Litecoin and FPGA/ASIC resistance

[Bullbear]

I've seen a few posts around here and at least one listing on cryptostocks or LTCGlobal that are building FPGA's for mining Litecoins, is this really possible? I thought the whole idea behind Litecoin was that it was resistant to this stuff? Does this mean that Litecoin ASICs will roll out sometime too?

[1714807095]

1714807095

[1714807095]

1714807095

[Mahn]

Of course. Litecoin is resistant to what's currently out there in the market, but that doesn't mean someone can't eventually figure out how to build a FPGA or ASIC for Litecoin. It's theorized, though, that if these things ever get built, they won't be able to provide the equivalent computing boost to that of the current Bitcoin FPGAs and ASICs.

[thep33t]

Of course. Litecoin is resistant to what's currently out there in the market, but that doesn't mean someone can't eventually figure out how to build a FPGA or ASIC for Litecoin. It's theorized, though, that if these things ever get built, they won't be able to provide the equivalent computing boost to that of the current Bitcoin FPGAs and ASICs.

Especially at the cost/Hash.

[furuknap]

I've seen a few posts around here and at least one listing on cryptostocks or LTCGlobal [...]

What did I miss? Links?

.b

[tacotime]

I've seen a few posts around here and at least one listing on cryptostocks or LTCGlobal that are building FPGA's for mining Litecoins, is this really possible? I thought the whole idea behind Litecoin was that it was resistant to this stuff? Does this mean that Litecoin ASICs will roll out sometime too?

Yes, kind of, probably

http://www.openwall.com/lists/crypt-dev/2013/03/21/1

The issue is that C. Percival's algorithm has a "feature" called time-memory tradeoff (TMTO) also known as space-time tradeoff.  Basically, you can trade more ALU cycles for less memory usage.

Very smart people have tried to eliminate this problem and failed; I won't pretend I've solved it with MC2, I myself am only using an alternative algorithm to further delay FPGA introduction.

[Novus]

Well JasinLee has stated that this isn't their group's IPO, so either this is a scam or by a group that used the same name (which is actually a pretty straightforward name)

[iPaulito]

It is just a matter of time... there will allways be someone smarter
- just my humble opinion

[Littleshop]

It is just a matter of time... there will allways be someone smarter
- just my humble opinion

As litecoin has already changed the algo once, they can change it again.  It is doubtful that much can be done beyond GPU's with scrypt due to the memory requirements.  You could create an ASIC but with the RAM requirements it would be large and one quick change to the formula would increase the ram requirements in an instant killing the usefulness of the ASIC. 

[tacotime]

It is just a matter of time... there will allways be someone smarter
- just my humble opinion

As litecoin has already changed the algo once, they can change it again.  It is doubtful that much can be done beyond GPU's with scrypt due to the memory requirements.  You could create an ASIC but with the RAM requirements it would be large and one quick change to the formula would increase the ram requirements in an instant killing the usefulness of the ASIC.  

No, you just increase the number of scratchpad recalculations using the same amount of memory.  You can calculate a scrypt hash of any memory size on less memory, it just takes longer (memory usage is not fixed; it's just the amount of memory that yields the least number of ALU cycles possible).

[Bullbear]

Thanks for the explanations, I think I have a lot of reading to do before I can understand them though

[NickCoin]

It is just a matter of time... there will allways be someone smarter
- just my humble opinion

As litecoin has already changed the algo once, they can change it again.  It is doubtful that much can be done beyond GPU's with scrypt due to the memory requirements.  You could create an ASIC but with the RAM requirements it would be large and one quick change to the formula would increase the ram requirements in an instant killing the usefulness of the ASIC. 

Hopefully that's the case when new scrypt Asics come out. Otherwise, I predict new encryption algorithm will come out to replace Litecoin.

[fran2k]

If someone makes an ASIC for Litecoin, a new litecoin ASIC resistant with the re-adjusted paramesters should appear.

But an ASIC with the amount of memory required to mine LTCs will cost a lot of money.