Author

Topic: Devs: there'll always be an ASIC for your algo. (Read 863 times)

sr. member
Activity: 364
Merit: 250
Vertcoin made a direct jump to ASIC.

https://www.flowertechnology.com/product/lilac/

So this one comes with "a lot" memory to allow for scrypt-N mining?
Assuming this flower tech is even legit...
full member
Activity: 186
Merit: 100
Vertcoin made a direct jump to ASIC.

https://www.flowertechnology.com/product/lilac/

So this one comes with "a lot" memory to allow for scrypt-N mining?
hero member
Activity: 686
Merit: 500
Vertcoin made a direct jump to ASIC.

https://www.flowertechnology.com/product/lilac/
hero member
Activity: 686
Merit: 500
Speaking of which litecoin's half purpose is lost, so will it be replaced by something else (in 2nd place).

And what will be replacing scrypt in general? Quark seems to be the most popular, but it's slow on GPU.
sr. member
Activity: 420
Merit: 250
I highly doubt companies out there are going to keep chasing Devs with every new  "So Called Scrypt resistant"" algo? doubt it - they will be forever chasing... It doesn't take that long to change code , but it does to create Asic, Test, Market , ship etc.. Is it even worth for them?
newbie
Activity: 55
Merit: 0
You technically can make an ASIC for anything, but... in some cases it's just plain stupidity. For most algos, ASIC-capability was actually a design criterium, so I won't blame them for being implemented. For some, the memory constraints alone make it a less interesting proposal (scrypt for example, with scrypt-n being a further development), but it's still possible. However, there are quite some algos that for now run mostly on CPUs, even with programmable GPUs all over the place, and enough greedy developers to get it working. Why? Because you'd have to design the whole thing from scratch, using functions that are very very slow on a GPU (random memory access, pointer chasing) and probably by the time you finally made the super-ASIC for it... the algo is out of favour.

Personally I will not focus on scrypt ASICs, I'll just move on to the next type of coin with my GPUs. The ones deciding the value of coins (those who spend money on them) will follow suit, probably. Pay2win has never been the idea of cryptocoins, even though that became reality the past months... The problem with pay2win is that someone else will throw more money at it than you can, and you're out of the game.
hero member
Activity: 686
Merit: 500
Be it Scrypt-Adaptive-Nfactor, or scrypt-jane or whatever... even litecoin devs claimed there'll never be a GPU miner and now there's an ASIC on the way.

"But my algo requires an exponentially increasing amount of memory."

And ASIC miners can't have more than 10KB or memory, right? And they can't have modular memory slots like those DDR3. As a result --

Quote from: Vertcoin
Vertcoin N-factor increases with time to stay one step ahead of any possible ASIC development.

But magically GPU automatically increase in memory as the difficulty increases. But ASIC magically cant. WOW.

Reality is -- you can make an ASIC to take modular memory, but that'll never happen for a GPU. So these algos are GPU resistant and ASIC friendly in reality.

These devs are certified dumb people.

At best you can make your algo GPU resistant -- it's the same reason why GPU can't follow CPU instructions.

If you truly want a CPU only coin, it has to be a CPU instruction (or better yet, a changing CPU instruction) cause the CPU is also an ASIC for the instruction. If you want a GPU coin, it has to be changing OpenGL calls, cause GPU is ASIC for OpenGL.

Apart from that, all you can do to delay an ASIC is make the the algo complicated like with darkcoin, quark, chaincoin etc...

Speaking of which quark algo is the best yet to resist a GPU miner.
Jump to: