philipma1957
Legendary
Offline
Activity: 4270
Merit: 8661
'The right to privacy matters'
|
|
July 01, 2016, 03:02:59 PM |
|
There's a reddit post today indicating that the S9 supports ASIC boost. The thread included the stratum command for enabling it: {"id": 0, "method": "mining.midstate_change", "params": ["0x00000004", "0x04000004","0x08000004","0x0C000004"]} Apparently it only gets enabled on pools that support it. If there's any truth to this then I assume that ant pool enables ASIC boost. There's a fair amount of negativity around ASIC boost in general. If the S9 uses patented technology to mine faster for Bitmain or ant pool users than anywhere else then people are not going to be happy. Not that it will stop anyone from buying. If people are willing to buy without a hope of ROI I doubt that a theoretical philosophical issue will stop them. Bitmain is the only game in town... Please note that I have no idea if the reddit post is accurate. Probably a good idea to wait before taking out the pitchforks. It would be good to get an official statement from Bitmain, but I doubt that will happen. Well if the boost is 20% it makes antpool the only choice. a lot of details will be needed for this. I have 5 s-9's so it mining at antpool makes my 65 th become 78th it will be very hard to not use antpool. Maybe other pools will need to adopt the boost. The questions are: how big is the boost? what pools have the boost? can any pool get the boost? Ahh I see you all catching on to their implementation of "ASIC Boost". Selling a miner that may take a hair cut if this gets hard forked out of btc code which I've seen some talks of already. Maybe Phil this is the boost we keep seeing on the network. Them turning this boost on and off on pools that support it. Just a wild guess:) Very astute guess, the spike has come and gone since November 2015. You could be right. If true the game is going to change a lot way more then just the ½ing . Not so sure I like this.
|
|
|
|
Sierra8561
|
|
July 01, 2016, 03:04:58 PM |
|
One of my batch 1 miners had a hard time hashing at 14th. It's since been fixed with a firmware upgrade. One thing I noticed while troubleshooting is it always hashed at around 14 while mining on antpool. I thought it was a fluke until reading about asic boost.
|
|
|
|
d57heinz
Legendary
Offline
Activity: 1453
Merit: 1011
Bitcoin Talks Bullshit Walks
|
|
July 01, 2016, 03:17:31 PM |
|
There's a reddit post today indicating that the S9 supports ASIC boost. The thread included the stratum command for enabling it: {"id": 0, "method": "mining.midstate_change", "params": ["0x00000004", "0x04000004","0x08000004","0x0C000004"]} Apparently it only gets enabled on pools that support it. If there's any truth to this then I assume that ant pool enables ASIC boost. There's a fair amount of negativity around ASIC boost in general. If the S9 uses patented technology to mine faster for Bitmain or ant pool users than anywhere else then people are not going to be happy. Not that it will stop anyone from buying. If people are willing to buy without a hope of ROI I doubt that a theoretical philosophical issue will stop them. Bitmain is the only game in town... Please note that I have no idea if the reddit post is accurate. Probably a good idea to wait before taking out the pitchforks. It would be good to get an official statement from Bitmain, but I doubt that will happen. yea good luck with any word from them.. All i see from them is announcements of the next batch. No help here whatsoever. But hey there's a new batch come and get it. uffff Edit and if you think about it.. i remember an article about asic boost being 20% or thereabouts. So take bitmain 12 ths/ miner and add 20 % .. they sell it as a 14 for more btc and then i seen kano post about it only working on antpool.. Looks like there is def credibility behind kano post.! Best Regards d57heinz
|
As in nature, all is ebb and tide, all is wave motion, so it seems that in all branches of industry, alternating currents - electric wave motion - will have the sway. ~Nikola Tesla~
|
|
|
edonkey
Legendary
Offline
Activity: 1150
Merit: 1004
|
|
July 01, 2016, 03:29:33 PM |
|
For grins I scp'd a copy of bmminer from my S9 so that I could poke around with it. This is from a stock B3 unit without any firmware upgrade.
I used the strings command and searched for "midstate_change". I could not find it. Although it's clear from reviewing some of the function names that bmminer is a modified cgminer (as many have suspected).
Does anyone know what else to look for to get a sense of whether or not ASIC boost is included in this binary? Maybe someone knows how to disassemble the miner software? I'm a software engineer but have to admit that I'm better at writing code than reversing it.
|
Was I helpful? BTC: 3G1Ubof5u8K9iJkM8We2f3amYZgGVdvpHr
|
|
|
d57heinz
Legendary
Offline
Activity: 1453
Merit: 1011
Bitcoin Talks Bullshit Walks
|
|
July 01, 2016, 03:53:19 PM |
|
For grins I scp'd a copy of bmminer from my S9 so that I could poke around with it. This is from a stock B3 unit without any firmware upgrade.
I used the strings command and searched for "midstate_change". I could not find it. Although it's clear from reviewing some of the function names that bmminer is a modified cgminer (as many have suspected).
Does anyone know what else to look for to get a sense of whether or not ASIC boost is included in this binary? Maybe someone knows how to disassemble the miner software? I'm a software engineer but have to admit that I'm better at writing code than reversing it.
One thing i love about the internet. in time the truth is always revealed. Someone soon will come with this info. What really grinds my gears is that its not like bitmain needs this advantage at all. Phil that spike could have been them playing with the combination of 16 nm and asic boost.. Was a heck of a spike back then ehh!! Best Regards d57heinz
|
As in nature, all is ebb and tide, all is wave motion, so it seems that in all branches of industry, alternating currents - electric wave motion - will have the sway. ~Nikola Tesla~
|
|
|
|
d57heinz
Legendary
Offline
Activity: 1453
Merit: 1011
Bitcoin Talks Bullshit Walks
|
|
July 01, 2016, 04:42:39 PM |
|
Yes but kano made his remark before s9 went out i believe.. Which could have allowed them time to reflash .. But ya.. its def a shot in the dark. Save a lot of time hassle and speculation if they just open sourced the bmminer code. what they have to hide? And certainly i see no advantage to close sourced unless they have some advantage cooked into it. Best Regards d57heinz
|
As in nature, all is ebb and tide, all is wave motion, so it seems that in all branches of industry, alternating currents - electric wave motion - will have the sway. ~Nikola Tesla~
|
|
|
notlist3d
Legendary
Offline
Activity: 1456
Merit: 1000
|
|
July 01, 2016, 05:02:21 PM |
|
I did a decent amount of testing different pools for fun to see if any difference. During it I tested F2, Antpool, Kano.is, BTCC, Slush. Just to compare two here is Antpool (1 day): Here is Kano.is (2 day) So when I tested I got 2 pretty close numbers. So I wont speculate about "ASIC boost" as I just do not have any knowledge about it. I can say I did quite a few different pools from big to small, very different in my eyes, and never got near a 20 percent difference. I welcome others to compare and see what they get to. I think you will have same result's.
|
|
|
|
IITravel01
|
|
July 01, 2016, 06:05:34 PM |
|
Why does B1-B1 state +7% at the wall efficiency, but B5 state 1,172w + 12%? If true, then even at the lower hash rate it's less efficient?
|
|
|
|
philipma1957
Legendary
Offline
Activity: 4270
Merit: 8661
'The right to privacy matters'
|
|
July 01, 2016, 06:54:58 PM |
|
Why does B1-B1 state +7% at the wall efficiency, but B5 state 1,172w + 12%? If true, then even at the lower hash rate it's less efficient?
no just a bigger fudge margin to still be in spec. remember they are all we got so they can pretty much do as they please.
|
|
|
|
-droid-
Legendary
Offline
Activity: 1050
Merit: 1001
|
|
July 02, 2016, 12:06:02 AM |
|
Who's buying these so close to halving?
|
|
|
|
not.you
Legendary
Offline
Activity: 1726
Merit: 1018
|
|
July 02, 2016, 12:21:26 AM |
|
I wonder why the default freq on these is only 550. Same number of chips as previous batches, so no doubt the same board. Why lower the freq and thereby the hashrate? Are these QA failures on some level?
|
|
|
|
Sierra8561
|
|
July 02, 2016, 12:27:35 AM |
|
Rumor has it you can flash the low frequency firmware update on all batches. Weird thing, it wasn't available via their website. You had to download it on bitmain talk.
|
|
|
|
edonkey
Legendary
Offline
Activity: 1150
Merit: 1004
|
|
July 02, 2016, 01:39:35 AM |
|
Who's buying these so close to halving?
I got a B3. It was kind of a sweet spot because BTC was at a high point, and I got it with enough time on this side of the halving to be significant. I actually have a chance to ROI it in BTC terms. But the other batches have not made sense. In fact I just ran the numbers on B5 and it's worse than B3. These are going to have to come down in BTC price terms before I'll get any more of them. I expect that will happen soon after the halving. Or maybe monkeys will fly out of my butt and someone will start to compete with Bitmain. That would help drive prices down. Whatever happened to BW's miner, for example? They been mining with them and were supposed to offer them to the public last month.
|
Was I helpful? BTC: 3G1Ubof5u8K9iJkM8We2f3amYZgGVdvpHr
|
|
|
QuintLeo
Legendary
Offline
Activity: 1498
Merit: 1030
|
|
July 02, 2016, 01:46:48 AM |
|
I wonder why the default freq on these is only 550. Same number of chips as previous batches, so no doubt the same board. Why lower the freq and thereby the hashrate? Are these QA failures on some level?
Yield and manufacturing variation on a very NEW manufacturing process node, I'd guess. Batch 1 seemed to have some MAJOR issues with not being able to meet it's specs on a lot of units - similar to the early S7 batches, then they redesigned the hashboards on the S7 to use a buck and fewer chips running at higher voltage, thus managing to meet the hashrate spec while eating noticeably more power....
|
I'm no longer legendary just in my own mind! Like something I said? Donations gratefully accepted. LYLnTKvLefz9izJFUvEGQEZzSkz34b3N6U (Litecoin) 1GYbjMTPdCuV7dci3iCUiaRrcNuaiQrVYY (Bitcoin)
|
|
|
Swimmer63
Legendary
Offline
Activity: 1593
Merit: 1004
|
|
July 02, 2016, 03:17:20 AM |
|
I wonder if they sorted their chips on this first chip run and are adjusting the miners to work with the chip run. B1 had the best chips and they are working through the rest of the batch. Just a thought. Not my area of expertise. My B4 has arrived and started up. No hard data yet though.
|
|
|
|
sunny163
Member
Offline
Activity: 126
Merit: 10
|
|
July 02, 2016, 05:49:16 AM |
|
this temperature is ok for S9 batch 2 or I need to lower down the frequency or to set fans to manually ? http://imgur.com/SdSUMEz
|
|
|
|
Sierra8561
|
|
July 02, 2016, 05:51:33 AM |
|
this temperature is ok for S9 batch 2 or I need to lower down the frequency or to set fans to manually ? http://imgur.com/SdSUMEzThose Temps are acceptable
|
|
|
|
AriesIV10
Legendary
Offline
Activity: 1260
Merit: 1006
Mine for a Bit
|
|
July 02, 2016, 11:06:47 AM |
|
this temperature is ok for S9 batch 2 or I need to lower down the frequency or to set fans to manually ? http://imgur.com/SdSUMEzThose Temps are acceptable Nice temps!
|
|
|
|
sunny163
Member
Offline
Activity: 126
Merit: 10
|
|
July 02, 2016, 11:10:26 AM |
|
this temperature is ok for S9 batch 2 or I need to lower down the frequency or to set fans to manually ? http://imgur.com/SdSUMEzThose Temps are acceptable Nice temps! this temp is perfect or I need to do something to lower down it ?
|
|
|
|
|