0,216*1.0638 (ideal string design + platinum PSU at peak 94% efficiency) = 0.23
This 230W at wall is to idealistic. Real world numbers will be 250-260W/TH.
230W is at the wall, it includes PSU and any other inefficiencies. They've not been wrong on the previous 3 generations so I'm not sure why they'd get it wrong now. What's the PSU efficiency assumed in the calculation? How far in advance before producing actual miners have they announced the previous 3 (perhaps you meant 2) generations of chips?Nevermind, I'm too blind for Bitmain's fine print. Bitmain had to backpedal on S3 specs and they issued a compensation for Batch 1 purchases - so that's 2 out of 3 as far as not being wrong. That wasn't chips, that was inappropriate DC-DC which made it unstable for 24/7 use. |
|
|
|
The jurisdiction that the forum operates in is the US.
Just for future reference, don't rely on the forum being 100% US. Various countries can claim that internet activity happens in the country of the user and so local laws apply. Now they might not immediately be able to enforce a foreign judgement in the US but they can certainly come after individuals. |
|
|
|
You can use as many PSUs as you want as long as you don't have multiple PSUs powering the same individual board (of which there is 9).
Do all PCI ports need to be powered? Or will it run with 2 on each board? 2 is fine, still only about 165W per PCI-E. Are you certain that using 2 is fine? The product description page says all three are required. It may be that each PCIe plug powers a specific set of chips; the 3 PCIe connectors may not share a common power bus at all as in, say, the S3 design. Without knowing the actual S5+ design, or having an actual unit to look at, saying "2 is fine" is only speculation at this point unless something more is known than what bitmain has disclosed so far. Bitmain could confirm this here, but they aren't posting much these days. If anyone already has a unit, they could also confirm whether using 2 out of the 3 connectors works and post here. What Bitmain customer Service told us was we have to use all 3 connectors for each board, As my understanding if each board need at least 382W(3436/9), in average each connector should be bear 127w@10.6A , and if it works using only 2 connectors instead that will be 191w@15.9A for each connector , it is high risk to heat up or burn your cables if they are not good enough. This still doesn't answer the question of whether all 3 are *required* by design, or can 2 be used if the cables can handle the increased current? BTW the power requirements aren't as high as you mention: 3436W quoted by bitmaintech is at the wall, using a 93% efficient PSU. So the DC power to the S5+ would only be 3436*0.93 or 3196 watts, or 355W per board. *If* the design allowed the use of 2 connectors per board, this would put 178W per connector (14.8A). This isn't bad, considering S3+ units can easily be powered with 2 PCIe cables and those use around 355W as well. They make a big deal about using 3 per board, so I would not try to personally. They actually ship a piece of paper with some general instructions, and an email with it to. And in all of them it made sure you use 3 per board. So under pcie specs should it run if high quality psu with nice cables... yes. But they make sure to tell you only run with 3 in each. If you do 2 i'm sure it violates warranty. Many comments I can reply to in one quote! The reason it says 3 connectors are required is because I wrote that long ago for S5, and they just updated the numbers in the announcement without reconsidering what it was actually saying. If you've got 16AWG then you could use 2 PCI-E from the same branch, 18AWG and you'll need 2 branches and you should keep checking on it for the first 30 minutes. |
|
|
|
It won't happen, the CN pools won't support anything along the lines of a fork even if it does allow for larger block sizes. Even Bitmain alone can block a fork.
It is my understanding that the chinese pools, for the most part do not own the hardware they are mining with, so switching pools to one that does support XT (or any other fork) would be as simple as changing a few settings. It is also possible that something could change in the future that would convince the Chinese pools to support XT, perhaps if major a major payment processor were to publicly support XT.....I am not sure if any of them will though. Many of the CN pools control large amounts of hash rate which they may not own but instead manage (inc cloud) so they can dictate what we do. And that will not be supporting XT (unless there is an overwhelming consensus). What everyone wants is for the developers to stop acting like children and just agree on a block size. Make it 1.1MB as a first step, see if everyone magically runs out of hard drive space and go from there. |
|
|
|
0,216*1.0638 (ideal string design + platinum PSU at peak 94% efficiency) = 0.23
This 230W at wall is to idealistic. Real world numbers will be 250-260W/TH.
230W is at the wall, it includes PSU and any other inefficiencies. They've not been wrong on the previous 3 generations so I'm not sure why they'd get it wrong now. |
|
|
|
Definitely no. The problem isn't XT nor the forum's policy. It is the users that are constantly opening threads for the same topics in the wrong subsection.
If BitcoinXT deserves a section for itself, doesn't DASH (just used as example, I'm not advocating it) deserve one too? Are you trying to tell me there are more posts about XT than about DASH (currently ~105k)?
To simplify, the main reason for which neither have or are going to have their section is because this is BITCOINtalk, not bitcoinXTtalk, nor DASHtalk.
Dash is an altcoin, XT isn't. XT is. Not yet. But, it will be. Right now, XT is just an alternative coin. If they can gain 75%+ hash power, they'll allow bigger blocks and then that would be an alt coin. It won't happen, the CN pools won't support anything along the lines of a fork even if it does allow for larger block sizes. Even Bitmain alone can block a fork. I wish they dont. But, did not bitmain sign the letter supporting 8MB block ? Supporting the idea that block sizes should increase, NOT any form of forming or XT which is considered dangerous. They want the core devs to work together. |
|
|
|
Definitely no. The problem isn't XT nor the forum's policy. It is the users that are constantly opening threads for the same topics in the wrong subsection.
If BitcoinXT deserves a section for itself, doesn't DASH (just used as example, I'm not advocating it) deserve one too? Are you trying to tell me there are more posts about XT than about DASH (currently ~105k)?
To simplify, the main reason for which neither have or are going to have their section is because this is BITCOINtalk, not bitcoinXTtalk, nor DASHtalk.
Dash is an altcoin, XT isn't. XT is. Not yet. But, it will be. Right now, XT is just an alternative coin. If they can gain 75%+ hash power, they'll allow bigger blocks and then that would be an alt coin. It won't happen, the CN pools won't support anything along the lines of a fork even if it does allow for larger block sizes. Even Bitmain alone can block a fork. |
|
|
|
The mods could dedicate a child board for XT in
the altcoin section, and that might even have
a positive effect and encourage more constructive
dialogue.
The only healthy dialogue regarding XT is that it should die immediately. Nevermind double spends, nevermind "51%s", nevermind transaction malleability, forking is the true threat. |
|
|
|
Sooo .... anyone noticed what the effect of this announcement has had on the bitcoin price today?
You mean the announcement went back in time 8 hours to cause a drop? |
|
|
|
|
40% power consumption of previous gen is a pretty huge jump.
|
|
|
|
|
We don't need a new thread for this, use one of the other two.
|
|
|
|
I'm fairly confident they will sell more batches of the s5+, it just makes sense from a business perspective.. Why release new gen when the competition is non existent.. S5+ was tempting but I opted to get some sp20's instead for the stealth factor
Depends, the S2 and S3++ were used to help consume the last of a generation's chip supplies and the S5+ may be the same. There's probably a finite number of S5+s planned and even if they're selling better than expected there may not be significant follow up stock. The S7 has to appear at some point. |
|
|
|
Interesting that in that press release, unless I'm just not seeing it, the words "Bitcoin" and "Mining" don't appear anywhere in it. Instead they use "Transaction Verification Services" and "Hashing" as the only words that describe what they are doing. Yeah, I think there's either a press blurb or an interview somewhere that explains the reasoning for it, would have been around may/june, shortly after they put out their bit about hooking up with Spondoolies-Tech; https://bitcointalk.org/index.php?topic=521520.msg11412144#msg11412144iirc, came down to them feeling it more accurately describes what they're doing, although I'd imagine somebody with a 'marcom' itch had something to do with it as well  I wonder if they take visitors.. I should be in that area in ~2 months Most of the big companies switched to that last year, 21 inc, Bitfury etc will all say transaction verification. It helps investors understand why we're burning millions of KWh's and why they should give us money to continue doing so. |
|
|
|
Whenever an account's password is changed, everyone who left positive feedback on it should be automatically notified that they should investigate whether the account was just sold and if so, to delete their feedback that applied only to the previous owner.
I think you're onto a good idea, but it would get a little annoying for users who do a ton of trades (Blazedout419 for example) because they'd be getting a lot of notifications whenever a user they sent positive trust to changes their password. I think a better idea would be notifying everyone who left positive trust for a user, if that user changes their password more than once every 12 or 24 hours. Because it seems often when an account is sold the password is changed at least twice in less than 6 hours: once when the seller sends the account details and then twice when the new buyer secures the account for themselves. Something like that might be a better idea because it wouldn't be as annoying as getting a notification every password change, but would still provide useful information to users who have sent positive feedback. I too like the idea but I don't think we've come up with quite the right formula yet. Maybe rather than an automated system, several members could vote to activate a warning system as described. That way its getting human checked that something looks iffy or the behaviour has changed. |
|
|
|
There are no official resellers any more and there is no qualification by Bitmain. The only recommendation is to buy from the official webstore (which is still going to be the cheapest anyway), anything else you do so at your own risk, and with them for order and warranty support.
not true > "Customer Service : Yanhua Qin 2015-08-05 04:21:26 > Hi, > you may contact support_at_hadsystems.de directly, it's our official > distributor in Germany." Well you've sort of just proven my point, if you buy from a 3rd party then support and warranty is via a 3rd party not Bitmain. Hence if you want to guarantee you don't get a situation like Minersource again, you purchase directly from Bitmain. |
|
|
|
I see you have been posting a lot. Are you a official reseller? (IE bitmain warrenty is still valid if bought from you, as it does not transfer people which kills most resellers).
Just was not sure if you are official reseller or unofficial. Thanks.
There are no official resellers any more and there is no qualification by Bitmain. The only recommendation is to buy from the official webstore (which is still going to be the cheapest anyway), anything else you do so at your own risk, and with them for order and warranty support. |
|
|
|
|
Typically you tell someone when you write an "open letter", especially when its in off topic.
1) I haven't reported a thing in that thread for months
2) I don't know who has reported things
3) I don't know who has deleted things
4) I don't know what has been deleted
5) I don't know when things are deleted.
6) You tell me to grow up, yet without any evidence conclude that someone elses actions must be mine and rage delete a trust rating left FOR A TRADE, as well as exclude me from your list.
All I know is that either 3, 23, 43 etc posts were deleted as my "16 August 2015, 09:49:21" post was the first on a new page.
|
|
|
|
nice what is the real performance in the pool?
we are seeing 7.7- 8.0 TH at stock settings Damn those fans are going flat out, and seemingly without reason. If you're at 25C ambient, then even at a 40C ambient those chips would be at 60C = still far below the cutoff. May be able to get away with removing / unplugging the front 3 fans, especially at that ambient. |
|
|
|
|
Show Posts
