First Asic for SIA - Obelisk SC1

[Hakkane]

gpu mining is the most secure way to secure the network

No it is not for small coins. Sia mining network is equivalent to 1/8th of the total GPUs mining ETH.

That means that if any of the 3 biggest ETH mining pools turns to mine SIA, they can instantly do a 51% attack.

There are several other reasons to explain why GPUs are bad for the security of small coins. Check the post I mentioned of the dev team: https://blog.sia.tech/choosing-asics-for-sia-b318505b5b51

[1715557807]

1715557807

[1715557807]

1715557807

[1715557807]

1715557807

[1715557807]

1715557807

[ekrififi]

If this "first ASIC" is built on an older process node like 55nm, it's not going to be a huge cost or risk to whoever designed it.

Keep in mind that the Baikal is 28nm yet sells faster than Baikal can make them right now because it's the best performance option currently available.


 The Gridseed GC3355 back in Scrypt days was a 55nm device - yet Gridseed appears to have made a TON of money on them before folks started selling 28nm ASIC (KNC, Innosilicon A2, Alcheminer, among others).


 I don't see ASIC having ANY price effect on SIA - they will affect PROFITABILITY but that's not the same thing at all.



 There are quite a few coins that have dedicated ASIC - all SHA-256 coins (NOT just Bitcoin), ALL Scrypt coins (Litecoin, Doge, DGB among MANY others), ALL X11 coins (DASH and some more have MULTIPLE ASIC devices available for them - and the Baikal ASIC also does X13/X15/Qubit and 1 or 2 more algos (DGB I know has a qubit option).

 Litecoin price did NOT explode when ASIC showed up for it - it was dropping slowly from it's ATH by that point and CONTINUED to drop 'till it eventually got down to around $2 for a long time.  (it kicked up into the $3.50-$5 range after the halfing, and STAYED there for about a year, then SegWit adoption got approved and it started jumping).

 DASH - price didn't change any more than it's normal variability when ASIC showed up for it, but that's when I dropped GPU mining on X11 as it became totally unPROFITABLE very quickly.

 I'm not sure on the timing for Bitcoin vs ASIC, but I'm sure that the price did not change because ASIC showed up for it, ASIC showed up for it because it's price was rising enough to make it a good risk for ASIC makers TO make machines for it.

 It could be argued that DOGE died in part due to ASIC - the final nail in the coffin for DOGE was when they decided to make it merge-mineable, it's not been worth more than some small number of Satoshi since.

Thank you so much for youe insight, QuintLeo. Much appreciated...BUT the optimist in me has to disagree ob a few points...

1) With Dash and Litecoin, they more than paid out in the long run, no? Also, I was under the impression that if you're mining a crypto, you kind of WANT it to tank in price so the dificulty goes down and you can stack more of it... then just as your hardware has run its course, you'd want that coin to take off. Seems like (unless you're planning on reinvesting your mined crypto in the short term), Litecoin's post ASIC trajectory is ideal.

2) DogeCoin may have a low price per coin, but still has a pretty large market cap (20th largest, I think). Also Doge actually gets used for SOME goods and services, which is rare for an altcoin. You could buy mining contracts with Genesis using Doge.

The one thing Sia has going against it is that it already has a pretty large market cap. it's valued much higher than any other data storage crypto coin (burst, storj etc)... BUT it's valued much lower than dropbox...

[Metroid]

gpu mining is the most secure way to secure the network

That means that if any of the 3 biggest ETH mining pools turns to mine SIA, they can instantly do a 51% attack.

You know that with asics the chances are much higher right? few people can totally obliterated the network and so on, Its very unlike any of the 3 biggest pools will get a 51% stake as for asics yes, they can do that with few resources. GPU mining is the most secure cause anybody owns a gpu and can mine and secure the network, now for asics only selected few and if they group each other then bye bye.

If asics come to SIAcoin then will it end? My guess is in the long term yes, see how sha256 coins are done for? nobody invest on them, there is no way to control the asics.

[ekrififi]

Webpage of Obelisk is finally live: https://obelisk.tech/

Tomorrow they will release the rest of the info.

Pre-orders start in 6 days

EDIT --> BTW, this is the explanation of why they are releasing an ASIC: https://blog.sia.tech/choosing-asics-for-sia-b318505b5b51 It is a question of protecting the network from attacks. GPU mining can be dangerous for small coins in that sense

Oh boy... ohhh boy..Thanks for the info. Can't wait to find out Price, lead time, hash rate so they can shut up and take my money.

I was about to purchase another L3+, but now I'm thinking of pre-orderibg this instead for the novelty, diversification... not to mention FOMO appeasement...

[Hakkane]

You know that with asics the chances are much higher right? few people can totally obliterated the network and so on, Its very unlike any of the 3 biggest pools will get a 51% stake as for asics yes, they can do that with few resources. GPU mining is the most secure cause anybody owns a gpu and can mine and secure the network, now for asics only selected few and if they group each other then bye bye.

If asics come to SIAcoin then will it end? My guess is in the long term yes, see how sha256 coins are done for? nobody invest on them, there is no way to control the asics.

Check the thread, seriously, it may give you a different perspective. What is good for a big coin like ETH may not be good for small coins. If they fall in small hands you are right. But currently, most of GPU mining is of big centralized farms too.

Also, there is another factor: the value of the ASIC depends on the value of the coin. With GPUs, I do a 51% attack on ETH, I will make the price of ETH to dump, but I can still resell my GPUs to gamers or computation centers. With ASICs, if I do a 51% attack... I will have tons of useless and worthless hardware around in the aftermath :-P

[leowonderful]

28NM is also a matured process making production easier than 16/14nm, possibly lowering costs. Unless there's a serious competitor that comes out for this ASIC, it will do fine. Remember Bitmain started at around .78W/GH at 28nm with the S3 and dropped to .25W/GH at 28nm with the S7.

[Eyedol-X]

If the price point and power requirements are right for me as I said previously, I believe in the SIA architecture and will likely be buying one as soon as pre order goes live.

I hope it's cheap

[Metroid]

ASICs, if I do a 51% attack... I will have tons of useless and worthless hardware around in the aftermath :-P

That is a very good excuse for the sia asics. The 51% attack is not only to kill the coin, is also to pump and manipulate its value and in turn make it a mafia coin. If you people are trying to find an excuse to implement and make the people, miners, investors to accept it, you people need to try it harder cause as i have a phd in computer science, you trolls will never convince me because there is nothing good about asics in any way concerning crypto. I'm an anti-asic person, things should never be controlled by few and yet that is how many asic coins are and in their long term they are bound to die. The same thing can be said about LTC, if it was not for asics then ltc value would be around $500 now or probably more.

[Hakkane]

That is a very good excuse for the sia asics. The 51% attack is not only to kill the coin, is alsoto pump and manipulate its value and in turn make it a mafia coin. If you people are trying to find an excuse to implement and make the people, miners, investors to accept it, you people need to try it harder cause as i have a phd in computer science, you trolls will never convince me because there is nothing good about asics in any way concerning crypto. I'm an anti-asic person, things should never be controlled by few and yet that is how many asic coins are and in their long term they are bound to die.

Keep calm buddy, I am not part of the team or are related anyhow to it. Nice you have a PhD in computer science. Mine is in Molecular Biology, and we are just two users having a calmed discussion about ASICs. I am also a an anti-ASIC but for the simple reason it is the start of a technology race that ends up wasting thousands of MegaWatts. And be sure by any means I will buy one for that reason.

What I am discussing is just that from the perspective of securing a blockchain, using a dedicated hardware instead general computers can have advantages. There are pros, there are cons, and it is nice to discuss about them.

Best wishes ;-)


EDIT --> No problem at all, Metroid ;-)

[Metroid]

wasting thousands of MegaWatts.

securing a blockchain, using a dedicated hardware instead general computers can have advantages. There are pros, there are cons, and it is nice to discuss about them.

Best wishes ;-)

Sadly is the "thousands of MegaWatts" that keeps the price as it is, electricity is expensive, concerning "securing a blockchain", I believe is in the power of millions of peers and not in hundreds of them, but that is my opinion on the matter. Good discussion with you and by the way I apologize if I disrespected you in any way 

Best wishes ;-)

[ekrififi]

It's nice to be nice. Happy to see the power of forums to resolve conflicts...

Not to rehash the whole affair, but it seems that this thread is not alone in voicing an anti-ASIC backlash. The backlash was such that David Vorick (from Sia Coin) wrote this blog post on the subject.

Enjoy

https://blog.sia.tech/choosing-asics-for-sia-b318505b5b51

[Eyedol-X]

Well looks like the Asic is going to be $2499.

Not cheap by any means but not too expensive either.

Supposedly they going to NOT accept SIA as a payment option but will accept BTC.

https://www.reddit.com/r/siacoin/comments/6j1gyg/obelisks_sia_asics_full_details/

[mikbal]

Asics will save against %51 attack.  because they will keep %51 of miners themself and sell the %49. In return sia coin will be controlled by a single entity,  which is their goal. Want proof.? Look at the bitcoin  segwit discussions and how bitmain behaves.

[QuintLeo]

Thank you so much for youe insight, QuintLeo. Much appreciated...BUT the optimist in me has to disagree ob a few points...

1) With Dash and Litecoin, they more than paid out in the long run, no? Also, I was under the impression that if you're mining a crypto, you kind of WANT it to tank in price so the dificulty goes down and you can stack more of it... then just as your hardware has run its course, you'd want that coin to take off. Seems like (unless you're planning on reinvesting your mined crypto in the short term), Litecoin's post ASIC trajectory is ideal.

2) DogeCoin may have a low price per coin, but still has a pretty large market cap (20th largest, I think). Also Doge actually gets used for SOME goods and services, which is rare for an altcoin. You could buy mining contracts with Genesis using Doge.

The one thing Sia has going against it is that it already has a pretty large market cap. it's valued much higher than any other data storage crypto coin (burst, storj etc)... BUT it's valued much lower than dropbox...

 1) Sometimes. A LOT of folks lost money on the Gridseed "blade" miners on Litecoin as they were priced too high in the first couple pricing incriments, and litecoin profitability was dropping too fast.
 Folks that went with the later 28nm gear though probably mostly HAVE made their money back, if they kept them running long enough.
 
 Crypto that tanks is a BAD thing in general, as you've already spent a lot of time mining it at the higher diff and then LOSE a lot of your profits. The best option for a miner is a slow but steady price RISE - like Bitcoin did for a while after it hit the low point a year and a half or so ago.
 Second best option is a steady price - like Litecoin for *most* of the last 2 years.
 Basic idea is that you're reinvesting your earnings into MORE hardware to earn MORE money - and you're not going to be able to do much of that if the coin is tanking.
 If you're holding onto the coin, that's INVESTING more than mining and INVESTORS do like to see large price swings.

 2) DOGE has benefited from the last 3 months of "all altcoins get money tossed at them due to the instability of Bitcoin" - there have been NO changes to it's fundimentals and once the current bubble pops for most altcoins I anticipate it crashing hard then slowly fading down into the single-digit or VERY low double-digit Satoshi range again.

[cashen]

So the big question.

Who is ordering?

[HagssFIN]

Not me. July 2018 estimated delivery is a big bummer 

[Msobh]

Not me. July 2018 estimated delivery is a big bummer 

Exactly . i was thinking to buy 1 , but after knowing 2018  . i will not join that

[Eyedol-X]

So the big question.

Who is ordering?

Not me. July 2018 estimated delivery is a big bummer  

Exactly . i was thinking to buy 1 , but after knowing 2018  . i will not join that

This is my thought as well, if I could pay for it in SIA coin today that I previously mined, I wouldn't feel the hit if it didn't produce.

Since they want to be paid in BTC today with the promise of pay back 1 year from now with now earning or interest gained on the $2500 payout from today is a huge gamble.

I think there is less of a gamble on return if you were to buy $2500 of SIA today and wait a year, it could be argued which would make more money by the time the Asic launches.

I'm a believer in what SIA does but there is not enough incentive to "pre-order" this thing.  I could just GPU mine what I can for 12 months and plan on stopping with the Asic hits and see what happens.

I'm still tossing it around in my head right now but the fact I can't pay in SIA coin today is a huge impact to whether or not I will purchase. I already dumped a ton on the Minebox earlier this year and that was only a ~4 month estimated wait.... and I still don't even have that yet or an ETA when it will really ship.

[ekrififi]

So the big question.

Who is ordering?

Not me. July 2018 estimated delivery is a big bummer  

Exactly . i was thinking to buy 1 , but after knowing 2018  . i will not join that

This is my thought as well, if I could pay for it in SIA coin today that I previously mined, I wouldn't feel the hit if it didn't produce.

Since they want to be paid in BTC today with the promise of pay back 1 year from now with now earning or interest gained on the $2500 payout from today is a huge gamble.

I think there is less of a gamble on return if you were to buy $2500 of SIA today and wait a year, it could be argued which would make more money by the time the Asic launches.

I'm a believer in what SIA does but there is not enough incentive to "pre-order" this thing.  I could just GPU mine what I can for 12 months and plan on stopping with the Asic hits and see what happens.

I'm still tossing it around in my head right now but the fact I can't pay in SIA coin today is a huge impact to whether or not I will purchase. I already dumped a ton on the Minebox earlier this year and that was only a ~4 month estimated wait.... and I still don't even have that yet or an ETA when it will really ship.

You're making a LOT of sense, Eyedol-X and company. I wish you weren't, but you are.

BUT alas, I AM going to put in an order come Wednesday, good points notwithstanding. Here's MY thinking, and please feel free to explain how foolish and misguided my thinking is...

1) Both the lead time and hash rate could be consservative estimates. Asic manufacturers got fileted when their release dates were postponed. I think Obelisk learned their lesson from their mistakes and out of fear of people like us, gave themselves some cushion. According to their post, it COULD arrive months sooner. For the same reason, it could be more powerful than 100 Gh/s... although it doesn't really matter exactly HOW powerful the ASICS are. So long as they are the only ASICs on the blockchain, and no GPUs can effectively mine SIA, what really matters is how many ASICS get produced and how long before additional Sia ASICS get are built and shipped.

2) i believe in Sia in the long run. If it manages to shoot up in price after the asics are shipped, then (unlike any other coin) the block rewards would not decrease because all of the asics would already be mining it. In effect, anybody who puchases these first batch of asics would have a monopoly to mine SiaCoin no MATTER how much it shoots up in price.

3) Umbrellas don't make it rain, but every other time an ASIC has been introduced to a coin, it has more than 10Xd afterwards. Any and all profitability psuedo-predictions should consider this potential price increase. Asics do not CAUSE price increases in and of themselves, but I do think they are inddicative of stable infrasture / network and a coin with a future.

4) Even if I DON'T want to mine Sia a year from now, I GOT to think there will be a market for this. Top of the line ASICS are rare, especially when the only people who will own them are people who were willing to buy them with a 12 month lead time. baikal cubes are going for thousands pf dollars. Baikal sold them for 780. The novelty, scarcity and demand for alt coin ASICS may make this a resale play.

What do you guys think? Am I way off on this?

[szafa]

Good news.First deliver in 2018 if it is true.Otherway can be fake.