Note: I was away for a couple days, but not too much else has happened so here's a response to an earlier post:
Scrypt-Jane was a specific implementation of Scrypt-Chacha with pre-programmed N-Factor changes, and fundamentally my problem is that they were poorly chosen. Now we're at the point where things are slowing down, but it's also "too difficult" to get miners up and running, so only those that got in early are still worried about mining UTC. Really, can you imagine anyone new to the cryptocurrency scene saying, "Oh, look at UTC -- I wonder how I can get set up to mine at NF-13?" They can do Scrypt, X11, X13, etc. and mine any of a couple hundred coins, whereas Scrypt-Jane requires tweaking parameters for just one coin (or at least on N-Factor). It's a major pain in the butt! Scrypt-N is practically the same algorithm but with NF changes more widely spaced; in January the first Scrypt-N coin will move to NF-11, and most of the difficulties with that N-Factor are now known; the next change will be in another year or more, so SJ gets to pave the way with little reward for doing so.
I have VERY limited knowledge in the realm of coding. But from what I understand, Scrypt-Jane is a
software library and Scrypt-Chacha is a
mixing algorithm:
https://bt.irlbtc.com/view/206577.msg2332885#msg2332885.
You say tweaking the parameters is a "pain in the butt," but I would rather tweak parameters than have to drop thousands on an ASIC just to be profitable. Oh, and when that ASIC is no longer profitable, I have to sell it at the right time because it eventually just becomes an inefficient space heater. Now you are saying Scrypt-N is fine because the scrypt-chacha coins 'paved the way' by showing some people how to tweak their parameters? BTW people will definitely wonder how to mine UTC if it is the most profitable for their GPUs as with anything else. Go to the cudaminer thread... when the 750ti came out, tons of people were mining YACoin because it.was.the.most.profitable. I agree the start of the NFactor schedule was painful but that has passed and the changes are spaced out now and CPUs with high cache are relatively profitable--at least with YAC.
You're correct, and I suppose I phrased that wrong. Scrypt-Jane is a library and it can use a variety of mixing algorithms, of which Scrypt-Chacha was selected for all of the coins that use Scrypt-Jane implementations (YAC, YBC, UTC, CACH, MRC, THOR, etc.), at least as far as I'm aware. Basically, Scrypt-Jane instead of becoming a "library" just became a specific implementation in my opinion. Then Scrypt-N looked at things and said, "let's do something similar but change the way the Scrypt-Chacha N-Factors change" -- or maybe it's not really the same? Anyway, I don't worry too much about the distinction between the two approaches to adjustable N-Factor Scrypt as it's not really pertinent to most users.
As far as getting a rig mining with a specific N-Factor, in my experience it's far more easy to do this on the NVIDIA side than it is on the AMD side of things. Cudaminer/ccMiner is often not faster (though sometimes it is), but there's no mucking around with thread concurrency, intensity, etc. to worry about. Yes, the launch config can help, but I've always found it a little odd that we get HW errors in the first place -- I guess I don't know what they really mean, but an error suggests that the hardware performed all the calculations that were asked of it and somehow got the wrong answer. And with SGminer or whatever, if you go past certain levels of intensity these become very real, and also a real pain to debug. If you only have one rig, or perhaps if you have one type of rig (e.g. all of them are running R9 280X GPUs or similar), it's not the worst thing in the world, but if you have a variety of hardware (which I do), it means debugging on every single rig. It's why I eventually quit mining UTC/THOR/MRC/etc. (never mind that THOR ended up being a complete bust regardless).
The ironic thing is that all of this was done to "protect us from the evil ASICs", and yet I'm not even convinced ASICs are the real enemy here. This has become a big business, and that means the small fries (you, me, and anyone else that can't invest millions of dollars) are probably just lucky to have gotten in early enough to have made some good earnings. I don't think we'll actually see a fully functional Scrypt-N ASIC in a time frame that will be profitable for the buyers, but the manufacturers of the ASIC will still make money -- or just scam people and disappear with the coins they're paid. But how do you stop that from happening? Government oversight, laws, regulations, etc. are all things that BTC was trying to avoid early on, yet now we're trending more and more towards having all of them. Oops.
I personally would be ok with ASICs if basically all of the companies didn't screw people with empty promises. ASICs, in a way, are a sign of success for a coin. But I think the most successful coin
long-term (very) must have one of the best
perceptions of being fairly distributed among the
most people. Scrypt-Chacha coins are currently the most poised to earn that perception over time, even though the marketcaps
currently do not even compare to other algos. YBCoin is the highest chacha coin btw at #26 ($1.5 million).
I think the egalitarian viewpoint of things being "fairly" distributed just isn't going to happen in our current alt-coin world. Bitcoin and Litecoin had a chance at it (and still didn't succeed, because not enough people knew about it early on, or didn't bother, or whatever), but today? No way. Unless you can somehow force everyone to mine with some common set of hardware, and prevent anyone from renting servers (or mining rigs), and also stop people from mining with too many rigs, etc. what we now live in is a world where lots of people know enough that everything is always "unfair" to a large majority of the people. But do you know what? As a judge once told me, the concept of "fair" is usually just a lie perpetrated by people who simply want to get their way (or get more than others) often use as an excuse. "It's not fair that a doctor earns more money" is the same as saying "It's not fair that someone else got better grades on a test." Go read Harrison Bergeron for my perspective of what talk of "fairness" ends up becoming if taken too far. Forget the perception of being "fair" because at best you'll simply have a facade -- either the devs end up controlling an unfair percentage of coins, or the big miners do, or the IT professionals who take the time to set up multiple addresses do, or.... You get the point. The best thing for a coin is to have people use it for real stuff. Bitcoin is now at this point (I figure when Newegg is accepting BTC payments, we're basically at the level of mass acceptance), but everything else is lagging far behind.
The power efficiency issue is also real -- I have given up on Scrypt-N and anything else above NF-10 for that reason
What is this power efficiency issue? Scrypt-Chacha uses LESS power as the NFactor increases as I measured myself and posted on this very thread. Do I need to post pictures of my measurements? (serious question)
It's not a question of whether Scrypt-Chacha is using less power as N-Factor increases, so much as it is a question of how much power Scrypt coins in general are using. Anyone that has been mining for a while knows that Scrypt was harder on GPUs and used more power than SHA256, and Scrypt-N is harder still and uses more power than Scrypt. Scrypt-Chacha tends to end up being somewhere between Scrypt and Scrypt-N in terms of power and stress on the GPUs. I think at one point you or someone else measured a decrease in power use of 7W going from NF-12 to NF-13 or something (maybe it was NF-9 vs. NF-13?) Anyway, let's say it's 7W less on a GPU that uses 225W. Now compare that with X11/X13/X14/X15/Cryptonight... all of these use about half as much power as Scrypt mining, the cards run cooler (and quieter), and long-term the hardware is thus less likely to fail.
Incidentally, I've had all of the fans on my dual-fan Sapphire 7950 cards fail thanks to Scrypt/Scrypt-N/Scyrpt-Chacha -- I replaced them with 120mm fans and box fans, as buying real replacements was a joke. ($40 for two small fans!? No thanks, and they'd just fail again in another 3-6 months.) I have a friend that purchased three single-fan 7970 cards and ended up having all three cards die after about six months (yes, overclocking was involved, so it's partly his fault). The number of R9 290X GPUs that have failed due to overclocking + mining is rather high considering the age of the Hawaii GPUs as well. But if you mine X11 or some other "less stressful" algorithm, it's more like running games in that the GPUs aren't pegged at 100% use and 85C+ temperatures 24/7.
