Bitcoin Forum

Wow, KnC just upped the ante on the ASIC market.  Titans are now going to be at least 250MH.  Wowzers


https://www.kncminer.com/categories/litecoin-mining-hardware

"Q2/Q3" is the only thing you needed to know anyways.

Move on, there's nothing to see here.

As i posted in another thread - assuming that each titan does 250mhs @ 10k and a Gridseed (minimum) 350khs minimum - each Gridseed would have to drop to $14-16 to match that type of khs/cost ratio. I highly doubt they can be sold at that amount for profit.

Glad i stuck to my GPUs.  When Gridseeds are worthless, my GPUs won't be.  

I hear it.  I wasn't around for the KnC bitcoin days but I have read that they aren't the best at delivering when projected.  I'm not investing in one of these until I can pay and have a product in hand.  Of course, it will probably be too late then anyway.  I have been thinking about selling everything and just buying coin at this point.

Wow  :o

250 MH/s ... that's some heavy hashing for $10k

the arms race has begun!

Yep - keep your GPUs cause vanilla Scrypt is about to be worthless to mine(soon)!

I unfortunately sold off 1/3 of my mining rigs for Gridseeds.  If I can get a good 6 months out of them before these KnC Titans hit they will pay for themselves by electricity savings.  I knew it was a risk but my power costs were starting to kill me.

Hell, I've actually bought more GPUs. Got 2 Gigabyte 7950s for $225 each on Fleabay.  

Believe it or not, now is a good time to be picking up cards.  I am completely anticipating vanilla Scrypt to worthless and the profitability shift to non-vanilla-Scrypt coins or not-Scrypt coins.  

However...KGW/DGW on so many of these vanilla Scrypt coins will only serve to exponentiate their individual difficulties because of the extreme amount of hash being thrown at them in such a short amount of time. Their profitable window will be tiny or non-existent.  All of the old vanilla Scrypt coins are about to get blown up.

I would dare say that Litecoin or any other non-KGW Scrypt coin may be only viable option for miners like the Titan.

I agree.  Before there wasnt a damn AMD card on Kijiji in my area, only wanted ads.  This week, there have have been 10-15 cards listed for sale.  Scrypt miners are starting to get out of the game.  I think the combination of ASICS and Summer temps have scared miners into trying to recoup their money before the dump starts.

It's all much simpler than that. The large drop in profitability has caused it.
The cause of that drop is debatable, but the cause of GPU miners exiting is pretty clear.

It isn't the threat of ASICs or summer temps. But simply profitability getting really close to dropping below the electricity costs in many regions.
Products posted by OP are very irrelevant still due to the projected release date, which will probably be late Q3... and that is even if it isn't a giant scam.

I want one 8)

K.

KNC needs $5 million of more pre-orders :P

Vert is not really ASIC resistant as it still uses scrypt and ASIC's could be modified to mine this as well...

The best ASIC resistant coins are ones that use multiple/unique algos as well as random hashes. Coins like Quark, Hirocoin, EMC2(soon to be), Darkcoin etc .. as well as the POS coins like blackcoin and mintcoin, that use proof of stake to mint...

This just got ridiculous...

Yes , its now a cat and dog between asic  makers and developers. GPU mining will stay as long as new algos will keep on emerging.

I do not want it - it is the same risky business as with bitcoin ASICS. When these arrive to market they strive to pay themselves and electricity. It is real lottery to order it. You risk much less when you invest into coins. And you can guess, which are becoming attractive now.

Vert is REALLY ASIC resistant, it is not because of it cannot produce ASIC to mine Vert, but due to N-Factor changing, it is limit time ASIC cycle.
For example, when you created an ASIC for Vert, you will likely to be able to use only 1 year or less due to N-Factor changing. So, it will cost you a lot.

While other algorithm using in Dark, Quark is NOT ASIC resistant anymore. Why I say like that ?
According to NIST (National Institute of Standards and Technology), if you want to become SHA3 candidate, you need to be able to __create by ASIC___. AND every algorithms in X11 used to be SHA3 candidate until Keccak win the competition and become SHA3.

Here is the paper : http://csrc.nist.gov/groups/ST/hash/sha-3/Round2/Aug2010/documents/papers/SCHAUMONT_SHA3.pdf

So, You cannot say those coin is ASIC resistent because when they come out, it is done, while Vert is not the same due to N-Factor which keeps changing.

lol i just read the post above my own...answered my question.

250 Mh is insane. What will alpha-t do now?


X11 the algorithm of DRK has blake, bmw, groestl, jh, keccak, skein, luffa, cubehash, shavite, simd and echo. The ASIC must be able to solve all those algorithms. Or at least that's my understanding.

pump it to 1000Mh/s please  :D

So they just more than doubled their previous 100mh/s on the same model, seems legit.

This is fantastic news, its stiff competition and is very welcome in this space...I'm extremely excited

They did with their SHA Asics too and delivered.

If theyre anything like theyre Neptune on delivery then we are all screwed.  And they just upped it again so we'll see how much of it is real and how much is bullshit.

Good luck ROI'ing since the majority of Scrypt coins (other than maybe Catcoin) have accelerated retarget schedules. Most Scrypt chains will adjust for the additional hash power almost immediately and you'll be competing against KnC's own Scrypt mining farm.

That's why they increased the hash so more ppl would buy them.  Now they added more customers and more money to start their farm.  By the time we get the titans ROI will be so high you will need to buy their next gen miner (which they have already started mining with) just to catch up.  Never ending cat and mouse game..we always lose KNC makes money off our "loans".

So.. Vertcoin and its clones will have value up!

This is an interesting conjecture. Can you explain your reasoning in more detail ?

I, too, am really curious as to how you could back up your estimate. Are you assuming there is a minimal amount of transistors you have to use and a minimal amount of reads/writes on a on-chip memory?

Overall energy consumption is affected by many factors. Like layout of the board. for example in an ASIC, you can have memory very close to a processor significantly lowering wattage.

Agreed i think it has many variables. One company is going to try run a modular type design with one power board that feeds a calculation board for better efficiency, also making it upgradeable being it modular.

Hopefully more people will adapt to asic-resistant coins til then. Scrypt is almost dead for us GPU-miners.

I am debating on dumping my GPUs but if x11 picks up, I might just cruise craigslist looking for deals from miners dumping hardware. Aim that firepower at x11 and continue mining until x11 gets ASICS.

I refuse to enter the Scrypt ASIC arms race, the only winners are the hardware makers and the resellers.

and litecoin bagholders who are looking to offload when it kicks off  for network security  8)

Minimum 300/MH/s of performance from a 28nm chip(Not 250)

I'm thinking about getting some. I'm crowdfund raising for one right now. How it works is everyone will get payouts based on buy-in ownership.
https://bitcointalk.org/index.php?topic=553599