asic will be coming？
- Thread starter darkasic
- Start date
Lzeppelin
Active member
They'll just let it go, Evan has said he's not opposed the ASICs as long as the coin follows the traditional model of CPU->GPU->ASIC he would have a had a problem with it if there were ASICs at launch making the barrier of entry way too high.
Last edited by a moderator:
I have serious mixed feelings. If ASICs are really open to the world (like gridseeds are), then ok. Anyone can have a piece of the pie, fair and honest.
But look what happened to BTC, and what's on the verge of happening to LTC !! Mega centralisation. Huge contracts and huge mega-watt data-centers. Preorder headaches and scams.
The worse is centralisation. I mean, to me I saw mining as freedom gateway to anyone in the world. Power to the people. p2p and all that. Wasnt that Satoshi's philosophy after all?
And think of this, with a 4 year history of cryptos and what can be done, new ASICs for new algo's is an instant target. Be it bad intentions to nuke the network, be it capitalist interest in raping the coin for profit.
The latter would only strengthen the network, but would smash the small miner. Just like torrents became a monster with a life of its own, why revert back to an "iTunes" model ?
Aren't crypto all about decentralisation? Having no central authority/persons with huge power over anyone else? And now with BTC seriously gaining mainstream attention and adoption and investment... I'm really scared of ASICs.
Anyone can go down to their local computer store and buy GPU's, scale up or down. Distribution is fair, even when local store stocks are eaten up. You can find your way around online merchants with proper warranties and guarantees, or second hand markets. With ASICs its the exact opposite, historically speaking.
Only people with financial power can blow away cash on pre-orders and wait month. Multi-million dollar deals direct to the ASIC source completely ruin anyone's chance at getting a single unit for their own profit pleasures. By the time they can,
a) The ASIC companies have used it
b) sold to big money customers, who then use it
c) dumps on second market when virtually worthless.
I do understand that Darkcoin is a serious coin with super serious intentions, and obviously need a strong network. To me that is made by numbers, not hash-rate. My only problem is the ethics of ASIC companies.
If I have 10 million dollars to buy GPU's and you have 250 bucks... you'll still get one, and I'll still have my mega-farm. Its s direct, fair, entrepreneur driven scaled economy. With ASICs its financial fascism.
Except for Gridseeds. No pre-orders, they shoot them out of China for anyone to grab. I dont know any other company that does this.
But look what happened to BTC, and what's on the verge of happening to LTC !! Mega centralisation. Huge contracts and huge mega-watt data-centers. Preorder headaches and scams.
The worse is centralisation. I mean, to me I saw mining as freedom gateway to anyone in the world. Power to the people. p2p and all that. Wasnt that Satoshi's philosophy after all?
And think of this, with a 4 year history of cryptos and what can be done, new ASICs for new algo's is an instant target. Be it bad intentions to nuke the network, be it capitalist interest in raping the coin for profit.
The latter would only strengthen the network, but would smash the small miner. Just like torrents became a monster with a life of its own, why revert back to an "iTunes" model ?
Aren't crypto all about decentralisation? Having no central authority/persons with huge power over anyone else? And now with BTC seriously gaining mainstream attention and adoption and investment... I'm really scared of ASICs.
Anyone can go down to their local computer store and buy GPU's, scale up or down. Distribution is fair, even when local store stocks are eaten up. You can find your way around online merchants with proper warranties and guarantees, or second hand markets. With ASICs its the exact opposite, historically speaking.
Only people with financial power can blow away cash on pre-orders and wait month. Multi-million dollar deals direct to the ASIC source completely ruin anyone's chance at getting a single unit for their own profit pleasures. By the time they can,
a) The ASIC companies have used it
b) sold to big money customers, who then use it
c) dumps on second market when virtually worthless.
I do understand that Darkcoin is a serious coin with super serious intentions, and obviously need a strong network. To me that is made by numbers, not hash-rate. My only problem is the ethics of ASIC companies.
If I have 10 million dollars to buy GPU's and you have 250 bucks... you'll still get one, and I'll still have my mega-farm. Its s direct, fair, entrepreneur driven scaled economy. With ASICs its financial fascism.
Except for Gridseeds. No pre-orders, they shoot them out of China for anyone to grab. I dont know any other company that does this.
Overd I hope you are right...
I have been a miner for several years and I hate what the ASICS have done to the crypto currency market.
DyslexicZombei
New member
D
Don't underestimate the board design.
Problems with this have badly hurt Hashfast, Cointerra, Black Arrow, and KnC among others. It's not just enough to design and tapeout an ASIC. As we've unfortunately seen: board design is often underestimated or not given enough time/resources. At my day job: I see this same baloney in firmware and drivers for some products from companies in the IT world obviously not expending enough time and resources on building solid drivers/firmware and working on bug fixes.
ASICs will come, but I think it'll be a while. They just don't grow on trees or out of the ground. If made publicly available, like in the BTC world, the FPGAs used to design x-11 ASICs could be profitable themselves for a window before ASICs are fully ready for x-11.
Someone could correct me on this if I'm misunderstanding this: but wouldn't an x-11 ASIC be the equivalent of trying to combine 11 different ASICs onto one CPU die? If so, potential issues with not only designing this but pipelining all of this to actually work on a custom designed board could slow x-11 ASIC development.
I suppose you could bruteforce it and actually have 11 different ASICs on a mother- or daughterboard but that seems to me to be cost prohibitive.
Also, I agree with you about centralization and greed. I was contacted by someone who shared some info with me while looking for a cheaper DC. This person contacted me because Cex.IO literally charged them 44% fees for the year in host/management fees. Really? 44%? That's some moxie or chutzpah IMO.
Don't underestimate the board design.
Problems with this have badly hurt Hashfast, Cointerra, Black Arrow, and KnC among others. It's not just enough to design and tapeout an ASIC. As we've unfortunately seen: board design is often underestimated or not given enough time/resources. At my day job: I see this same baloney in firmware and drivers for some products from companies in the IT world obviously not expending enough time and resources on building solid drivers/firmware and working on bug fixes.
ASICs will come, but I think it'll be a while. They just don't grow on trees or out of the ground. If made publicly available, like in the BTC world, the FPGAs used to design x-11 ASICs could be profitable themselves for a window before ASICs are fully ready for x-11.
Someone could correct me on this if I'm misunderstanding this: but wouldn't an x-11 ASIC be the equivalent of trying to combine 11 different ASICs onto one CPU die? If so, potential issues with not only designing this but pipelining all of this to actually work on a custom designed board could slow x-11 ASIC development.
I suppose you could bruteforce it and actually have 11 different ASICs on a mother- or daughterboard but that seems to me to be cost prohibitive.
Also, I agree with you about centralization and greed. I was contacted by someone who shared some info with me while looking for a cheaper DC. This person contacted me because Cex.IO literally charged them 44% fees for the year in host/management fees. Really? 44%? That's some moxie or chutzpah IMO.
Last edited by a moderator:
well you do know X11 is actually 11 algorithms and not 1 like bitcoin?
If litecoin asic development took about 3-4 years. Thats at least 20-30 years for 11 algo's in 1 chip.
But do know they are further than Intel atm with there technology. I've read somewhere that there was a company producing sha 256 ASIC at 20nm (intel is at 22nm going to 14nm later this year.
If litecoin asic development took about 3-4 years. Thats at least 20-30 years for 11 algo's in 1 chip.
But do know they are further than Intel atm with there technology. I've read somewhere that there was a company producing sha 256 ASIC at 20nm (intel is at 22nm going to 14nm later this year.
vertoe
Three of Nine
Asic development does not take 3-4 years. Board design, prototypes, tapeout, asic - maximum 12 months. The number of algorithms just increases the complexity of the design, not the time required to build one.
BTW for litecoin it took about 6-12 months. Prior to that nobody really thought that could be profitable.