contactlight
|
|
July 11, 2014, 03:20:59 AM |
|
And surprise surprise, they are not shipped yet!
They stated earlier it would be the 14th. Let's not freak out yet, this is BITMAIN we're talking about. Actually, it's my bad. I thought today was the 14th when I wrote that message.
|
|
|
|
edonkey
Legendary
Offline
Activity: 1150
Merit: 1004
|
|
July 11, 2014, 06:04:29 AM |
|
While we're waiting, does anyone have suggestions for the difficulty to set for S3 units (for pools that allow configurable difficulty, of course)?
My current experience is limited to GridSeeds and Scrypt mining. In that case, where multiple ASIC units are effectively behind a single IP address, I have had more manageable results by configuring relatively low difficulty.
What I mean by "manageable" is that I can generally tell at a glance whether my rigs are doing well by how recently they've submitted shares. With lower difficulty, I rarely see the units struggling to submit. With higher difficulty, the GridSeeds would work in some cases for several hundred seconds before submitting. This sometimes mades it difficult to differentiate between a device that has gone non-responsive and one that's just working hard and hasn't submitted yet.
Also, and this is harder to quantify, but since I've moved to low-ish difficulty, none of my GridSeed units have hung. This is after many weeks of continuous uptime. Prior to overriding the difficulty, I'd see a unit go non-responsive (where no shares would be submitted for hours on end) every few days or so. The affected unit was not always the same, which might have indicated a software problem.
Maybe high difficulty, or long times between shares enabled the problem, so this observation might not translate to other rigs. Given that, it's hard to conclude that low difficulty equals stability in all cases. But with my rigs, that certainly appears to be the case.
I realize that from a statistical standpoint that difficulty has no effect on long term profitability. But as I've said, my experience has been that lower difficulty for my current ASIC-based rigs has made it easier to manage and spot problems.
I have no experience with any Antminer units. But if the S3 is made up of lots of ASIC chips, all of which are under the same IP address and effectively appearing as one 500 GH/s unit to the pool, would it be better to set a low-ish difficulty that more matches the compute power of the individual chips rather than the unit as a whole?
|
Was I helpful? BTC: 3G1Ubof5u8K9iJkM8We2f3amYZgGVdvpHr
|
|
|
Collider
|
|
July 11, 2014, 06:19:51 AM |
|
Set your difficulty between 5XX and 2XX for units of ~480GH.
|
|
|
|
oskuro
|
|
July 11, 2014, 07:10:21 AM |
|
Set your difficulty between 5XX and 2XX for units of ~480GH.
whats the purpose of doing that?You earn more? and how do you set up difficulty?
|
|
|
|
allcoinminer
|
|
July 11, 2014, 07:23:58 AM |
|
Set your difficulty between 5XX and 2XX for units of ~480GH.
whats the purpose of doing that?You earn more? and how do you set up difficulty? Technically setting low difficult or high difficulty will not give you any benefit in time. But for lower hash power miners, setting higher difficulty will take a lot of time for the ASIC to hash and will report less frequently. In that time higher has power miners may already be reported back and solved a block so the lower hashrate miner will always report late and the work will go waste. Same time if a lower difficulty is set the lower hash power miner will solve fast and report in time. Likewise, if we set lower difficulty for higher hash rate miners, they will solve it fast and report very often lead to unwanted traffic and its better to set a higher difficulty.
|
|
|
|
oskuro
|
|
July 11, 2014, 07:35:59 AM |
|
Set your difficulty between 5XX and 2XX for units of ~480GH.
whats the purpose of doing that?You earn more? and how do you set up difficulty? Technically setting low difficult or high difficulty will not give you any benefit in time. But for lower hash power miners, setting higher difficulty will take a lot of time for the ASIC to hash and will report less frequently. In that time higher has power miners may already be reported back and solved a block so the lower hashrate miner will always report late and the work will go waste. Same time if a lower difficulty is set the lower hash power miner will solve fast and report in time. Likewise, if we set lower difficulty for higher hash rate miners, they will solve it fast and report very often lead to unwanted traffic and its better to set a higher difficulty. but that difficulty has to be configured on the pool? or in the antminer ?. I mine @ bitminter. And default difficulty of S1 / S3 its not good then?(i know no one has S3 yet but i suppose will the same as S1)
|
|
|
|
Maidak
Legendary
Offline
Activity: 1890
Merit: 1058
Vave.com - Crypto Casino
|
|
July 11, 2014, 08:35:42 AM |
|
I purchased one of these for 8 of them to run wireless http://www.amazon.com/gp/product/B004TLIVBG/ref=oh_details_o04_s00_i01?ie=UTF8&psc=1 hopefully it works without issue I went with this one since I currently use a cisco e3000 (Wish I would have went with an asus) I don't see any reason why these wouldn't work a little on the pricey side however.
|
|
|
|
GigaBit
|
|
July 11, 2014, 09:05:54 AM |
|
While we're waiting, does anyone have suggestions for the difficulty to set for S3 units (for pools that allow configurable difficulty, of course)?
I set my S1's at 256 as what is recommended and is also what I have had the most success with. Yes difficulty can make a significant difference, especially if a pool sets it automatically or has a very low default difficulty setting. Since the S3's are about 2.5x faster than the S1, then I'd say a difficulty 512 would be a good logical place to start; 200 x 2.5 = 500 Where 200 is the max speed of a unit at 256 difficulty; I never don't overclock something. I'd also try over and under 64, 128 & 256 differences to push the envelope and get that extra bunch of satoshi a day for doing nor spending on anything else
|
..Stake.com.. | | | ▄████████████████████████████████████▄ ██ ▄▄▄▄▄▄▄▄▄▄ ▄▄▄▄▄▄▄▄▄▄ ██ ▄████▄ ██ ▀▀▀▀▀▀▀▀▀▀ ██████████ ▀▀▀▀▀▀▀▀▀▀ ██ ██████ ██ ██████████ ██ ██ ██████████ ██ ▀██▀ ██ ██ ██ ██████ ██ ██ ██ ██ ██ ██ ██████ ██ █████ ███ ██████ ██ ████▄ ██ ██ █████ ███ ████ ████ █████ ███ ████████ ██ ████ ████ ██████████ ████ ████ ████▀ ██ ██████████ ▄▄▄▄▄▄▄▄▄▄ ██████████ ██ ██ ▀▀▀▀▀▀▀▀▀▀ ██ ▀█████████▀ ▄████████████▄ ▀█████████▀ ▄▄▄▄▄▄▄▄▄▄▄▄███ ██ ██ ███▄▄▄▄▄▄▄▄▄▄▄▄ ██████████████████████████████████████████ | | | | | | ▄▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▄ █ ▄▀▄ █▀▀█▀▄▄ █ █▀█ █ ▐ ▐▌ █ ▄██▄ █ ▌ █ █ ▄██████▄ █ ▌ ▐▌ █ ██████████ █ ▐ █ █ ▐██████████▌ █ ▐ ▐▌ █ ▀▀██████▀▀ █ ▌ █ █ ▄▄▄██▄▄▄ █ ▌▐▌ █ █▐ █ █ █▐▐▌ █ █▐█ ▀▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▀█ | | | | | | ▄▄█████████▄▄ ▄██▀▀▀▀█████▀▀▀▀██▄ ▄█▀ ▐█▌ ▀█▄ ██ ▐█▌ ██ ████▄ ▄█████▄ ▄████ ████████▄███████████▄████████ ███▀ █████████████ ▀███ ██ ███████████ ██ ▀█▄ █████████ ▄█▀ ▀█▄ ▄██▀▀▀▀▀▀▀██▄ ▄▄▄█▀ ▀███████ ███████▀ ▀█████▄ ▄█████▀ ▀▀▀███▄▄▄███▀▀▀ | | | ..PLAY NOW.. |
|
|
|
xjack
|
|
July 11, 2014, 10:01:32 AM |
|
That's not gonna work for a 240v setup.
|
|
|
|
StuffOfInterest
|
|
July 11, 2014, 10:15:28 AM |
|
That's not gonna work for a 240v setup. Look for a European version of the adapter. I've saw one in an apartment a few months ago with a standard western European 2-prong plug. Should be able to find it on one of the European Amazon sites.
|
|
|
|
cech4204a
|
|
July 11, 2014, 11:45:59 AM |
|
Am is blind or i can see that S3 is slower than S2 ? S2 is in another topic stated to mine at 1THS?
|
Bitcoin is DEAD
|
|
|
hurricandave
Legendary
Offline
Activity: 966
Merit: 1003
|
|
July 11, 2014, 11:51:03 AM |
|
Am is blind or i can see that S3 is slower than S2 ? S2 is in another topic stated to mine at 1THS?
Buuuuuuuzzz. S3 is the next Gen successor of the S1,....an S2 is an entirely different machine.
|
|
|
|
oskuro
|
|
July 11, 2014, 11:51:38 AM |
|
Am is blind or i can see that S3 is slower than S2 ? S2 is in another topic stated to mine at 1THS?
?? S3 its 478GH/s @366W S2 its 1 TH/s @less than 1100W from the wall. I think its way better S3 because of power .S2 its entirely diffent also different machine
|
|
|
|
atledenin
Newbie
Offline
Activity: 29
Merit: 0
|
|
July 11, 2014, 11:52:52 AM |
|
http://i.ebayimg.com/00/s/NDI3WDY0MA==/z/TTgAAOSw9NxTsnPQ/$_57.JPGJust spotted this on an ebay auction. Looks like the heatsink arrangement is different, unless they have added those on the chip to help venting heat with the cover on. Edit : Just spotted the other heatsinks (as with S1) through the fan grill.
|
|
|
|
Maidak
Legendary
Offline
Activity: 1890
Merit: 1058
Vave.com - Crypto Casino
|
|
July 11, 2014, 12:11:05 PM |
|
Just spotted this on an ebay auction. Looks like the heatsink arrangement is different, unless they have added those on the chip to help venting heat with the cover on. Edit : Just spotted the other heatsinks (as with S1) through the fan grill. Can't wait to get my batch one order! So I am currently in the works of a 1400 watt power supply with 6x 6pin pci-e cables which I can offer at a good rate wondering if anyone would have interest in them? I am getting the sample unit delivered sometime around the 16th. The price on them will be about $150~ish If i'm able to get enough crowd funding for 2-500 units.
|
|
|
|
cech4204a
|
|
July 11, 2014, 12:14:26 PM |
|
Am is blind or i can see that S3 is slower than S2 ? S2 is in another topic stated to mine at 1THS?
?? S3 its 478GH/s @366W S2 its 1 TH/s @less than 1100W from the wall. I think its way better S3 because of power .S2 its entirely diffent also different machine Oh ok i didn't know that i guess the manufacturer is different than?
|
Bitcoin is DEAD
|
|
|
allcoinminer
|
|
July 11, 2014, 12:15:10 PM |
|
[mg]http://i.ebayimg.com/00/s/NDI3WDY0MA==/z/TTgAAOSw9NxTsnPQ/$_57.JPG[/img]
Just spotted this on an ebay auction. Looks like the heatsink arrangement is different, unless they have added those on the chip to help venting heat with the cover on.
Edit : Just spotted the other heatsinks (as with S1) through the fan grill.
The image was already out the same day of the announcement by BitMainTech. Bitmain already said the heat sink is different but, it will be compatible with S1 to upgrade. More DC-DC converters are there now, we may easily add additional miniature heatsinks over them. As per the image its a rectangular block instead of the S1's square block. The fan is placed to the bottom end. When the cover is placed and the voltage controllers are at the top chances are there to concentrate hot air over the top side of the box.
|
|
|
|
jonnybravo0311
Legendary
Offline
Activity: 1344
Merit: 1024
Mine at Jonny's Pool
|
|
July 11, 2014, 12:32:37 PM |
|
Am is blind or i can see that S3 is slower than S2 ? S2 is in another topic stated to mine at 1THS?
?? S3 its 478GH/s @366W S2 its 1 TH/s @less than 1100W from the wall. I think its way better S3 because of power .S2 its entirely diffent also different machine Oh ok i didn't know that i guess the manufacturer is different than? No... same manufacturer. The difference is in the ASIC used in each machine. By the way, you forgot to mention that the S3 is cheaper, too. S2 - 1TH/s @ 1100W is $1625 2 S3s - 956GH/s @ 732W is ~$900
|
Jonny's Pool - Mine with us and help us grow! Support a pool that supports Bitcoin, not a hardware manufacturer's pockets! No SPV cheats. No empty blocks.
|
|
|
oskuro
|
|
July 11, 2014, 12:34:18 PM |
|
Am is blind or i can see that S3 is slower than S2 ? S2 is in another topic stated to mine at 1THS?
?? S3 its 478GH/s @366W S2 its 1 TH/s @less than 1100W from the wall. I think its way better S3 because of power .S2 its entirely diffent also different machine Oh ok i didn't know that i guess the manufacturer is different than? No... same manufacturer. The difference is in the ASIC used in each machine. By the way, you forgot to mention that the S3 is cheaper, too. S2 - 1TH/s @ 1100W is $1625 2 S3s - 956GH/s @ 732W is ~$900 yeah, didnt think it was necessary to talk about price hehe. its a nonsense to buy S2 having S3
|
|
|
|
atledenin
Newbie
Offline
Activity: 29
Merit: 0
|
|
July 11, 2014, 12:49:28 PM |
|
[mg]http://i.ebayimg.com/00/s/NDI3WDY0MA==/z/TTgAAOSw9NxTsnPQ/$_57.JPG[/img]
Just spotted this on an ebay auction. Looks like the heatsink arrangement is different, unless they have added those on the chip to help venting heat with the cover on.
Edit : Just spotted the other heatsinks (as with S1) through the fan grill.
The image was already out the same day of the announcement by BitMainTech. Bitmain already said the heat sink is different but, it will be compatible with S1 to upgrade. More DC-DC converters are there now, we may easily add additional miniature heatsinks over them. As per the image its a rectangular block instead of the S1's square block. The fan is placed to the bottom end. When the cover is placed and the voltage controllers are at the top chances are there to concentrate hot air over the top side of the box. Cool, is it a better image from the press release? Would like to see a bit closer up!
|
|
|
|
|