Author

Topic: MONERO adopts a new POW Algorithm to fight ASIC (Read 736 times)

full member
Activity: 1148
Merit: 132
The last miner hardware which was released or used to mine XMR last algo wasn't an ASIC. I am pretty sure it was an FGPA and from what I've read it was also developed in private.

Only way the developers knew its existence is from some 3rd party researcher that noticed anamolies in some of the block data and he posted his findings on medium.

They probably kept it a secret because they didn't want the XMR dev to fork the algo earlier however they were caught anyways. This was for an FGPA but an ASIC would have too long of a timeline and cost to make it feasible for any algo that XMR uses.

One of the reason monero has been unable to be a gpu profitable coin, too many fpga's on it and those people can change fast to whatever xmr algo it is using and reason investors are not buying xmr anymore.

hopefully randomx slows the fpga crowd for six months at least
sr. member
Activity: 2142
Merit: 353
Xtreme Monster
The last miner hardware which was released or used to mine XMR last algo wasn't an ASIC. I am pretty sure it was an FGPA and from what I've read it was also developed in private.

Only way the developers knew its existence is from some 3rd party researcher that noticed anamolies in some of the block data and he posted his findings on medium.

They probably kept it a secret because they didn't want the XMR dev to fork the algo earlier however they were caught anyways. This was for an FGPA but an ASIC would have too long of a timeline and cost to make it feasible for any algo that XMR uses.

One of the reason monero has been unable to be a gpu profitable coin, too many fpga's on it and those people can change fast to whatever xmr algo it is using and reason investors are not buying xmr anymore.
legendary
Activity: 3808
Merit: 1723
The last miner hardware which was released or used to mine XMR last algo wasn't an ASIC. I am pretty sure it was an FGPA and from what I've read it was also developed in private.

Only way the developers knew its existence is from some 3rd party researcher that noticed anamolies in some of the block data and he posted his findings on medium.

They probably kept it a secret because they didn't want the XMR dev to fork the algo earlier however they were caught anyways. This was for an FGPA but an ASIC would have too long of a timeline and cost to make it feasible for any algo that XMR uses.
full member
Activity: 980
Merit: 132
Well done guys who are struggling with asic, but most likely they will come up with a different coin mining algorithm: pos, dpos, since often changing algorithms is quite expensive for developers, since you need to rewrite the code for a new type of mining.
legendary
Activity: 2590
Merit: 1022
Leading Crypto Sports Betting & Casino Platform
no matter what they change, chinese can reprogram a new asic board with the correct code to work with the new monero algo, let alone fpga that can be reprogrammed again and again, even if this is memory intensive, monero is big enough to encourage the makiing of an asic for it

Its not that easy to reprogram a new ASIC board. Its needs extensive research to be effecient and needs to be fabricated at a plant. Last estimate I heard it was a few million at least just to get it fabricated. After you get it fabricated you can easily just put it into an existing ASIC board like an Antminer.

However keep in mind that the Miners revenue for XMR is about $90K a day according to,
https://bitinfocharts.com/monero/

If you look at ETH its $2 million and BTC is $13.7 million. Hence its not wise to make a new ASIC for their algo because there isn't enough money to be made, and also the risk that they might fork to a different algo is great. BTC most likely will never switch algos hence why money is spent on research and developement for Antminers instead of alt-coin ASICS.

chinese don't care too much about huge profit, they will follow every profit there is out there, take a look at baikal miner, they made an asic miner for every shit algo, even those that were almost dead, if there is profit, asic will come one day, you don't need bitcoin capitalization to develop an asic for an altcoin

chinese also can have chips and other stuff at cheaper price
sr. member
Activity: 857
Merit: 262
no matter what they change, chinese can reprogram a new asic board with the correct code to work with the new monero algo, let alone fpga that can be reprogrammed again and again, even if this is memory intensive, monero is big enough to encourage the makiing of an asic for it

Its not that easy to reprogram a new ASIC board. Its needs extensive research to be effecient and needs to be fabricated at a plant. Last estimate I heard it was a few million at least just to get it fabricated. After you get it fabricated you can easily just put it into an existing ASIC board like an Antminer.

However keep in mind that the Miners revenue for XMR is about $90K a day according to,
https://bitinfocharts.com/monero/

If you look at ETH its $2 million and BTC is $13.7 million. Hence its not wise to make a new ASIC for their algo because there isn't enough money to be made, and also the risk that they might fork to a different algo is great. BTC most likely will never switch algos hence why money is spent on research and developement for Antminers instead of alt-coin ASICS.

exactly. you actually need to develop a chip that will be efficient enough to perform a particular algo and then fab it and bring fast enough to market before algo changes - ASICs for monero becoming less realistic. btw what's up with CN8 asics or what was it?
legendary
Activity: 3808
Merit: 1723
no matter what they change, chinese can reprogram a new asic board with the correct code to work with the new monero algo, let alone fpga that can be reprogrammed again and again, even if this is memory intensive, monero is big enough to encourage the makiing of an asic for it

Its not that easy to reprogram a new ASIC board. Its needs extensive research to be effecient and needs to be fabricated at a plant. Last estimate I heard it was a few million at least just to get it fabricated. After you get it fabricated you can easily just put it into an existing ASIC board like an Antminer.

However keep in mind that the Miners revenue for XMR is about $90K a day according to,
https://bitinfocharts.com/monero/

If you look at ETH its $2 million and BTC is $13.7 million. Hence its not wise to make a new ASIC for their algo because there isn't enough money to be made, and also the risk that they might fork to a different algo is great. BTC most likely will never switch algos hence why money is spent on research and developement for Antminers instead of alt-coin ASICS.
member
Activity: 574
Merit: 14
ASIC's "undermining the stability and security of the network"... pffft! What a load of crap.
It is easier to concentrate power in the hands of a miners when you run on ASICS. Compare that to GPU farms or tokens you can mine with your laptop
sr. member
Activity: 857
Merit: 262
it makes perfect sense for monero. some coins should not be minable by "specialized" hw.
I'm also thinking of implications of the more cpu-centric algo from the perspective of "involuntary" mining, which is very popular on XMR
legendary
Activity: 2590
Merit: 1022
Leading Crypto Sports Betting & Casino Platform
no matter what they change, chinese can reprogram a new asic board with the correct code to work with the new monero algo, let alone fpga that can be reprogrammed again and again, even if this is memory intensive, monero is big enough to encourage the makiing of an asic for it
legendary
Activity: 3318
Merit: 1247
Bitcoin Casino Est. 2013
This is another great move from the Monero developers which has already done this in continuity.Randomx algorithm is a great one and I agree with them that ASICS should be kept off the network and a chance should be given to everyone including hobby home miners.
jr. member
Activity: 188
Merit: 2
In Blockchain we trust!
There are only one way to stay away "ASICs problem" - switching to any POS algo. That is only one way to progress, to all coins, and MONERO is not an exception.
legendary
Activity: 3808
Merit: 1723
GPUs however are pretty much also almost decentralized equipment since they are sold by lots of manufactures at various prices, most affordable and they are available pretty much anywhere in the world.

I do agree that CPU is actually the best way to be completely decentralized and GPUs a close second. Also consider that there are many devices like tablets and mobile phones which also have CPUs and might be able to also apply to the "One CPU One Vote" which was discussed in the original Satoshi whitepaper for Bitcoin.

We will see how this algo does within the next few months, would be nice if it actually worked and didn't force XMR to fork every few months the algo.
newbie
Activity: 5
Merit: 1
"Banning" Asics does not aid decentralization. There are rich farms who will  buy lots of VGA's than, if there is RandomX than they will invest in CPU's.


This is accurate. I think a lot of people will move to using companies like ours for their mining purposes. Essentially treating bitcoin miners just like server rack space. We're already doing both at our facilities, whatever the customer prefers to use our electricity for.

Right now if you have dedicated server equipment it's sometimes better to have it sent to an offsite location and managed there too. Even avoiding services like AWS so you can have something private and not being utilized by the Amazon overlords. We have Servers, ASICs and GPU miners at our locations. Love what Monero is doing but feel like it has more to do with preventing ASIC Manufacturers specifically from having a majority of hashing power and not mining operations around the world like ours which will continue to grow power capacity and operations.
legendary
Activity: 1512
Merit: 1011
"Banning" Asics does not aid decentralization. There are rich farms who will  buy lots of VGA's than, if there is RandomX than they will invest in CPU's.


You are right, there always will be "rich farms" and commercial entities in any industry.
Still, ASICs (war) was the reason to stop mining BTC years ago. So I fully support the Dev's decision. This will continue to allow small fishes like me to play as miners. And have some 'totally clean' coins.
legendary
Activity: 1470
Merit: 1114
So a RandomX ASIC or FPGA miner is just a CPU, compiler and runtime system on a chip that
serves a single purpose.

not the only difference, while I agree that all cpus and asics made from logic gates, but the difference could be significant in architectures of the chips.
take for example cpu and gpu, despite they all are used in one pc - their architecture differs so much as their performance
cpu have 4 huge cores while gpu have 8000 small cores - this is just an example.


If you need a processor you NEED a PROCESSOR, the architecture is secondary.

Edit: Existing algos don't need processors the algorithms are hard coded.
hero member
Activity: 756
Merit: 507
The thing is there is no way to stop ASICs or FPGA in this case. Period.

Techically correct. It will always be possible to build an ASIC or FPGA to do whatever a CPU can do
because they are all built from logic gates. The most significant difference between an ASIC and a CPU
is it's purpose.

RandomX is essentially a compiler, runtime system and processor all in one. It takes as input
a program which it compiles into its own native machine code then runs it to produce the hash.

So a RandomX ASIC or FPGA miner is just a CPU, compiler and runtime system on a chip that
serves a single purpose.

Seems simple enough. Or is it?


not the only difference, while I agree that all cpus and asics made from logic gates, but the difference could be significant in architectures of the chips.
take for example cpu and gpu, despite they all are used in one pc - their architecture differs so much as their performance
cpu have 4 huge cores while gpu have 8000 small cores - this is just an example.
hero member
Activity: 2352
Merit: 905
Metawin.com - Truly the best casino ever
"Banning" Asics does not aid decentralization. There are rich farms who will  buy lots of VGA's than, if there is RandomX than they will invest in CPU's.

Of course there are rich mining companies which will be able to buy thousands of GPUs and almost every equipment which will be needed for Monero mining but still that's a great news. Things won't be as easier for such companies as it was recently.
Hope eth will follow Monero, btw this coin is really one of the great and most underrated coin out there cause it truly demonstrates nature of crypto (Most anonymous, even better than bitcoin in this job, great tx fees and etc). Also developers don't hang on things too which is another bonus here.
member
Activity: 449
Merit: 24
keep your all eth coin trolls, the moment asics are banned from mining eth then eth price will be back to $ 1400 or higher maybe around $3000.

When will eth switch algos?  
full member
Activity: 584
Merit: 106
"Banning" Asics does not aid decentralization. There are rich farms who will  buy lots of VGA's than, if there is RandomX than they will invest in CPU's.
jr. member
Activity: 199
Merit: 1
when's the deadline?
legendary
Activity: 1470
Merit: 1114
The thing is there is no way to stop ASICs or FPGA in this case. Period.

Techically correct. It will always be possible to build an ASIC or FPGA to do whatever a CPU can do
because they are all built from logic gates. The most significant difference between an ASIC and a CPU
is it's purpose.

RandomX is essentially a compiler, runtime system and processor all in one. It takes as input
a program which it compiles into its own native machine code then runs it to produce the hash.

So a RandomX ASIC or FPGA miner is just a CPU, compiler and runtime system on a chip that
serves a single purpose.

Seems simple enough. Or is it?


sr. member
Activity: 2142
Merit: 353
Xtreme Monster
keep your all eth coin trolls, the moment asics are banned from mining eth then eth price will be back to $ 1400 or higher maybe around $3000.
sr. member
Activity: 784
Merit: 282
i'm afraid it won't help too much.. this is 3rd change of algorithm for Monero, and within 2-6 month they are facing again with the same problem : increasing numbers of asics or fgpas.
no idea how to fight it and eliminate the problem completely, maybe the strategy they choose - is the only one is possible in this situation.

i wonder, if it possible to change POW somehow that only high level computer language could described and solve it. so then asics will be out.


At least they are doing something and making progress. While the number of ASICs and FPGAs are increasing (for all coins and algorithms btw), the constant changes in algo make it less susceptible to such things even if it only happens every 2 to 6 months. Can you imagine what the network hash rate would be if Monero devs accepted defeat and never changed their algo?

I salute the Monero team and support their continued fight against centralized mining.
sr. member
Activity: 861
Merit: 281
The thing is there is no way to stop ASICs or FPGA in this case. Period.
After an algo change, the FPGA programmers start working to make it run under the new algo and if they succeed, they will start mining silently.
If FGPA is successful, then big companies like Bitmain or Innosilicon have so much capital that they can manufacture large scale ASICs for it.
I believe in time, ASICs can be manufactured for any algo and the only way to tackle this is to cycle a change of algo in a specific period of time.
Similar to what Monero is doing and believe me, there are doing one great of a job.
hero member
Activity: 756
Merit: 507
i'm afraid it won't help too much.. this is 3rd change of algorithm for Monero, and within 2-6 month they are facing again with the same problem : increasing numbers of asics or fgpas.
no idea how to fight it and eliminate the problem completely, maybe the strategy they choose - is the only one is possible in this situation.

i wonder, if it possible to change POW somehow that only high level computer language could described and solve it. so then asics will be out.
legendary
Activity: 3808
Merit: 1723
ETH Dev should learn something from Monero and do the same. Its been a few years already since the first ETH asics hit the market and there is no news of any potential algo switch. They did an audit and it came back in favor and we will see if they will actually take action. However from what I remember this was suppose to be implented in the last fork which is what the devs discussed at the beginning of the year.

Its too bad the overall daily miner revenue for XMR is low, would be a game changer otherwise. Either way I got some idle rigs and will switch over and see how long it stays profitable, should be interesting.

Regarding XMR, was it even ASICs which were dominating the network? I assumed it was just FGPA's. I don't think the ASICs had enough time to fabricate a new chip.?
legendary
Activity: 1820
Merit: 1092
~Full-Time Minter since 2016~
Very good to hear this
of course, miners will not be surprised, XMR has ALWAYS vowed to keep ahead of ASICs
So this is to be expected
 /clap 2 Fluffypony & Team, ive always liked that guy  Grin
member
Activity: 129
Merit: 17
ASIC's "undermining the stability and security of the network"... pffft! What a load of crap.
member
Activity: 574
Merit: 14



Monero is adopting a new Proof-of-Work algorithm to fight off ASIC miners from taking over the network and thereby undermining the stability and security of the network.
According to an article 
" Monero is planning to switch to an entirely different proof-of-work algorithm to remain resistant."

 RandomX will be introduced to replace the current CryptoNight Algorithm.


Credit : https://bitcoinist.com/monero-xmr-adopts-new-pow-algorithm-to-fight-off-asics/

Good move to enable smaller miners stay relevant in a fast changing ecosystem.
Jump to: