Topic: ASIC resistent coins? (Read 1425 times)
It does not matters how hard we try, everything that can be computed is subject to faster computing, someway, somehow, with more CPU or more RAM, but somebody will always invent a faster CPU, cheaper RAM and everything else. Sorry
why do you think there are no asic-resistance algorithm? Without time or profitabe factor?
I agree x-11 is my favorite algo too. 
Awesome
Hey, PM me if you're on Reddit (@luxkore), as I'd like to talk more there if you're down to -- I just am more comfy using RES than this forum, which feels a little clunky (and personally, a bit annoying cause the 6 minutes limit).I'm not exactly a newbie to crypto, just finally decided to get an account and start commenting (not just lurking now)
Look forward to speaking with a fellow enthusiast of X11
Wasn't intend for me but... There are +'s and -'s with both. I'm a redditor too. I like the messaging here better, and the central hub of knowledge. I like reddit's altcoin scene a little better. Your 6 minit wait limit will shrink after you get to 15 posts I believe, and it will eventually go away all together. fyi
Hey, that sounds great! Right now I'm just dual-tabbin', haha -- wait until roughly six minutes, come back and post something I wrote up already. Go back to reddit.
Rinse, wash, repeat
I agree x-11 is my favorite algo too.
Awesome
Hey, PM me if you're on Reddit (@luxkore), as I'd like to talk more there if you're down to -- I just am more comfy using RES than this forum, which feels a little clunky (and personally, a bit annoying cause the 6 minutes limit).I'm not exactly a newbie to crypto, just finally decided to get an account and start commenting (not just lurking now)
Look forward to speaking with a fellow enthusiast of X11
Wasn't intend for me but... There are +'s and -'s with both. I'm a redditor too. I like the messaging here better, and the central hub of knowledge. I like reddit's altcoin scene a little better. Your 6 minit wait limit will shrink after you get to 15 posts I believe, and it will eventually go away all together. fyi
I agree x-11 is my favorite algo too.
Awesome
Hey, PM me if you're on Reddit (@luxkore), as I'd like to talk more there if you're down to -- I just am more comfy using RES than this forum, which feels a little clunky (and personally, a bit annoying cause the 6 minutes limit).I'm not exactly a newbie to crypto, just finally decided to get an account and start commenting (not just lurking now)
Look forward to speaking with a fellow enthusiast of X11
I agree x-11 is my favorite algo too.
I am mining heavycoin which uses hefty1 algorithm although is not fpga proof, it looks like it create stalls so fpgas can't be more efficient than gpus. Asics (for hefty1, x11, x13 etc) will probably be faster but there is a ROI question for end user. We will see how it will go with existing scrypt asics.
This ^
Even though scrypt ASICs might seem daunting now, the overall question that matters most is if ROI can happen. And right now, they're basically glorified GPU farms that are packed into smaller spaces, and consume less power. Definitely not like Bitcoin ASICs, which just turned everything on its head.
So a new coin that is scrypt wont make it? I find it hard to believe. I've seen a few launch recently and they still are going.
It's not that they won't make it... It is just that it is very difficult for a new coin to launch with scrypt. Litecoin has so much hashing power, that someone could stop mining LTC for a day at very little lost revenue, and overwhelm the network of another coin, either 51%-ing it, double-spending, or flat out running it into the ground. As a result, a lot of scrypt coins have moved toward alternatives... AuxPOW, PoS, X11, etc... in an effort to stay out of danger.
No known asics for x-11 x-13 x-15, and so on there are some for scrypt and scrypt-n. If a coin is say scrypt like litecoin and don't want asics mining they can fork and switch to another algo. Rented hash can be a bigger issue than asics in my opinion. Instead of buying a machine you can rent even more hash power for the time you want.
I'm not very knowledgeable with the x11/x13/etc, but why is that? Could there be one developed for them soon?
There's nothing to prevent them... they are just newer algos that ASIC manufacturers haven't had time to develop for yet.
Time is one factor, definitely, but another aspect to consider is how profitable ASIC manufacturers think the coin is. If there's only a few, and are fairly unknown, there's no incentive to push forward. After all, custom hardware ain't cheap.
some algos may never get asics. Some just are not used enough to justify making asics.
Well, it all depends how many people decide to jump on the ASIC-resistance bandwagon. If it becomes a popular idea, someone funds development of an ASIC. Case in point: scrypt. Was never meant to be truly ASIC-proof, and things like scrypt-jane and scrypt-N knock out current mining techniques (like GPU mining) over time.
Some argue that with the advent of cheaper hardware, ASIC development will be around the same speed as GPU clock speed increases, so overall there won't be an issue as long as you stay one step ahead -- that's why X11 is my favorite right now
some algos may never get asics. Some just are not used enough to justify making asics.
So a new coin that is scrypt wont make it? I find it hard to believe. I've seen a few launch recently and they still are going.
It's not that they won't make it... It is just that it is very difficult for a new coin to launch with scrypt. Litecoin has so much hashing power, that someone could stop mining LTC for a day at very little lost revenue, and overwhelm the network of another coin, either 51%-ing it, double-spending, or flat out running it into the ground. As a result, a lot of scrypt coins have moved toward alternatives... AuxPOW, PoS, X11, etc... in an effort to stay out of danger.
No known asics for x-11 x-13 x-15, and so on there are some for scrypt and scrypt-n. If a coin is say scrypt like litecoin and don't want asics mining they can fork and switch to another algo. Rented hash can be a bigger issue than asics in my opinion. Instead of buying a machine you can rent even more hash power for the time you want.
I'm not very knowledgeable with the x11/x13/etc, but why is that? Could there be one developed for them soon?
There's nothing to prevent them... they are just newer algos that ASIC manufacturers haven't had time to develop for yet.
No known asics for x-11 x-13 x-15, and so on there are some for scrypt and scrypt-n. If a coin is say scrypt like litecoin and don't want asics mining they can fork and switch to another algo. Rented hash can be a bigger issue than asics in my opinion. Instead of buying a machine you can rent even more hash power for the time you want.
I'm not very knowledgeable with the x11/x13/etc, but why is that? Could there be one developed for them soon?
So a new coin that is scrypt wont make it? I find it hard to believe. I've seen a few launch recently and they still are going.
Would have to agree with the other replies, to some point. No, but all the new algo's that come out are & make it tough to mine with ASIC's. Just like Scrypt though, they will just build new tech that is able to mine with the new algo's.
All ASIC-resistent coins have to be mined by CPU or GPU farms...
You know, hot racks of graphics cards in air conditioned sheds...
Run by inbred miners strumming 2 string banjos... this is hardly the future of Crypto.
Massive low watt hash power from ASICs is the future...
And one can address the problem of centralization with P2Pools...
Or by including ASIC chips with commodity hardware like motherboards, routers, etc.
Scrypt ASICs have dropped in price by 80% in 3 months...
That gives you an idea how dirt cheap ASIC chips will be one year from now.
I suppose the only problem is that it is very hard for a new coin to have a shot if it uses Scrypt or SHA. Even Dogecoin, with it's huge following, struggled to hold hashing power so much that they are implementing AuxPOW. If you aren't scared of a 51% attack, one of the ASIC algos is great. And if your coin is good and does well, ASICs will be developed for it and that's okay too. But it is almost impossible for a new coin to start off as an ASIC coin from what I've seen.
Would have to agree with the other replies, to some point. No, but all the new algo's that come out are & make it tough to mine with ASIC's. Just like Scrypt though, they will just build new tech that is able to mine with the new algo's.
All ASIC-resistent coins have to be mined by CPU or GPU farms...
You know, hot racks of graphics cards in air conditioned sheds...
Run by inbred miners strumming 2 string banjos... this is hardly the future of Crypto.
Massive low watt hash power from ASICs is the future...
And one can address the problem of centralization with P2Pools...
Or by including ASIC chips with commodity hardware like motherboards, routers, etc.
Scrypt ASICs have dropped in price by 80% in 3 months...
That gives you an idea how dirt cheap ASIC chips will be one year from now.
No known asics for x-11 x-13 x-15, and so on there are some for scrypt and scrypt-n. If a coin is say scrypt like litecoin and don't want asics mining they can fork and switch to another algo. Rented hash can be a bigger issue than asics in my opinion. Instead of buying a machine you can rent even more hash power for the time you want.
Would have to agree with the other replies, to some point. No, but all the new algo's that come out are & make it tough to mine with ASIC's. Just like Scrypt though, they will just build new tech that is able to mine with the new algo's.
how does heftycoin differ from x11? Also, what is this crytponote stuff I saw mentioned in that article? I havent heard of that type of algo before
i guess it is always just a matter of time, till they make asics or FPGA for those algo's so i think the answer is more like a NO
Is there such thing as a true ASIC resistent coin? I thought I had heard of a few attempts, but then I've seen ASIC developed for those types of encryption. We all know how quickly technology advances.. do you think its even possible to create a truely asic resistent coin, or will someone simply see the opportunity to make some quick cash and develop technology for it?
There are newer algorithms that try to be asic resistant. The main ones (scrypt/sha256) are not asic resistant anymore. Scrypt was for a while, though.
I am mining heavycoin which uses hefty1 algorithm although is not fpga proof, it looks like it create stalls so fpgas can't be more efficient than gpus. Asics (for hefty1, x11, x13 etc) will probably be faster but there is a ROI question for end user. We will see how it will go with existing scrypt asics.
Maybe this article answers your question:
http://blog.timmattison.com/archives/2014/06/17/why-proof-of-work-algorithms-need-to-use-memory-bound-functions/
http://blog.timmattison.com/archives/2014/06/17/why-proof-of-work-algorithms-need-to-use-memory-bound-functions/
Jump to: