Hey
UdjinM6, sorry for bothering, but what do you think about
this,
this,
this and
this threads? :smile:
There is a long post in Russian on bits.media which was also quoted here
https://dashtalk.org/threads/jakoby-sozdali-asic-dlja-dash.8091/page-7#post-86125
Here's the same thoughts in English (at least it should be pretty close, I tried my best :smile
.
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
That's a very interesting but quite unexpected discussion, I have to say.
I'll try to answer all the points that were put here one by one, at least those that I can remember. Sorry if I missed smth.
About situation in general. I can hardly understand the fear of ASICs unless you see it just as a threat to existing miners in terms of losing some share of their mining profit (but such is life). As for me - ASICs are a more expensive but more efficient specialized mining device. That basically means that full ROI time of the hardware itself is way longer but it has lower running cost (electricity, etc). Yes, if everything goes well for these first batches there surely will be more proposals with even more effective devices and even more "tasty" price/speed rates, soon they will start to fight not just the GPUs but also each other on the market and most likely this will also result in the end of GPU mining in Dash. That's exactly what happened in Bitcoin. But what is wrong with that? What is wrong from the point of evolution and security of the network? We could fear a new next gen GPU or a new optimized software then, and change algo when this happened (
shall we?).
About algo change. The point of "incompatible" algo was also to avoid the situation when an older and more powerful network (bitcoin or litecoin for example) would attack us - if you dig a little bit in the history you can find a case where bitcoin (or one large pool to be precise) already killed a coin once. The task for X11 was to walk all the way from CPU through GPU and further but not only for the sake of the way itself and not even for the sake of "cold" mining hardware as a result of algo switching, but also to avoid such situation and have a nice period of natural growth. Dash is the most powerful X11 network right now (or at least I think so, correct me if I'm wrong) and it is (
was? see next paragraph) quite hard to find even more power to attack it.
About "ASICs are going to kill the network". GPUs (and non-specialized devices in general) can mine whatever they want and from the network point of view they are quite unstable basis. If there is a disaster in the network (
by disaster I meant a huge uncontrollable fork, double spend event etc that could ruin trust completely) GPUs will just switch to another network and/or algo or they can even be used "directly" - to play games. ASICs can do nothing but produce hashes for specific algos and they are highly incentivized to have a successful and growing network, otherwise their "specialized" investment are going to be burned in flames. (
what I meant here is that GPUs don't really care about 51% and such or at least they are not that incentivized to do so)
About asic monopoly/centralization. I don't want to make miners feel bad, but they don't actually supply blocks to the network, pools do.
So from the network point of view there is simply no difference either it's 10 miners with 4 GPUs each or 1 miner with 1 ASIC behind the pool - network knows nothing about shares because it simply doesn't see them. I mean, centralization risk always was and it is still there - in pools, and it has nothing to do with ASICs.
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
