I looked at an image of the Haswell die and appears to be less than 20%. The APU (GPU) is taking up more space on the consumer models. On the server models there is no GPU and the cache is probably a higher percentage of the die.


I believe that is wrong. I suspect an ASIC can be designed that vastly outperform (at least on a power efficiency basis) and one of the reasons is the algorithm is so complex, thus it probably has many ways to be optimized with specific circuitry instead of generalized circuitry. My point is isolating a simpler ("enveloped") instruction such as aesinc would be a superior strategy (and embrace USB pluggable ASICs and get them spread out to the consumer).

Also I had noted (find my post in my thread a couple of months ago) that the way the AES is incorrectly employed as a random oracle (as the index to lookup in the memory table), the algorithm is very likely subject to some reduced solution space. This is perhaps Claymore's advantage (I could probably figure it out if I was inclined to spend sufficient time on it).

There is no cryptographic analysis of the hash. It might have impossible images, collisions, etc..

Yes you are right, I had looked a a server layout.

But the point being that even if it is 10%, then you can't include more than 10 copies on the same size die using the same process. Using a cheaper/inferior process, you will be including far fewer than 10 copies.

I make no claim about more qualitative design improvements, cryptographic flaws (though in some cases these might translate to CPU/GPU mining improvements as well as ASIC), etc.. I was simply addressing the "thousands of copies" comment. That approach is not feasible.

Finally, I spent enough time trying to understand how to use XMR. My level is too low.
BTW I think it is enough to start trading.

I decided to make an arbitrage. To buy XMR on the Poloniex exchange and to sell it on the Hitbtc exchange.
There is a nice difference between XMR prices on this exchanges. So if you did not try it before - do it. You can earn additional money. Check rates by yourself and make your own conclusions.
After I'll test it with XMR - I'll write here some additional information. If you id it before - write what you think.

I edited my prior post. I think the complexity is working against you, not for you as you assume. Defer to your cryptographers. You've got to characterize it before you can make any such qualitative assumptions with certitude.


The worst thing that can happen is a few have a significant advantage. If that degenerate case is the true risk (we can't say since we don't know), much better to have ubiquitous cheap ASICs.

I assume nothing. ASICs may rip this to shreds. I merely state that it isn't inevitable.

We don't know how much cryptographic analysis there was, since it was all done in secret by whoever designed it (also secret). Maybe there was a lot, maybe none at all, or maybe something in between.

Re: cryptographers, yes, we are paying cryptographers to review it, as has been stated on this thread before. We are conducting a top-to-bottom review of everything. Cryptography is important, of course, but there are plenty of opportunities for other problems throughout the code as well. And this was indeed the case with the deoptimized mining code we discovered when we first started looking at the code we inherited, but certainly the scope of this effort is not now limited to (or even focused on) mining.

the emission halves every 512 days?

Quite interested in having a purely OpenCL miner. I know Claymore is OpenCL, but it only works with AMD GPUs. The Intel HD series of GPUs are very poor I know, but they too utilize OpenCL - for those of us that are hoping to squeeze every little last bit from our systems using the onboard GPU would be very advantageous.

And your CPU, yes. They are separate devices with separate resource pools, despite the Intel HD GPU being on the same unit as the CPU. I have read up on it and other coin algos have miners for the Intel HD series and the use of it doesn't impact the CPU mining speed.

on scrypt, gpu mined slower than cpu.
but probably using less power.

New exchange: lazycoins.com

Disclaimer: please take note that referencing it do not imply we endorse it. Trade at your own risk.

Updated by David Latapie