And it's followed up by a 22 minute block. Whoever this monster pool hopper is please sod off. And if my conspiracy theory is right please also sod off because it's not funny.
... The average blocks per day for the last few months are as follows: - 4.6 in October - (est. average pool hash rate = 7.8PH) 5.1 in November (est. average pool hash rate = 8.6PH) 5.2 in December (est. average pool hash rate = 10.3PH) 4.0 in January (est. average pool hash rate < 10.3PH) 5.2 so far for February (est. average pool hash rate < 9.3PH) Looking at January as the worst month in recent times, we see that it started the month at 11.6PH and remained above 10PH until the 15th Jan (finding 65 valid blocks in 14.5 days). For the remaining 16.5 days of the month, the pool never went over 10PH (and was on average closer to 9PH) and found 57 blocks. I personally would therefore call it luck (both good and bad) rather than being a bug in the code. I'd rather look at it on the face of it rather than compartmentalise this based on months .... aside from the difficulty mined in certain periods, there is nothing to make a month's period any special to the bitcoin network. So here goes ... I was only able to get data for the period 2015-01-08 18:50:01 to 2015-02-18 13:15:21 (i.e the last block) and we have: 53 Blocks where the Pool Scoring Hash Rate was below 9 Ph/s 66 Blocks where the Pool Scoring Hash Rate was above 9 Ph/s BUT below 9.5 Ph/s 23 Blocks where the Pool Scoring Hash Rate was above 9.5 BUT less than 10 Ph/sPh/s 23 Blocks where the Pool Scoring Hash Rate was above 10 Ph/s So there .... there may be some mileage in the conspiracy theory but we do not go without! |
|
|
|
If I had the option to buy (used from a truster source) 3 S3 and 1 S3+ for $500 as a package complete w/2 psu's and a router or put together an S5, what would the better option be? They were never overclocked. Pretty excited to discover mining and excited to get started.
any other links or advice would be appreciated. Thanks!
It looks like a good deal for the S3's as at the very minimum you'll get 2Th/s whereas the S5 is, at a push, 1.3Th/s. Power consumption wise, you can not beat the S5, so if your electricity costs are high its a no brainer, S5. The other considerations include the space .... you'll need at least 5 times the space for the S3's than the S5 AND the noise as you'll get more noise from a single stock S5 than you will from your 4 S3's put together, and that is not exaggerating. |
|
|
|
Thanks for the info, but I'm past 90 days on it.
If powering the rig off for a few minutes results in the same result, then you may have to browse to the Miner Configuration -> Advanced Settings tab and if the voltage is set to 0000, change it to 0725 if you are running at 250 or less frequency then press Save & Apply. When you re-check the Miner Status page, it will take a few moments before the chips reset and show o's, they may eventually turn to x's though. The trick is to find a stable voltage setting at the frequency you are running at that results in less HW errors, so you may have to increment the voltage. It is an involved process but will surely yield results if you have the patience. |
|
|
|
As an aside it's a bad idea to use plastic that doesn't meet 94V- standards like you are since if you have an over heating incident you are more likely to have a serious fire instead of just some melted plastic.
Not sure if that was in specific reference to the images I posted, but if it is, I can not think how you came to the conclusion as to the standards it meets / complies with. I did (after your post) look up the details on the perspex material I am using and I have no concerns whatsoever in that regard. I mean using it in the manner you are sandwiching components that are going to get warm if they are over loaded. Tot that the particular plastic you are using is or isn't 94V-0 or better rated. 94V is a fire retardant spec for those following along. Thanks for clarifying and I hope you won't take it personal to learn that none of your assertions bear out. 1. There are NO components being sandwiched. The "plastic" simply holds the connector (to ease unplugging) and the PSU switch. 2. The "plastic" is not voltage rated but it is fire rated, and I can confirm to you that it is fire retardent with a high enough combustive temperature. PS. Any pointers on the 750W PSU connector ...? 94V has nothing to do with voltage ratings. It is a UL flammability standard. What ever you want to call it it is in contact with pieces that are carrying high current and can get hot but if you've checked the specs on your material and you are satisfied then good. Don't have the exact connector p/n off hand but I've used 51915-176LF and then hack it down to the correct size with a saw and glue it back together with some epoxy. http://en.wikipedia.org/wiki/UL_94I suppose there are different standards' names applied for different territories (which can be compliant with internationaal standards), but seeing it is a flammability standard, the "plastic" I use is sourced in the EU and complies with / achieves UK fire / flammabilty levels ... aka high combustive temps. On the connector, I'll pass on the saw hacking bit and wait for the PSU .... introducing foreign material in the shape of epoxy resin to the composite connector may compromise its flammability capability ..... if they make the receptacle, they certainly make the connector! |
|
|
|
As an aside it's a bad idea to use plastic that doesn't meet 94V- standards like you are since if you have an over heating incident you are more likely to have a serious fire instead of just some melted plastic.
Not sure if that was in specific reference to the images I posted, but if it is, I can not think how you came to the conclusion as to the standards it meets / complies with. I did (after your post) look up the details on the perspex material I am using and I have no concerns whatsoever in that regard. I mean using it in the manner you are sandwiching components that are going to get warm if they are over loaded. Tot that the particular plastic you are using is or isn't 94V-0 or better rated. 94V is a fire retardant spec for those following along. Thanks for clarifying and I hope you won't take it personal to learn that none of your assertions bear out. 1. There are NO components being sandwiched. The "plastic" simply holds the connector (to ease unplugging) and the PSU switch. 2. The "plastic" is not voltage rated but it is fire rated, and I can confirm to you that it is fire retardent with a high enough combustive temperature. PS. Any pointers on the 750W PSU connector ...? |
|
|
|
As an aside it's a bad idea to use plastic that doesn't meet 94V- standards like you are since if you have an over heating incident you are more likely to have a serious fire instead of just some melted plastic.
Not sure if that was in specific reference to the images I posted, but if it is, I can not think how you came to the conclusion as to the standards it meets / complies with. I did (after your post) look up the details on the perspex material I am using and I have no concerns whatsoever in that regard. |
|
|
|
The issue being, properly soldering a 6 awg wire to a plastic plug...
Same way you were going to solder the 1mm terminal blocks .... only this time, since 6 (or 8 ) AWG wires are bigger, easier! Just pre-tin them, put them in place and away you go. |
|
|
|
Well, I was planning to mount the PSUs a meter or two from the miner's, so having less wirer would be best.
I was thinking of running 8AWG from the psu to beside the miner than breaking out with smaller terminal boards. .....
And what is stopping you from doing that now? That you got the 32A terminal blocks does not mean you have to use them on the connector! I know it looks neat, but it is not the only practial way to achieve what you want. Of-course, nearly everyone's setup is different, but if I were in your position, all I'd do was to solder the 8AWG wires onto the connectors and use the terminal blocks on the other end of the 8AWG wires to break out to the rigs ..... or am I over simplifying this one ? ! PS. I hope those 8AWG's are copper wires ..... and you've factored in the 1-2 meter resistance they introduce to the circuit, thus the output the other end. |
|
|
|
Well by imagination your mean math..
I'm expecting to see 2000-2400 off my psu if I max it out
That works out to be 200 amps on the high end.
200/32 = 6.25 so let's say 6.
I only want to run to four lines.
for 50 amps max per line.
Even that with an est. 3% loss would have me using 6 AWG wire... Which is nuts.
That runs to a terminal than breaks out into dual PCI connectors.
If I'm unable to find a 50amp terminal for the back of the plug, I need to somehow split the wires into either more wire' at an uneven/unblanaced number or overload them little 32 ampers.
I'd rather not have 12 terminals on the back of the plug, 4 and 4 would be as ideal as they come.
Finding a 50 terminal I can solder to the back of the plug, is my next issue!
Short of finding your prefered 50A rated terminal block, I can not see another solution aside from side mounting the 32A terminal blocks onto your connector, which'd make each power connector capable of serving 64A (as opposed to the max out put of 50A). In my estimation, you'd then have a total of 16 terminals on each connector (or 12 if you prefer it that way). For bitcoin mining, I can not find any reason to make it more complicated than that, but of course, you may have other use cases I am not privy to. I am assuming here that you'd like to do this on the basis of cost reduction, if not, then you have the breakout board option, and with that you do have options, literally! |
|
|
|
I can tell by looking that it won't work on the 750W, at least not the ones I carry.
What's the part number for the Dell 750W PSU's connector if you care to share? I'd like to build them for the dps 2000bb but at 2000watts, I feel a need to find a larger terminal block to effectively build them. If you think outside the box, you can actually put one together capable of handling 2000W + with just the 32A terminal blocks ... just use some imagination .... @sidehack - the actual model / description is Dell NY526 PowerEdge 2950 750W Redundant Power Supply |
|
|
|
I can tell by looking that it won't work on the 750W, at least not the ones I carry.
What's the part number for the Dell 750W PSU's connector if you care to share? |
|
|
|
Anyone interested in buying Dell 550W server PSU endplate connectors? I am putting together some of these for my own use but have to buy larger quantities and thus offload the excess. These PSU's can be used to power S1's, S3's and you can use 2 to power an S5 (or one S5 underclocked). The connectors can be assembled with either one, two or no switches. 1. For one switch it can either be used for reducing the fan RPM (this noise) or the power. If you choose to have it control the fan, then the plug will ship always switched on. 2. For two switches, one os for the fan and the other for powering up. 3. No switches means the plug will be switched on. The terminals I use are rated at 32 amps @ 300V each so easily cope with the 550W Dell server PSU. I can also source the PSU's (when available) + the 6pin PCI-e cables. Here are some images of the WIP - only remaining bit is adding the switches and putting them together. The Parts  The semi assembled plug  I am initially looking to ship to the UK and EU for a start as shipping farther may be too expensive, however, if you are not cowed by that, by all means get in touch! |
|
|
|
That is exactly what I'm wanting, but I'd like a heavier duty terminal,
The 32Amp terminals are the biggest (amperage wise) that I have had reason to search for, but there could be higher rated ones .... look at the usual places, aka digikey et al. I mentioned the Mk2's I had in progress, here are some images .... all that is left is wiring in the switch and gluing them together. The individal parts The semi assembled plug |
|
|
|
|
Another double-bubble ..... ! Surely, this makes a difference ....
|
|
|
|
Got the perspex fittings to put together the Mk1 with two three pole connectors ..... this is for a 550W Dell server PSU (not sure it fits the 750W one, but I'll be getting that later this week). I also added a switch for the power and had also designed a spare hole for the fan swicth (where the wires are protruding) but left it out on this one since the model I am going to run this on does not seem to turn down the fan. Front view  Rear view  In situ  I also have a Mk2 which is top monted, as opposed to the side mounted Mk1 above, but have not put it together yet. |
|
|
|
Been a bit of a slowdown ... hey!? And that wih the entire network too ... (at least it seems to me). Wonder whether that recent update to the blockchain.info website has anything to do with it  , it is taking a tad longer to load now it displays the entire transaction tree by default. Still ... I expect business as usual at slush soon .... |
|
|
|
|
All 3 confirmed ...... then another one! Whey hey!!!
|
|
|
|
|
Whoa! Three in a row within 2 minutes .... better not be invalid.
|
|
|
|
|
I have a sneaky feeling that is the same 2PH that was parked at kano.is until a few days back ..... it sure does not move around with any luck though!
|
|
|
|
|
Show Posts
