branches of a branched cable to the 6 pin connectors (look whitish on your photo) on the same side. CX750 has two branched cables, so I cable per board, each branch engaging a connector.
BTW, i still have two CX750 and two CX500 if anybody wants them.
Thanks for the tip on jumping it. Last one I got was already done. So I need to insert two of those 6+2pin PCI cables and connect them to the same side of the miner long ways like in this picture?  Yes plug all 4 connectors, don't mix power supplies on the same board but you,re only using one so no problem. Once running for a while make sure the connectors aren't burning hot. If they are you will have to unplug the +6 and use 4 individual cable. Thats it. |
|
|
|
This guy in the video is using a peg to upvolt it, maybe you're able to downvolt it too? https://www.youtube.com/watch?v=RYOgZqfjimwNot sure if its a home made thing but you definitively should be running it closer to 12v in your case. But if you have one at 11.7 that sound pretty good too for stock speed. It should run stock speed without hashlost and still be more efficient. 11.7 was sharing my computer's PSU with it. I wish I could undervolt the HP supply to about 11.5, as I've heard people are getting decent results below 11 even. The thread we're working on, do get good results but pretty much everyone that has earlier versions don't get good results. So in effect, only with v1.91 boards people have been able to lower the Volt and get efficiency gain. Older versions were report to drop chip when you start dropping the volt significantly under 12v and there wasn't much gain to it. Please check your board version and post your results there that would be nice. https://bitcointalk.org/index.php?topic=1151460.180 |
|
|
|
|
Show Posts
