8th Alt coin thread. Or what to do now that asics are all over the place.

[citronick]

Just back from long consulting work with limited Internet access and catching up with the thread now.

In my inbox, saw a few interesting emails... one of them... Bitmain firesales... all SHA256 miners are at $7xx and $6xx (without PSU) -- are we looking at the coming of the mega-ASICs-miner?

Knowing how Bitmain works.... are we starring at a possible 100TH miner at 1000w by the end of the year?

Make sure to answer the invite from generalt

Free contest for a silver coin.


No way will a 100th miner with 1000 watts goes on sale.

I think they have a 1300 watt 19th machine

Looks like the firesale is because of GMO's product announcement and sales of their miner tomorrow

https://news.bitcoin.com/japans-gmo-specs-price-7nm-bitcoin-mining-rigs/

$1999, 24TH @ 1950watts with 2200w PSU + taxes etc

[1714496335]

1714496335

[1714496335]

1714496335

[1714496335]

1714496335

[1714496335]

1714496335

[soothaa]

Looks like the firesale is because of GMO's product announcement and sales of their miner tomorrow

https://news.bitcoin.com/japans-gmo-specs-price-7nm-bitcoin-mining-rigs/

$1999, 24TH @ 1950watts with 2200w PSU + taxes etc

Not sure why.. this thing is over 2x the price for less than 2x the performance...

[fr4nkthetank]

Looks like the firesale is because of GMO's product announcement and sales of their miner tomorrow

https://news.bitcoin.com/japans-gmo-specs-price-7nm-bitcoin-mining-rigs/

$1999, 24TH @ 1950watts with 2200w PSU + taxes etc

Not sure why.. this thing is over 2x the price for less than 2x the performance...

well its what 20% (just rough guess) more efficient ?  Warrants a premium on the initial purchase price for an equal ROI since running costs are lower / hash.  but yeah.  and have you guys seen the difficulty rise .______.  

[shield132]

Link to thread 7

https://bitcointalk.org/index.php?topic=2674194.0


I self modded the thread. I will simply delete nasty attacks on each other about the good or bad of asics vs gpus vs fpgas.

I want focus to be on how to survive the shift.

Do I go all asics?

Do I sell all gpus?

Do I go diverse?

My position is the industry needs to be widespread as possible to create demand for coins.

This year I am no longer focusing on what gear to use.

But how to expand my cheap reliable power.

We know some people can not do this and some can.

My gear today is

16x 1080tis
1rx 560
1rx 460
1 1050ti
13 moonlander usb ltc sticks
12 Antminer L3+ with jstenop low watt software.
5 Hanlong T1
3 Avalon 841
2 ⅔ S-9

Currently GPU mining works, doesn't it? Despite the fact that Bitmain tries to ruin everything. You have GTX1080ti, why to sell them? Personally I would go 40% GPU, 60% ASIC (remember how profitable your GPU was some months ago).
What's about future, to my mind game industry mustn't be turned into mining industry so GPU's mustn't be used in mining anyway, my vote to asic for that.
And to be fair you don't need my advice but just stated my opinion, I look you from outside and you are amazing member!

[R0mi]

Currently GPU mining works, doesn't it? Despite the fact that Bitmain tries to ruin everything. You have GTX1080ti, why to sell them? Personally I would go 40% GPU, 60% ASIC (remember how profitable your GPU was some months ago).
What's about future, to my mind game industry mustn't be turned into mining industry so GPU's mustn't be used in mining anyway, my vote to asic for that.
And to be fair you don't need my advice but just stated my opinion, I look you from outside and you are amazing member!

Yes, but let's dust off the MagikSmoke 8 ball and look at the next few months:

Z9 will destroy Equihash in short order.  Probably will be a replay of the L3+.  There goes your 1080 Ti cash cow.

Baikal Giant does X11/Quark/Quibt/Myriad-Groestl/Nist5/Skein/X11Gost/ and maybe a few more in a few months.  At your doorstep in about 5 days for $1,600.

Tribus is number 1 on Whitefire's FPGA bitstream release list:

http://zetheron.com/index.php/fpga-performance-profit/

And Tribus can support about 130 FPGAs until profit goes to Bitmain ShiteHeaven.  And people are already mining Tribus using FPGAs, I mined it for quite some time with 1080 Ti's and it is a waste of money to even try anymore and Whitefire's bitstreams are not even fully out in the wild yet.

Blake variants have been destroyed by A3, Halong, BK-B.  Ditto for Decred.  And I haven't even received my damn Obelisk SIA miner yet!

A word about the whole FPGA thing:  if you follow all the Bitcoin threads and the discord, the FPGA roll out is shaping up in the *short* term to be every bit as difficult as the Vega's were and then some.  But it took what?  About three or four months for people to figure out ways to make the Vega profitable.  Yes, cooling mods required at first.  Yes, Senseless will not support Windows.  GPUHoarder is interested in cost-efficient accelerators.

But Vosk will be posting YouTube FPGA videos by Thanksgiving, if not before then.  You wait and see.

Cryptonight and all of its variants are squarely in the sights of the FPGA guys.  I will go on record to say right now Vega's won't be able to compete on Cryptonight come December.

That leaves X16 variants -- and those are next in line for FPGA domination, the only drawback is that you need to link a few FPGAs together and that won't happen immediately on a large, easy to implement scale.

And then we come to Eth hash.

That seems like a very narrow doorway for an awful lot of GPUs.

Please feel free to point out what I am missing and where I am wrong.  Unless the Volta is really cheap, I don't think it changes anything near term.  And Vega 20?  Who the hell knows, and even when it comes out, it will be every bit as unoptimized by the AMD genius techs as was Vega 10.  So, unless you are part of OhGodACartel/OhGodASlovenianMiningFarm/OhGodWhyDoIFixAMD?, good luck to you.

[ZedZedNova]

... and have you guys seen the difficulty rise .______.  

Holy crap! 14.7%. Man my Gekko's never gonna ROI at this rate 

[philipma1957]

Currently GPU mining works, doesn't it? Despite the fact that Bitmain tries to ruin everything. You have GTX1080ti, why to sell them? Personally I would go 40% GPU, 60% ASIC (remember how profitable your GPU was some months ago).
What's about future, to my mind game industry mustn't be turned into mining industry so GPU's mustn't be used in mining anyway, my vote to asic for that.
And to be fair you don't need my advice but just stated my opinion, I look you from outside and you are amazing member!

Yes, but let's dust off the MagikSmoke 8 ball and look at the next few months:

Z9 will destroy Equihash in short order.  Probably will be a replay of the L3+.  There goes your 1080 Ti cash cow.

Baikal Giant does X11/Quark/Quibt/Myriad-Groestl/Nist5/Skein/X11Gost/ and maybe a few more in a few months.  At your doorstep in about 5 days for $1,600.

Tribus is number 1 on Whitefire's FPGA bitstream release list:

http://zetheron.com/index.php/fpga-performance-profit/

And Tribus can support about 130 FPGAs until profit goes to Bitmain ShiteHeaven.  And people are already mining Tribus using FPGAs, I mined it for quite some time with 1080 Ti's and it is a waste of money to even try anymore and Whitefire's bitstreams are not even fully out in the wild yet.

Blake variants have been destroyed by A3, Halong, BK-B.  Ditto for Decred.  And I haven't even received my damn Obelisk SIA miner yet!

A word about the whole FPGA thing:  if you follow all the Bitcoin threads and the discord, the FPGA roll out is shaping up in the *short* term to be every bit as difficult as the Vega's were and then some.  But it took what?  About three or four months for people to figure out ways to make the Vega profitable.  Yes, cooling mods required at first.  Yes, Senseless will not support Windows.  GPUHoarder is interested in cost-efficient accelerators.

But Vosk will be posting YouTube FPGA videos by Thanksgiving, if not before then.  You wait and see.

Cryptonight and all of its variants are squarely in the sights of the FPGA guys.  I will go on record to say right now Vega's won't be able to compete on Cryptonight come December.

That leaves X16 variants -- and those are next in line for FPGA domination, the only drawback is that you need to link a few FPGAs together and that won't happen immediately on a large, easy to implement scale.

And then we come to Eth hash.

That seems like a very narrow doorway for an awful lot of GPUs.

Please feel free to point out what I am missing and where I am wrong.  Unless the Volta is really cheap, I don't think it changes anything near term.  And Vega 20?  Who the hell knows, and even when it comes out, it will be every bit as unoptimized by the AMD genius techs as was Vega 10.  So, unless you are part of OhGodACartel/OhGodASlovenianMiningFarm/OhGodWhyDoIFixAMD?, good luck to you.

pretty much spot on.  Gpu mining would need a new algo so  yeah I will mine my 1080ti's  for now

14x of them made 150 usd since friday  they used to do 150 usd a day  but they are paid off

2x on a shit coin  made 150 usd since may 15th about 6 a day for 2   or 3 a day for 1.

sooner or later  people will find that coin and kill it off.

[jmigdlc99]

Yes, but let's dust off the MagikSmoke 8 ball and look at the next few months:
-snip-

Great analysis and summary of GPU outlook overall. This should be stickied for all those still asking "will GPU mining still be profitable?"

I'd like to get your opinion though, does this mean to you that the best time to sell GPUs would be today? I'm asking because i'm curious if we can still ride out our GPUs for 1 or 2 more months before the shit hits the fan. I'm pretty happy with ROI and all the coins i already have for long-term holding.

[martyroz]

Yes, but let's dust off the MagikSmoke 8 ball and look at the next few months:
-snip-

Great analysis and summary of GPU outlook overall. This should be stickied for all those still asking "will GPU mining still be profitable?"

I'd like to get your opinion though, does this mean to you that the best time to sell GPUs would be today? I'm asking because i'm curious if we can still ride out our GPUs for 1 or 2 more months before the shit hits the fan. I'm pretty happy with ROI and all the coins i already have for long-term holding.

It is a great analysis, for sure. But perhaps it's missing the most important and hardest to predict factor; forks.

This is also the key to your question. If all Equihash coins fork, nVidia family cards are back in business till 2019.

[dragonmike]

Yes, but let's dust off the MagikSmoke 8 ball and look at the next few months:
-snip-

Great analysis and summary of GPU outlook overall. This should be stickied for all those still asking "will GPU mining still be profitable?"

I'd like to get your opinion though, does this mean to you that the best time to sell GPUs would be today? I'm asking because i'm curious if we can still ride out our GPUs for 1 or 2 more months before the shit hits the fan. I'm pretty happy with ROI and all the coins i already have for long-term holding.

It is a great analysis, for sure. But perhaps it's missing the most important and hardest to predict factor; forks.

This is also the key to your question. If all Equihash coins fork, nVidia family cards are back in business till 2019.

...unless fpga's do it better.
Haven't yet read about any equihash forks, but arguably it's only a matter of time until a good fpga bitstream can be programmed for any of them.

[ATCkit2]

Currently GPU mining works, doesn't it? Despite the fact that Bitmain tries to ruin everything. You have GTX1080ti, why to sell them? Personally I would go 40% GPU, 60% ASIC (remember how profitable your GPU was some months ago).
What's about future, to my mind game industry mustn't be turned into mining industry so GPU's mustn't be used in mining anyway, my vote to asic for that.
And to be fair you don't need my advice but just stated my opinion, I look you from outside and you are amazing member!

Yes, but let's dust off the MagikSmoke 8 ball and look at the next few months:

Z9 will destroy Equihash in short order.  Probably will be a replay of the L3+.  There goes your 1080 Ti cash cow.

Baikal Giant does X11/Quark/Quibt/Myriad-Groestl/Nist5/Skein/X11Gost/ and maybe a few more in a few months.  At your doorstep in about 5 days for $1,600.

Tribus is number 1 on Whitefire's FPGA bitstream release list:

http://zetheron.com/index.php/fpga-performance-profit/

And Tribus can support about 130 FPGAs until profit goes to Bitmain ShiteHeaven.  And people are already mining Tribus using FPGAs, I mined it for quite some time with 1080 Ti's and it is a waste of money to even try anymore and Whitefire's bitstreams are not even fully out in the wild yet.

Blake variants have been destroyed by A3, Halong, BK-B.  Ditto for Decred.  And I haven't even received my damn Obelisk SIA miner yet!

A word about the whole FPGA thing:  if you follow all the Bitcoin threads and the discord, the FPGA roll out is shaping up in the *short* term to be every bit as difficult as the Vega's were and then some.  But it took what?  About three or four months for people to figure out ways to make the Vega profitable.  Yes, cooling mods required at first.  Yes, Senseless will not support Windows.  GPUHoarder is interested in cost-efficient accelerators.

But Vosk will be posting YouTube FPGA videos by Thanksgiving, if not before then.  You wait and see.

Cryptonight and all of its variants are squarely in the sights of the FPGA guys.  I will go on record to say right now Vega's won't be able to compete on Cryptonight come December.

That leaves X16 variants -- and those are next in line for FPGA domination, the only drawback is that you need to link a few FPGAs together and that won't happen immediately on a large, easy to implement scale.

And then we come to Eth hash.

That seems like a very narrow doorway for an awful lot of GPUs.

Please feel free to point out what I am missing and where I am wrong.  Unless the Volta is really cheap, I don't think it changes anything near term.  And Vega 20?  Who the hell knows, and even when it comes out, it will be every bit as unoptimized by the AMD genius techs as was Vega 10.  So, unless you are part of OhGodACartel/OhGodASlovenianMiningFarm/OhGodWhyDoIFixAMD?, good luck to you.

FWIW, Bitcoin Gold plans to have a new algo:

https://forum.bitcoingold.org/t/our-new-equihash-equihash-btg/1512

[DesertMiner]

Heads up CNv7's price spiked hard on Nicehash, could it be a 51% attack on some coin ? or is there a coin I'm missing?

[Mashy81]

Heads up CNv7's price spiked hard on Nicehash, could it be a 51% attack on some coin ? or is there a coin I'm missing?

Electroneum difficulty has finally dropped after the fork so I think they are hitting that

[lexele]

Heads up CNv7's price spiked hard on Nicehash, could it be a 51% attack on some coin ? or is there a coin I'm missing?

Could be Electronero, an electroneum fork.

[adaseb]

I have no idea if ETH will do anything but most likely since Bitmain is slashing these Equihash ASIC prices like this AND all those 51% attacks that happened last month... most Equihash coins will change their algo. At least modify the parameters to make it mineable only with 2.5GB GPUs like Bitcoin Gold proposses.

Then and only then we will finally see exactly how much dominance all these secret hidden ASICs are mining.

[philipma1957]

innosilicon  has announced a 50000 sol zec miner

uses only 620 watts.

10k  for 1

this will be the end of zec for my 1080ti's

3 1080tis  do 2100 sols at 600 watts


so  this gear is 25 x more efficient.

Buying 75 1080tis cost about 50k

this is 10 k

http://www.innosilicon.com/html/news/26.html

they will do smaller sales  with lower moq in august

at the moment moq is 100 at 10k = 1 million

so my l3+ earn about 2.80 a day  and burn about 2.89 a day due to my summer rates

May be shutting down  everything but sha 256

[dragonmike]

innosilicon  has announced a 50000 sol zec miner

uses only 620 watts.

10k  for 1

this will be the end of zec for my 1080ti's

3 1080tis  do 2100 sols at 600 watts


so  this gear is 25 x more efficient.

Buying 75 1080tis cost about 50k

this is 10 k

http://www.innosilicon.com/html/news/26.html

they will do smaller sales  with lower moq in august

at the moment moq is 100 at 10k = 1 million

2 months to get your investment back at today's profitability level, i.e. you'll be lucky to get your 10K back at all (considering many Equihash coins will fork).
So another case of the ASIC producer probably having used this for several months already, flogging it to the masses now with the consequence that Equihash will cease being profitable for everyone pretty soon and these machines being reduced to the role of doorstoppers within 6 months.

[Eyedol-X]

innosilicon  has announced a 50000 sol zec miner

uses only 620 watts.

10k  for 1

this will be the end of zec for my 1080ti's

3 1080tis  do 2100 sols at 600 watts


so  this gear is 25 x more efficient.

Buying 75 1080tis cost about 50k

this is 10 k

http://www.innosilicon.com/html/news/26.html

they will do smaller sales  with lower moq in august

at the moment moq is 100 at 10k = 1 million

so my l3+ earn about 2.80 a day  and burn about 2.89 a day due to my summer rates

May be shutting down  everything but sha 256

Well don't forget that multiple equihash coins have planned forks so this could all change quickly and these new asics could be door stops in a matter of months.

I shut down my L3+ a couple of weeks ago because of similar math

[melpheos]

innosilicon  has announced a 50000 sol zec miner

uses only 620 watts.

10k  for 1

this will be the end of zec for my 1080ti's

3 1080tis  do 2100 sols at 600 watts


so  this gear is 25 x more efficient.

Buying 75 1080tis cost about 50k

this is 10 k

http://www.innosilicon.com/html/news/26.html

they will do smaller sales  with lower moq in august

at the moment moq is 100 at 10k = 1 million

so my l3+ earn about 2.80 a day  and burn about 2.89 a day due to my summer rates

May be shutting down  everything but sha 256

Zec's choice to welcome (if there is no split in the dev team as it looks right now that there is some conflict) ASICs is a poor decision the same way SUMO did revert back to ASIC friendly.
Effectively a moq 100 units for $1 million is clearly against small farm and for centralization which is contrary to the main idea of ZEC.

[DesertMiner]

innosilicon  has announced a 50000 sol zec miner

uses only 620 watts.

10k  for 1

this will be the end of zec for my 1080ti's

3 1080tis  do 2100 sols at 600 watts


so  this gear is 25 x more efficient.

Buying 75 1080tis cost about 50k

this is 10 k

http://www.innosilicon.com/html/news/26.html

they will do smaller sales  with lower moq in august

at the moment moq is 100 at 10k = 1 million

so my l3+ earn about 2.80 a day  and burn about 2.89 a day due to my summer rates

May be shutting down  everything but sha 256

Kudos to the David Vorick for warning us about this earlier, here's a quote from his post at : https://blog.sia.tech/the-state-of-cryptocurrency-mining-538004a37f9b

"Needless to say, we weren’t the least bit surprised when Bitmain released powerful ASICs for equihash. The Bitmain ASICs are actually substantially less performant (5x to 10x) than our own internal study suggested they would be. There could be many reasons for this, but overall we think that it’s pretty reasonable to assume that more powerful equihash ASICs will be released in the coming months."

Nonetheless, I believe the 1080 ti's wont go out of business, in a couple of days LUX will be releasing their new algo, BTG will be forking and ETC (could be already done) will be removing the difficulty time bomb.

And I expect a lot of other coins will be forking too.