Ethereum Anti-ASIC fork, is it the right time for bitcoin too?

[Wind_FURY]

the real reason behind ProgPoW is to eliminate bad actors from Bitmain
It's much more accessible to make an 'arms race' using GPU than having to rely on buying chips from China

What assures everyone that there won't be any more efficient hardware developed for ProgPow. It would bring Bitcoin back to step one again, instead of staying with SHA256 to make it easier for different companies to make ASICs, and make mining hardware creation more distributed.

IMO this will be an inevitable step on bitcoin's road map

Inevitable is the fork-causing chain-split. But there is Bitcoin Gold. Plead them to experiment with ProgPow.

Ethereum and other coins might be the first to experiment with, but it will eventually make it's way into bitcoin's code

[1714717388]

1714717388

[1714717388]

1714717388

[1714717388]

1714717388

[1714717388]

1714717388

[1714717388]

1714717388

[1714717388]

1714717388

[kano]

I believe Charlie Lee used Scrypt because of Tenebrix's claims of its resistance to GPU, FPGA, and ASICs? How was it Charlie Lee's fault that Scrypt was not resistant to the development, and creation of specialized hardware for it?

...

Scrypt has setting that decide how much resources are required.
The Scypt setting used in Litcoin were too low and allowed it to fit in a GPU.

Charlie was GPU mining long before the GPU code was released.

I posted about this before Litecoin started in the Tenebrix thread:
https://bitcointalk.org/index.php?topic=45849.msg548778#msg548778
September 28, 2011, 03:35:01 PM

There are other comments there about it.

Litcoin launch post was: October 09, 2011, 06:14:28 AM
https://bitcointalk.org/index.php?topic=47417.0

He knew exactly what he was doing ... and even made comment about Tenebrix being a good choice of algorithm in the LTC post.

...
Proof of Work

We really liked Tenebrix's Scrypt proof of work.
...

[kano]

@WIND_FURY

We have been thru this for a while, your assumption about ASICs-eventually-win is false and an advertised hype. There is no way to make FETCH_MEM operations to be ASIC friendly, no scientific way, not speaking of magical and mysterious powers of Bitmain  

Above-thread I've extensively discussed why and how ProgPow is provably safe against ASIC optimizations higher than 1.2x and I've provided supplementary resources as well, why don't you bother researching a bit more instead of rehashing Bitmain ads ever and ever?

You are missing a rather obvious point ...

Some recent miners cost as much as $1000 or more.
So, what stops someone making an 'ASIC' that has the necessary hardware to match some other PoW similar to CPU+RAM+GPU
Of course it would have aspects similar to a full computer, but would end up costing less, and well, that's not a sudden jump in cost or price vs current Bitcoin ASIC miners.

Sorry,but I overlooked this post. Your argument is wrong, Bitmain just can't make a more specialized thing for ProgPow without ending in a complete gpu technology development project and competing with AMD, and Nvidia it is the same for RAM and memory bus architecture.
...

You keep saying Bitmain ... there are many better, faster, more power efficient miners available than Bitmain miners.
Bitmain lost their main hardware guy years ago, that's why they've produced nothing new for years since the S9, until the recent releases.
He now runs Poolin after the contractual agreements finally expired.

You clearly have some weird idea about what is in RAM and a GPU ... and how no one could replace the simple hardware in a GPU with other designed hardware customised for any PoW.
Unless your PoW uses absolutely every piece of circuitry in a GPU (which nothing does) you certainly cannot argue something cannot be replicated and made more efficient (though it probably could anyway even if it did)

However, ASIC SHA256 miners use way more power and require very specific design constraints that the GPU companies clearly cannot match.
If making miners was so much easier than a GPU, AMD and nVidia would already have been doing it - Bitmain clearly made a crap ton of money (and profit) that can be compared to the GPU producers.

[aliashraf]

You keep saying Bitmain ... there are many better, faster, more power efficient miners available than Bitmain miners.
Bitmain lost their main hardware guy years ago, that's why they've produced nothing new for years since the S9, until the recent releases.
He now runs Poolin after the contractual agreements finally expired.

I don't believe you here. S9 was out like 3 years ago and now they've launched S11 and S15, the delay is not due to lack of expertise or some genius engineer leaving a multi-billion dollars firm, nonsense.

Going down from 28 nm tech to 14 was hard enough and getting to 7-8 nm was a nightmare. It is not about design it is about chip making industry, Bitmain designs shits and orders chips to its manufacturer.

Plus, Jihan is smart, he knows how sensitive is bitcoin community about monopoly, and it is absolutely reasonable for Bitmain to maneuver in a sophisticated way to hide their dominance but it is real and undisputable: Bitmain is the dominant party in crypto ASIC industry.

You clearly have some weird idea about what is in RAM and a GPU ... and how no one could replace the simple hardware in a GPU with other designed hardware customised for any PoW.
Unless your PoW uses absolutely every piece of circuitry in a GPU (which nothing does) you certainly cannot argue something cannot be replicated and made more efficient (though it probably could anyway even if it did)

Your idea of modern gpu and RAM architecture, looks to me pretty much weird too
GPU/RAM design/manufacturing industry has their own markets and rules and competition sphere is overloaded already, believe me  

There is no need for utilizing absolutely every piece of circuitry in a GPU, we are engineers and would be  satisfied with %80.
Let Bitmain or any other cracker try making a multi thousand core thing that  saturates a HBM memory bus and its cores simultaneously while running ProgPow and making it more efficient and/or cheaper than AMD by eliminating the unused 20% percent circuits. Little chance to succeed tho, but what would be the gains(just in case)?
A semi-gpu with 20% better price/efficiency and no application other than mining! Does it worth shit? A S15 is thousands of times more efficient than a gpu based system with comparable price, remember?

It is what I mean when I say ProgPoW is  ASIC-proof, it is provably safe against ASIC attack not because because it is impossible to make a hypothetical ASIC chip with a slightly (1.2x) better performance, (of course it is possible) it is secure against such an attack in the same way that bitcoin is secure: game theory and rational behavior of parties guarantees its security.

Crypto is not a mere engineering/programming domain, mathematics and socioeconomics are kings here, ProgPoW is secure by virtue of its brilliant game theoretic concept: making it infeasible for ASIC attacks to be started in the first place by de-incentivizing such attempts.

However, ASIC SHA256 miners use way more power and require very specific design constraints that the GPU companies clearly cannot match.
If making miners was so much easier than a GPU, AMD and nVidia would already have been doing it - Bitmain clearly made a crap ton of money (and profit) that can be compared to the GPU producers.

AMD/NVIDIA are leaders in gaming market which is a 200 billion industry that is doubling every year and is predicted to keep the pace for near future, they just don't take mining that serious.  

[philipma1957]

Given the size of the industry nowadays I also have my doubts that a mere switch to GPU / CPU mining would help decentralization all that much; at least not without additional measures to make pool mining significantly less attractive.

This seems to be the important question. What tangible benefits can we expect from ensuring the viability of GPU or FPGA mining? At large scale, how different are the economics of general purpose vs. application-specific processors? In both cases, it seems like mining hardware production would still be fairly concentrated among a few giant producers. Also, if we look to historical pool hash rate distribution of, for example, Ethereum vs. Bitcoin, the former is not necessarily less concentrated. So, what's the goal here?

I think a lot of people look at the advent of ASICs in 2013 and the subsequent rise of Bitmain, and they automatically blame ASICs for how the industry developed. However, I think a lot of it has to do with industry consolidation and concentration of capital that you'd see develop in any growing market over time.

Gpus are widely adapted many many many gamers.  not to many many many asic miners.

I look at this as an economical issue not a tech issue.

So I will not be as tech savvy as most on this thread.
I think  a fusion of sha256 with ProgPow  would be a more elegant solution then an either or either.

Simply make a way to mine btc with:
sha256 with an asic  be paid whatever
ProgPow with a gpu  be paid whatever
or mine with both an asic and a gpu be paid a bit better.

the bonus payment for mixed mining would be subject to some type of proof of gear linked method

But frankly  guys with  10,000 S15's won't like this idea.
and guys with 10,000 gpus won't like this idea.
only a mixed mining farm would say yes to it.

[Wind_FURY]

Crypto is not a mere engineering/programming domain, mathematics and socioeconomics are kings here, ProgPoW is secure by virtue of its brilliant game theoretic concept: making it infeasible for ASIC attacks to be started in the first place by de-incentivizing such attempts.

I was hoping the truth to ASIC-resistance would purely be technical. Because if you included socio-economics, socio-politics, and game theory into the debate, it would only be a question of incentives for an ASIC company.

I believe an ASIC company will build dedicated hardware, if the coin is successful enough, and when the price of the coin is right. If it pays to build specialized hardware, then someone will.

[aliashraf]

Crypto is not a mere engineering/programming domain, mathematics and socioeconomics are kings here, ProgPoW is secure by virtue of its brilliant game theoretic concept: making it infeasible for ASIC attacks to be started in the first place by de-incentivizing such attempts.

I believe an ASIC company will build dedicated hardware, if the coin is successful enough, and when the price of the coin is right. If it pays to build specialized hardware, then someone will.

It is not about price of the coin.

It is about mining profitability which is and remain marginal with the exception of short periods of price surge. Even in the case of a permanent highly profitable mining industry, costs of designing and manufacturing a special gpu-like hardware with 20% advantage, won't be compensated by the profits, as gpu industry would be competing with the manufacturer at the same time to produce a more powerful real gpu.

[philipma1957]

Crypto is not a mere engineering/programming domain, mathematics and socioeconomics are kings here, ProgPoW is secure by virtue of its brilliant game theoretic concept: making it infeasible for ASIC attacks to be started in the first place by de-incentivizing such attempts.

I was hoping the truth to ASIC-resistance would purely be technical. Because if you included socio-economics, socio-politics, and game theory into the debate, it would only be a question of incentives for an ASIC company.

I believe an ASIC company will build dedicated hardware, if the coin is successful enough, and when the price of the coin is right. If it pays to build specialized hardware, then someone will.

aliashraf mentioned a transitional asic to gpu transitional idea earlier in the thread.

I riffed on it saying  don't remove ASICs and sha256 figure how to blend it with ProgPow  to mine a coin from now on.

with 3 tiers
all asic
all gpu
a blend

this allows for the gamers which are many to coexist with the asic miners.

best of all asic builder have no incentive to crack ProgPow  as the most profitable way to mine is an asic sha256 such as bitmain's s15 linked to say a nvidia 2080ti doing ProgPow

there must be a way to link or pair the two pieces of gear to one another so that the pool mining sees the combined algos are linked to 1 of each piece of gear.

kind of  like 2 ssds or 2 hdds can be linked  into  a raid.  Once  done  speed and size in a striped raid are faster and bigger, but you could still break the pair up and of course sacrifice both speed and size.

As i said my technical expertise is meh on how to do that idea but if you did it asic builders lose the incentive to beat the gpu ProgPow incentive.

I suppose an identifier/ tag on each gpu and each asic would be needed to allow the linking of the two
pieces of hard ware for the better payout .  that would not be so easy and you would need to prevent faking the gpu with a copycat asic.  would be nice if technically possible

as bitmain/avalon/bitfury/whatsminer/innosilicon    would all have an incentive to partner with amd/nvidia/intel    rather then attack each other.

but what the heck I just a guy typing on a keyboard in NJ and my tech skills at math and coding are not those of OhGodaGirl or Kano or -ck

Still ould be nice to see some joint effort among the builders.  

They keep missing the big picture :

using this tech to better the world and make money at the same time.

[kano]

...

However, ASIC SHA256 miners use way more power and require very specific design constraints that the GPU companies clearly cannot match.
If making miners was so much easier than a GPU, AMD and nVidia would already have been doing it - Bitmain clearly made a crap ton of money (and profit) that can be compared to the GPU producers.

AMD/NVIDIA are leaders in gaming market which is a 200 billion industry that is doubling every year and is predicted to keep the pace for near future, they just don't take mining that serious.  

I'll take that to mean the rest of your arguments are bullshit also ...

https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-fourth-quarter-and-fiscal-2018

Edit:
That says nVidia was 9-10 Billion
So you could maybe argue 20 Billion

This: http://fortune.com/2018/07/30/bitmain-valuation-profits/

says Bitmain was 2-3 Billion
Bitmain was not the only one to sell miners in 2018 there are other manufacturers.

[aliashraf]

...

However, ASIC SHA256 miners use way more power and require very specific design constraints that the GPU companies clearly cannot match.
If making miners was so much easier than a GPU, AMD and nVidia would already have been doing it - Bitmain clearly made a crap ton of money (and profit) that can be compared to the GPU producers.

AMD/NVIDIA are leaders in gaming market which is a 200 billion industry that is doubling every year and is predicted to keep the pace for near future, they just don't take mining that serious.  

I'll take that to mean the rest of your arguments are bullshit also ...

https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-fourth-quarter-and-fiscal-2018

َAnd I'll take that you are not able to read, not a big surprise for a pool operator  

I'm talking about gaming industry (not gpu market), I say gaming industry is an order of magnitude bigger and more promising than mining industry Nvidia with like $10B revenue is active in a $200B industry and won't jeopardize this for an already destroyed (thanks to your friend, Jihan) mining industry, got it?

[kano]

...

However, ASIC SHA256 miners use way more power and require very specific design constraints that the GPU companies clearly cannot match.
If making miners was so much easier than a GPU, AMD and nVidia would already have been doing it - Bitmain clearly made a crap ton of money (and profit) that can be compared to the GPU producers.

AMD/NVIDIA are leaders in gaming market which is a 200 billion industry that is doubling every year and is predicted to keep the pace for near future, they just don't take mining that serious.  

I'll take that to mean the rest of your arguments are bullshit also ...

https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-fourth-quarter-and-fiscal-2018

َAnd I'll take that you are not able to read, not a big surprise for a pool operator  

I'm talking about gaming industry (not gpu market), I say gaming industry is an order of magnitude bigger and more promising than mining industry Nvidia with like $10B revenue is active in a $200B industry and won't jeopardize this for an already destroyed (thanks to your friend, Jihan) mining industry, got it?

Lol - change your argument when you are caught out

[aliashraf]

...

However, ASIC SHA256 miners use way more power and require very specific design constraints that the GPU companies clearly cannot match.
If making miners was so much easier than a GPU, AMD and nVidia would already have been doing it - Bitmain clearly made a crap ton of money (and profit) that can be compared to the GPU producers.

AMD/NVIDIA are leaders in gaming market which is a 200 billion industry that is doubling every year and is predicted to keep the pace for near future, they just don't take mining that serious.  

I'll take that to mean the rest of your arguments are bullshit also ...

https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-fourth-quarter-and-fiscal-2018

َAnd I'll take that you are not able to read, not a big surprise for a pool operator  

I'm talking about gaming industry (not gpu market), I say gaming industry is an order of magnitude bigger and more promising than mining industry Nvidia with like $10B revenue is active in a $200B industry and won't jeopardize this for an already destroyed (thanks to your friend, Jihan) mining industry, got it?

Lol - change your argument when you are caught out

My argument is quoted by you isn't it?

[kano]

...

However, ASIC SHA256 miners use way more power and require very specific design constraints that the GPU companies clearly cannot match.
If making miners was so much easier than a GPU, AMD and nVidia would already have been doing it - Bitmain clearly made a crap ton of money (and profit) that can be compared to the GPU producers.

AMD/NVIDIA are leaders in gaming market which is a 200 billion industry that is doubling every year and is predicted to keep the pace for near future, they just don't take mining that serious.  

I'll take that to mean the rest of your arguments are bullshit also ...

https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-fourth-quarter-and-fiscal-2018

َAnd I'll take that you are not able to read, not a big surprise for a pool operator  

I'm talking about gaming industry (not gpu market), I say gaming industry is an order of magnitude bigger and more promising than mining industry Nvidia with like $10B revenue is active in a $200B industry and won't jeopardize this for an already destroyed (thanks to your friend, Jihan) mining industry, got it?

Lol - change your argument when you are caught out

My argument is quoted by you isn't it?

If you want to jump to saying that the total gaming industry revenue including selling consoles and games and all the online stuff is relevant deciding if nVidia or AMD would bother to make miners, then I guess the total Bitcoin industry is what you are failing to compare against.

In which case your argument still loses:

~2 Billion in bitcoin mined last financial year (though it was a lot higher, I'm just using today's value)
~2-3 Billion in miners sold by one company (Bitmain) I've no idea how much was sold by the other companies ...
~2 Trillion in trading on exchanges
https://bitcoinist.com/bitcoin-trading-volume-2-trillion/

OK, so I'm not sure what you are comparing then ... ... ... ...

[aliashraf]

...

However, ASIC SHA256 miners use way more power and require very specific design constraints that the GPU companies clearly cannot match.
If making miners was so much easier than a GPU, AMD and nVidia would already have been doing it - Bitmain clearly made a crap ton of money (and profit) that can be compared to the GPU producers.

AMD/NVIDIA are leaders in gaming market which is a 200 billion industry that is doubling every year and is predicted to keep the pace for near future, they just don't take mining that serious.  

I'll take that to mean the rest of your arguments are bullshit also ...

https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-fourth-quarter-and-fiscal-2018

َAnd I'll take that you are not able to read, not a big surprise for a pool operator  

I'm talking about gaming industry (not gpu market), I say gaming industry is an order of magnitude bigger and more promising than mining industry Nvidia with like $10B revenue is active in a $200B industry and won't jeopardize this for an already destroyed (thanks to your friend, Jihan) mining industry, got it?

Lol - change your argument when you are caught out

My argument is quoted by you isn't it?

If you want to jump to saying that the total gaming industry revenue including selling consoles and games and all the online stuff is relevant deciding if nVidia or AMD would bother to make miners, then I guess the total Bitcoin industry is what you are failing to compare against.

In which case your argument still loses:

~2 Billion in bitcoin mined last financial year (though it was a lot higher, I'm just using today's value)
~2-3 Billion in miners sold by one company (Bitmain) I've no idea how much was sold by the other companies ...
~2 Trillion in trading on exchanges
https://bitcoinist.com/bitcoin-trading-volume-2-trillion/

OK, so I'm not sure what you are comparing then ... ... ... ...

Bitcoin Trading statistics in exchanges are both fake and irrelevant. Other figures conveniently show how small is the whole cryptocurrency market (total stakes involved) compared to gaming industry not mentioning uncertainties and institutional ambiguities.

What I'm saying is, Nvidia/AMD won't jeopardise their reputation and market share to act maliciously in crypto ecosystem. They won't run private mining farms with secret gpus for few more bucks and risk conflict of interests accusations and its legal consequences. On the other side Bitmain and its clones show no obligation to such a rule.

Actually an ASIC manufacturer has no obligation to sell its product, Bitmain cleverly did this to keep bitcoin in a sublime status: neither dead, nor alive. It was necessary for Bitmain to keep milking us.

[kano]

...
Bitcoin Trading statistics in exchanges are both fake and irrelevant. Other figures conveniently show how small is the whole cryptocurrency market (total stakes involved) compared to gaming industry not mentioning uncertainties and institutional ambiguities.
...

Lol, ignore it coz it doesn't suit your argument

...
What I'm saying is, Nvidia/AMD won't jeopardise their reputation and market share to act maliciously in crypto ecosystem. They won't run private mining farms with secret gpus for few more bucks and risk conflict of interests accusations and its legal consequences. On the other side Bitmain and its clones show no obligation to such a rule.

Actually an ASIC manufacturer has no obligation to sell its product, Bitmain cleverly did this to keep bitcoin in a sublime status: neither dead, nor alive. It was necessary for Bitmain to keep milking us.
...

Sigh you really don't know much about the mining industry do you.

Canaan existed well before Bitmain and made FPGA miner and of course made ASIC miners before Bitmain.

As for calling every other mining manufacturer a clone, then I guess ATI (bought by AMD) must be a clone of nVidia right?

Making the grossly incorrect assumption that all other miner manufacturers act the same as Bitmain clearly shows you don't know what's going on.

[aliashraf]

...
What I'm saying is, Nvidia/AMD won't jeopardise their reputation and market share to act maliciously in crypto ecosystem. They won't run private mining farms with secret gpus for few more bucks and risk conflict of interests accusations and its legal consequences. On the other side Bitmain and its clones show no obligation to such a rule.

Actually an ASIC manufacturer has no obligation to sell its product, Bitmain cleverly did this to keep bitcoin in a sublime status: neither dead, nor alive. It was necessary for Bitmain to keep milking us.
...

Sigh you really don't know much about the mining industry do you.

Canaan existed well before Bitmain and made FPGA miner and of course made ASIC miners before Bitmain.

As for calling every other mining manufacturer a clone, then I guess ATI (bought by AMD) must be a clone of nVidia right?

Making the grossly incorrect assumption that all other miner manufacturers act the same as Bitmain clearly shows you don't know what's going on.

No I don't and don't want to know more about what you call "mining industry" and I call ASIC manufacturing industry.

Actually, I've learned anything I need about this "industries" for not buying your sanitizing propaganda about them.

[vapourminer]

best of all asic builder have no incentive to crack ProgPow  as the most profitable way to mine is an asic sha256 such as bitmain's s15 linked to say a nvidia 2080ti doing ProgPow

there must be a way to link or pair the two pieces of gear to one another so that the pool mining sees the combined algos are linked to 1 of each piece of gear.

kind of  like 2 ssds or 2 hdds can be linked  into  a raid.  Once  done  speed and size in a striped raid are faster and bigger, but you could still break the pair up and of course sacrifice both speed and size.

thats kinda already here(ish)

smallish FPGA gpu accelerators, specifically sqrl acorns, or later the BCU1525s and other big block (for want of a better term) FPGAs to accelerate GPUs.

pool sees one miner for several discreet pieces of kit. mix and match for best results per algo. a BCU could accelerate a lot of gpus. acorns only one or two perhaps.

just takes time and devs, which are admittedly in short supply in the sqrl ecosystem at the moment. that will change IMO. well i sure hope so anyway 

[philipma1957]

best of all asic builder have no incentive to crack ProgPow  as the most profitable way to mine is an asic sha256 such as bitmain's s15 linked to say a nvidia 2080ti doing ProgPow

there must be a way to link or pair the two pieces of gear to one another so that the pool mining sees the combined algos are linked to 1 of each piece of gear.

kind of  like 2 ssds or 2 hdds can be linked  into  a raid.  Once  done  speed and size in a striped raid are faster and bigger, but you could still break the pair up and of course sacrifice both speed and size.

thats kinda already here(ish)

smallish FPGA gpu accelerators, specifically sqrl acorns, or later the BCU1525s and other big block (for want of a better term) FPGAs to accelerate GPUs.

pool sees one miner for several discreet pieces of kit. mix and match for best results per algo. a BCU could accelerate a lot of gpus. acorns only one or two perhaps.

just takes time and devs, which are admittedly in short supply in the sqrl ecosystem at the moment. that will change IMO. well i sure hope so anyway 

part of the reason I mention what I did is OhGodaGirl made progpow algo  and was involved in the acorn project.
She could be looking to develop  the s15 + rtx2080ti + linking fpga  to mine  a sha256 + progpow coin

[Wind_FURY]

Crypto is not a mere engineering/programming domain, mathematics and socioeconomics are kings here, ProgPoW is secure by virtue of its brilliant game theoretic concept: making it infeasible for ASIC attacks to be started in the first place by de-incentivizing such attempts.

I believe an ASIC company will build dedicated hardware, if the coin is successful enough, and when the price of the coin is right. If it pays to build specialized hardware, then someone will.

It is not about price of the coin.

It is about mining profitability which is and remain marginal with the exception of short periods of price surge. Even in the case of a permanent highly profitable mining industry, costs of designing and manufacturing a special gpu-like hardware with 20% advantage, won't be compensated by the profits, as gpu industry would be competing with the manufacturer at the same time to produce a more powerful real gpu.

Of course it is, if taken in the right context. Let's pretend that Ethereum made the change to ProgPow, and the coin has rallied to $10,000 per ETH, a huge success for the cryptocurrency. Would the ASIC companies not develop any specialized hardware for it? Impossible. They would risk millions in investment to be the first in the race to build one.

[philipma1957]

Crypto is not a mere engineering/programming domain, mathematics and socioeconomics are kings here, ProgPoW is secure by virtue of its brilliant game theoretic concept: making it infeasible for ASIC attacks to be started in the first place by de-incentivizing such attempts.

I believe an ASIC company will build dedicated hardware, if the coin is successful enough, and when the price of the coin is right. If it pays to build specialized hardware, then someone will.

It is not about price of the coin.

It is about mining profitability which is and remain marginal with the exception of short periods of price surge. Even in the case of a permanent highly profitable mining industry, costs of designing and manufacturing a special gpu-like hardware with 20% advantage, won't be compensated by the profits, as gpu industry would be competing with the manufacturer at the same time to produce a more powerful real gpu.

Of course it is, if taken in the right context. Let's pretend that Ethereum made the change to ProgPow, and the coin has rallied to $10,000 per ETH, a huge success for the cryptocurrency. Would the ASIC companies not develop any specialized hardware for it? Impossible. They would risk millions in investment to be the first in the race to build one.

Which is why  making a coin to have an asic and a gpu linked via an acorn type controller for max profit. Would make the asic company think twice, instead they have an incentive to cut a deal with  amd or nvidia and with an fpga builder.

that holy trinity concept  fuses  companies together.  You can then use the mining tech as a financial tool to expand renewable power.