Quote from: cryptomines on April 01, 2014, 04:32:52 PM

Quote from: g0re79 on April 01, 2014, 09:49:59 AM

Quote from: oinquer on April 01, 2014, 09:28:13 AM

Quote from: loicatraile on April 01, 2014, 09:06:37 AM

It´s good to increase the hash power but if everybody can increase it then is pointless and you will only increase power consumption .. so please hvc.1g stop updating the cgminer!

You maybe speak out the truth!
My happiness after increasing my hashpower has gone. Angry

Its better than some other guys squeezes AMD cards with new cgminer in secrecy. And at least we can fuck back all those 750ti noobs Cool

Just for note - hashrate incrase is +30% but power consumtion rised just by tiny 4%, so no big deal, here are results for my rig of 4* R9 280x at 1100 / 1500 (various brands, no voltage / BIOS optimalization at all):

cgminer v5 - 61MHs / 850W
cgminer v6 - 80MHs / 880W

So, 80Mh at 880 watts breaks down to 11 watts per Mh. My 5* 750ti's (no OC, -500 mem) give me 55 Mh at 220 watts which breaks down to 4 watts per Mh. You get 25Mhs more than I do for 660 watts. It seems to me that I could add 3 more 5*750ti rigs for that same 660 watts and get another 165Mh/s. Which would be a total of 220 Mh/s for the same 880 watts of power. 750ti's have far better performance per watt, and in the end less money to the electric company means more money in my wallet. Now, my math could be wrong, but I don't think so.

After writing this post I came back to the top to add this. I'm not a hater. Up until this post I've been mining it. I was cought up in the "cool aid" myself. But today I decided to think about this coin, it's implementation and it's long term survival. I don't like what I'm thinking. Below is my current thoughts. I haven't given this much thought other than what I just typed. So PLEASE correct me where I'm wrong (and I hope I am). I start with a response to another post which has useful information.

So at 80MH/s you right now would be generating about 227.74 coins per day on 4/8/14. HVC is currently trading at 0.00002933 so that is 0.006679589 BTC per day. BTC right now is at $452 so your rig is earning $3.02 per day on average before expenses. 880W is 21.12 kWH per day. If your electric cost is 13 cents (with all taxes included) that is $2.75 in cost per day. You are making $0.27 per day mining HVC at present. If you figure any type of depreciation or any other factors then you are most likely loosing money mining HVC at present. You will never get an ROI at present profit levels. So why mine HVC at present if there are other coins much more profitable? You could always use that money to purchase HVC if you think it has a future. Profitability would be higher for 750ti users as the power requirements are lower, but again, why not just mine a different coin that is more profitable?

BUT DOES THIS COIN HAVE A FUTURE?
To me this coin really has a lot of problems. I'm a programmer and very good business analyst. I don't think it was thought out well. More hype then delivery. Here are some of my reasons:
Coin was sold as CPU only by the devs. (see here for a good profile of how this was originally "sold": http://blog.gryfencryp.to/2014/03/19/heavycoin-hvc-democratic-cpu-coin/) Didn't even take a week to have GPUs mining this coin. The whole implementation of the coin was based on CPU and not GPU mining. Forget about any thoughts of ASIC mining as that just wasn't possible according to what the dev's told us. Then the tone switched to ASIC resistant but is it? See below:

Coin has what many consider a unique feature of decentralized voting but it's more gimmick then anything else due to implementation failures from my interpretation of the timeline. It's a novel idea but the implementation is/was horrible and automately will have no affect regardless of your vote. Vote 1 or 1024 and it will hardly make a difference to the life of the coin.

Devs made a horrible choice in having three phases of production which are Mint, Limit and Sustain (at least in the setup). For the sake of argument lets just look at the Sustain phase. 10,000,576 coins for a duration of 4.7 years to 38+ years. Doesn't matter what the vote is. The Sustain phase should have been called "Coin is Dead" phase. 10 million coins at 4.7 years (minimal). Really? I mean really? It hardly worth mining at present. Who in there right mind is going to mine it during the Sustain phase? That's 242 coins PER HOUR at the BEST CASE (only 4.7 years). This isn't per 2 minute block (well over 550 coins) but PER HOUR. Forget the diff value. No way/shape/form anybody with any sense will be mining at this point. From over 550 coins per 2 minute block (16,500 per hour minimal currently) to 242 at bast case per hour? That's one hell of a DIFF JUMP. Really? You guys really think this is a good idea, regardless of the fact that you can vote on it? This is the best case. Worst case isn't even worth looking at since best case doesn't work. ( extrapolate 4.7 years to 38 years)

Mint phase had 45,000,000 coins (regardless of duration). Limit phase is fixed at 56 days until coin is dead (from mining anyway). I honestly don't see how anyone would mine it after this period. You can do the math to determine how many coins based on current vote pattern will be mined during the limit phase. Really the only variable worth exploring.

IPO gave 5 million coins away (where did these coins come from?). There was a pre-mine (called 1-2%) but if you really look at what the numbers will look like it's a lot higher. Not a statement but a question: Was this coin not really pre-mined around 10% plus or minus? Regardless of what you call it, where did the 1-2% plus 5 million coins come from? What is that percentage compared to what will most likely be mined though out the mint and limit phase. What is that percent? That is really the pre-mine regardless of what you call it.

I'm not sure how anyone can see this being a good coin for anyone not in on the "ground floor" or being a dev with their hand in the pocket of the IPO when you factor in what I just typed.

Another MAJOR problem: A few miners hold the keys to most of the coins. What happens if they decide to dump? 3 or 4 people could single handily kill this coin outright with a complete dump of their coins. HVC in the current alt coin market right now could probably not survive this. gpools, want to make some extra cash? Get people to bribe you not to dump your HVC. I want 10% of bribe money for the idea.

One of the other characteristics of this coin being sold as unique is the Hefty1 algorithm (http://heavycoin.github.io/about.html#asic-resistant). This isn't some new algorithm per say but a combination of other algorithms that already exists. It does nothing more than interleave bits from each hash of 4 other hash functions into a combined 256-bit hash. Nothing special going on here by any means. Pretty simple/basic stuff. Not sure what all the attention is about this as this type of thing has been done for ages. Sounds good on initial inspection but the 4 algorithms chosen weren't the best to choose. SHA-256 should have never been one of them IMHO.

1/4 of the Hefty1 algorithm is not ASIC proof. SHA-256 is what ASIC machines in the wild currently run best!!! If I thought this coin had a true future I would dive right into the code and separate the algorithms to run on separate GPUs for faster processing. Sure the GPUs would run hotter but I'd produce more hash/s too. That's an increase right there. I'd run all SHA-256 via any type of ASIC device (even the "cheap" ones not worth running for BTC would be valuable here for HVC). Thereby freeing up all GPUs not having to process any SHA-256 stuff at all (another huge increase for the GPU hash rates). Assign different GPU(s) to each remaining hash function. Slower ones get more GPUs. The idea with a GPU farm is to have one ASIC device to process SHA-256 and other GPUs running dedicated algorithms for each. Then with a central "operator program" doing the dispatching I'd run circles around everyone else. I wouldn't be surprised if someone else is already doing one or both of these ideas! Not really that much involved. I'd do this myself but I'm not sure I see the long term value of the coin. Do any of you after reading this? Using this type of approach with X11 or Quark is much, much harder as your not just doing interleaving. The parallel processing would be much more difficult to get high level results then Hefty1.

From an HVC ASIC resistance standpoint it would have been far better to only use Keccak-512, Grøestl-512,BLAKE-512 and leave out SHA-256! Why was an ASIC hash function included in the first place? I can't figure out the reasoning behind it, can you? But, HVC was conceived as CPU only! Not even GPU, forget about ASIC. Failure on Devs part is all I can think of. Just really not well conceived.

Technically if someone had a good SHA-256 ASIC device and wanted to run a pool they could cut the SHA-256 out of the end-user's program and handle all SHA-256 at the pool level. Give a special client to pool members and this pool would dominate since they can hash much faster then everyone else as 1/4 of the hash functionality is done via ASIC device(s).

This is what I mean by "the coin is not thought out well" or done during Amature hour as I'll call it.

Again, with the first paragraph taken into consideration and again stating I'm not a hater but not sipping the "cool aid" anymore. I want to believe in HVC but after thinking about it and typing this, I just don't see it.

Any/Everyone, what is wrong with my logic? What am I overlooking? What really is unique about HVC that has a future? What do I have wrong in my logic?

Carlo