Pages:
Author

Topic: 8th Alt coin thread. Or what to do now that asics are all over the place. - page 98. (Read 81547 times)

sr. member
Activity: 1246
Merit: 274
I might have to look more closely at LUX with those proposed algorithm changes.  I tested that algorithm briefly at one point, just to gauge the hashrate for any profit calculations, but nothing beyond that.
legendary
Activity: 4256
Merit: 8551
'The right to privacy matters'
Just as an FYI, LUX has just pre-released our new PHI2 protocol that greatly improves on the original version Cheesy

PHI2 has been designed to be FPGA and ASIC resitant...and also includes a 30-60% power reduction, and lower temps.

Full release won't happen until we do a planned fork and introduce Segwit, Smart Contracts, POW/POS/MN reward changes and a few other things

Full details on our discord if anyone wants to check it out - https://discord.gg/Drbfbtz

Plenty more to come from the LUXcore team as well  Grin


The Luxcore concept is unique and, as I see it, will be a much needed feature in the crypto world. So this is really good news.

Food for thought-I wonder how Lux compares to Supernet (UNITY)? Guess I should pose this question on their threads.

 I Need to read discord closely
jr. member
Activity: 85
Merit: 6
Just as an FYI, LUX has just pre-released our new PHI2 protocol that greatly improves on the original version Cheesy

PHI2 has been designed to be FPGA and ASIC resitant...and also includes a 30-60% power reduction, and lower temps.

Full release won't happen until we do a planned fork and introduce Segwit, Smart Contracts, POW/POS/MN reward changes and a few other things

Full details on our discord if anyone wants to check it out - https://discord.gg/Drbfbtz

Plenty more to come from the LUXcore team as well  Grin


The Luxcore concept is unique and, as I see it, will be a much needed feature in the crypto world. So this is really good news.

Food for thought-I wonder how Lux compares to Supernet (UNITY)? Guess I should pose this question on their threads.
legendary
Activity: 4354
Merit: 3614
what is this "brake pedal" you speak of?
i wonder what the cards to keep are.. a few of each class of card? or go mostly with one class? or just go by power efficiency, ie nvidia at the moment. since most algos seem to run better on some class of card, how many bases to cover?

grouped by memory type

nvidia:
-------
GDDR5X    1080ti
GDDR5      1070ti

amd
----
GDDR5      480/580  <- unsure on these, nvidia 1070ti beats them on power efficiency in most algos?
HBM2        vega56


I have 16 1080ti's

and 1 rx560
and 1 1050ti

I am banking on the 1080 ti's hanging in there for a while.

most of my hashing is sha 256 and script.

yeah 1080ti is a keeper for sure.

the 1080ti and vega 56/64 seem the main cards to keep if the goal is to simplify the rig. between them they are pretty good (for gpus) at anything.

never got into asic mining beyond a bfl jalapeno (got one of the 1st which actually made me decent money fiat wise, but not btc wise). after that the asics were too power hungry and i was afraid of being burned, so i was not mining anything (even gpu wise) for a couple years.

wish id paid more attention those two years, i basically ignored mining and just hodled instead of trying to use btc to expand into asics. which worked out in many ways actually.
legendary
Activity: 4256
Merit: 8551
'The right to privacy matters'
even if gpu mining goes to more or less zero profit i think there will always be a use for a decent "standby" gpu rig (something thats already ROId).. a rig thats kept updated and ready to roll for any new algos that come along, just download a miner and go. theory being any really new algo wont have asics for a few months. get in early, hash with gpu till the big stuff appears, then back to standby status.

This is a good strategy that i would also like to follow through on. I'm currently on the rocks about selling off some of my GPUs and just keeping a reliable set of 1070s and 1080tis. As long as they are turning a profit above electricity costs they will continue to mine. Then when all goes to shit with ASICs, they will be on standby for new coins with new algorithms.

Thanks for the idea bud.

i wonder what the cards to keep are.. a few of each class of card? or go mostly with one class? or just go by power efficiency, ie nvidia at the moment. since most algos seem to run better on some class of card, how many bases to cover?

grouped by memory type

nvidia:
-------
GDDR5X    1080ti
GDDR5      1070ti

amd
----
GDDR5      480/580  <- unsure on these, nvidia 1070ti beats them on power efficiency in most algos?
HBM2        vega56

 



I have 16 1080ti's

and 1 rx560
and 1 1050ti

I am banking on the 1080 ti's hanging in there for a while.

most of my hashing is sha 256 and script.
legendary
Activity: 4354
Merit: 3614
what is this "brake pedal" you speak of?
even if gpu mining goes to more or less zero profit i think there will always be a use for a decent "standby" gpu rig (something thats already ROId).. a rig thats kept updated and ready to roll for any new algos that come along, just download a miner and go. theory being any really new algo wont have asics for a few months. get in early, hash with gpu till the big stuff appears, then back to standby status.

This is a good strategy that i would also like to follow through on. I'm currently on the rocks about selling off some of my GPUs and just keeping a reliable set of 1070s and 1080tis. As long as they are turning a profit above electricity costs they will continue to mine. Then when all goes to shit with ASICs, they will be on standby for new coins with new algorithms.

Thanks for the idea bud.

i wonder what the cards to keep are.. a few of each class of card? or go mostly with one class? or just go by power efficiency, ie nvidia at the moment. since most algos seem to run better on some class of card, how many bases to cover?

grouped by memory type

nvidia:
-------
GDDR5X    1080ti
GDDR5      1070ti

amd
----
GDDR5      480/580  <- unsure on these, nvidia 1070ti beats them on power efficiency in most algos?
HBM2        vega56

 

full member
Activity: 420
Merit: 184
One of my buddies just dropped off 2 units of X3 in my test lab - testing mining BCN and ETN.
...
Just sharing testing the soon to be paperweights - dont kill the messenger :-)

Point them at DERO; a few weeks ago I never would have recommended this, but network hashrate has eclipsed 300MH/s (ie - within spitting distance of XMR) so it's kind of pointless to warn people off now.

At least you'll be mining a coin that is likely to be worth more in the future (BCN and ETN don't exactly have the best prospects, you have to admit).

legendary
Activity: 1834
Merit: 1080
---- winter*juvia -----
One of my buddies just dropped off 2 units of X3 in my test lab - testing mining BCN and ETN.

He doesnt have the right PSUs, and need to borrow my Delta-2400w-94%-Platinum.

220-237khs per unit stable; Running 2 x X3, around 470-540khs range at 500-520 watts ! 400mhz stock settings.

Thats like over 60 x RX480s to get that kind of hash.

At 500mhz settings, I was able to get them up to 630khs for both total, 60c stable. Watt-meter says 1200w.

550mhz settings is still good, a bit louder as the fans work harder.  

Anything beyond that settings, the X3 doesnt even hash.

Its doesntwork with vanilla NH CN it just spits rejects.

Just sharing testing the soon to be paperweights - dont kill the messenger :-)
newbie
Activity: 2
Merit: 2
On a sidenote, since we're talking about power, my provider here in London (British Gas) was going to increase prices to over £0.16/KWh (that's US$ 0.22!)... so I'm in the process of switching. That'll bring me back down to £0.11 ($0.15). Every little helps!

That's interesting, I've been on £0.1282 with BG, they now want £0.16 as they did from you. Best I've been quoted is £0.1321 ($.18) from Scottish Power ... out of interest, who are you switching to?




I am not sure if I can help, but here in Somerset, I spend a flat £0.09029 ($0.12 based on the actual rate) with OTM (I'm not sure I can write down the full name of the Provider); and best/worst of all there's no contract, the tariff is variable, you can  switch anytime.

I hope I can help, I learned so much from this forum, thank you!
member
Activity: 104
Merit: 11
Phil, had a chance to look at the algo chances suggested by @ohgodagirl? interesting stuff
Just in case that someone doesnt know whats this about here is the link where you can check it out
https://twitter.com/OhGodAGirl/status/996813743102283776

their group is the one who released the "ETH largement pill" for cards with gddr5x...they know how to utilize gddr5x gpu rams......1080ti's here we go again! hehe

I spoke to the Dev of progpow and I love their approach. It's not about ASIC resistance but closing the loopholes that make ASICS faster and more efficient. The proposal was posted on the ETH GitHub and I strongly suspect that people were bought by ASIC manufacturers, just read the comments... It's sad. However they have a working algo and miner so anyone could fork and start using it. Unfortunately x16r is not exactly ASIC resistant either, there is a whole list of projected ASIC speed gains over GPUs and yes, we are screwed with the current algos. As much as I don't like ethash I gotta admit it was still the most ASIC resistant algo. (Only 2x gain on ASIC)
Hash rates with progpow:

RX580 | 9.4 | | Vega56 | 16.6 | | Vega64   | 18.7 | | GTX1070Ti | 13.1 | | GTX1080   | 14.9 | | GTX1080Ti | 21.8 |

There is hope for the resistance!
legendary
Activity: 3444
Merit: 1061
Phil, had a chance to look at the algo chances suggested by @ohgodagirl? interesting stuff
Just in case that someone doesnt know whats this about here is the link where you can check it out
https://twitter.com/OhGodAGirl/status/996813743102283776

their group is the one who released the "ETH largement pill" for cards with gddr5x...they know how to utilize gddr5x gpu rams......1080ti's here we go again! hehe
e97
jr. member
Activity: 58
Merit: 1
Not sure if folks have seen this: "DeepMind AI Reduces Google Data Centre Cooling Bill by 40% "https://deepmind.com/blog/deepmind-ai-reduces-google-data-centre-cooling-bill-40/

They use machine learning over complex workloads however, I've seen miners do it with a basic three tier system: peak, off-peak, night-time. then its a matter of adjusting (preferably automatically) your overclock / voltage adjustment to your tier to maximize power efficiency based on power rates.
full member
Activity: 1148
Merit: 132
Just as an FYI, LUX has just pre-released our new PHI2 protocol that greatly improves on the original version Cheesy

PHI2 has been designed to be FPGA and ASIC resitant...and also includes a 30-60% power reduction, and lower temps.

Full release won't happen until we do a planned fork and introduce Segwit, Smart Contracts, POW/POS/MN reward changes and a few other things

Full details on our discord if anyone wants to check it out - https://discord.gg/Drbfbtz

Plenty more to come from the LUXcore team as well  Grin

nice i have some gpu mining lux too
legendary
Activity: 4256
Merit: 8551
'The right to privacy matters'
Got real world busy today.

Trying to hire a bathroom contractor. Also doing spring cleaning.

I am playing with a smaller coin. Mining and holding it.
jr. member
Activity: 140
Merit: 2
Phil, had a chance to look at the algo chances suggested by @ohgodagirl? interesting stuff
Just in case that someone doesnt know whats this about here is the link where you can check it out
https://twitter.com/OhGodAGirl/status/996813743102283776

To add--BBT's live stream recording w/ OGAG in chat is here:

https://www.twitch.tv/videos/262095416

Even if Eth ignores it some other ethash coins have expressed interest, a good sign less than 24 hours after release
full member
Activity: 1148
Merit: 132
Phil I'm curious what's with the thinking that GPU mining is dying?

My RX500 series are easily pulling in >$1.00 per GPU per day after power.

My Vegas are >3x that

My 1080tis make $3-$-4 per day per after power

I'm pretty happy.

The problem is not NOW, we are good

But in a few weeks or months we are really screwed with Asics and FPGAs

we will be fine x16r and X16s as well as c7 will be asic free at least till oct nov
then cryptonite coins will be forking again anyway
sr. member
Activity: 349
Merit: 250
Just as an FYI, LUX has just pre-released our new PHI2 protocol that greatly improves on the original version Cheesy

PHI2 has been designed to be FPGA and ASIC resitant...and also includes a 30-60% power reduction, and lower temps.

Full release won't happen until we do a planned fork and introduce Segwit, Smart Contracts, POW/POS/MN reward changes and a few other things

Full details on our discord if anyone wants to check it out - https://discord.gg/Drbfbtz

Plenty more to come from the LUXcore team as well  Grin
member
Activity: 154
Merit: 10
Phil, had a chance to look at the algo chances suggested by @ohgodagirl? interesting stuff
Just in case that someone doesnt know whats this about here is the link where you can check it out
https://twitter.com/OhGodAGirl/status/996813743102283776
jr. member
Activity: 225
Merit: 1
I don't see how fpga are going to be as good as they say. They are super expensive already and as soon as someone developes some software to use them their price will double or triple just like the gpu boom.
Anyway gpu prices are in free fall. Rx570 prices have come down $100 in the last w weeks here in Australia.
sr. member
Activity: 1246
Merit: 274
What is the advantage of the natural gas generator?  Is it cheaper to power rigs via natural gas generator than with regulator electric?

You can get natural gas for as low as $100 per 1000 cubic meters, it is usually 10 000 kW*hours - $0.01 per kW*hour
Modern high-power turbine generators go as far as 42% efficiency so we can expect something around $0.03 per kW*hour
Not bad at all. But you should seriously consider how long it will take to ROI. If you have some need for secondary heat output (say hot-water for sale) it helps a lot in terms of ROI. Unfornutaly at the moment we got $300 for natural gas locally and $0.05 for electricity so I have no chance to proof my studies for-real :-(.

In our case the primary reason is that we have free heat for our house from the natural gas well in the woods behind us. We need to determine if it would be possible to run both the furnace in the house and the generator for mining power at the same time or not.... also we need to get a quote for installation costs.  Before all of that though, we have some repairs needed for our septic system, and if that ends up being a completely new septic we won't be able to do that and the natural gas generator in the same year.
Pages:
Jump to: