Topic: [POOL][Scrypt][Scrypt-N][X11] Profit switching pool - wafflepool.com - page 216. (Read 465668 times)

I'm using mostly used 5850s. They are around $100 and can do 350-400KH depending on the OC they allow. Used PSUs and used old Asus P5B and P5K motherboards with the cheapest CPU I can find. At least 1,5GB of RAM needed too. A rig with 700-750KH costs around $300. Electricity is free for me.

Thanks for the responses, that makes sense. Looking for Scrypt miners, there are some custom built units, for example a 900 kH/s for 800 dollars or 600kH/s - $500. Looks like they are based on ATI video cards and go for around $100 per 100kH/s.

Of course, I also found a Radeon 7850 for $160 that is supposed to do 350 kH/s and dual 6970 for $400 that will do 850 kH/s....hmmm, I've been meaning to upgrade my gaming rig.

Last question, is there a good profitability calculator for what we are doing? I understand now that those bitcoin ones out there do not apply.

So the first issue is we really can't switch a small enough amount that we would need to to make them worth it (200-300MHs), and even if we could, you're talking about switching 1.5% of the pool, for earning a coin that is (at best), 50% more valuable, and carries a much higher risk (orphans, price, etc).  You're looking at a potential upside of 0.75% (less than 1%) for a decently sized risk, and easily more than a 1% increase in server capacity to handle such switching capabilities.

So long story short, while it is possible to eek out that extra ~1%, its just not worth it (at least at the current time).

I don't think a 10% increase in profitability over mining something else is worth the risk that someone else is going to dump on the market before us.  Market depth is very important.  And I presume poolwaffle is busy enough without adding coin daemons that we may only use for a few minutes and then discard, for a tiny tiny increase in profit (if any).

thanks for ur answer,
what if pool can dedicate tiny part to them in order to fill those little buy supports in market, thats easy money to gain.

As mentioned, those are sha256, not scrypt, so wouldn't apply to wafflepool. You can use them in a similar way to wafflepool at megamultipool - they do profit switching sha256. I've got a single block eruptor (10Ghs) running on mega, but it's only pulling in about $1/day's worth.

At this point, if I were you, I'd spend the money on a couple Radeon 270s. You'd end up making about $3/day at current market conditions.

With the new system we are mining more coins than we were before, but even with the ability to mine more coins, there are limits to what amount of power we can switch at a time.  With the new system, we can switch (roughly) increments of 1ghs, which is awesome compared to what we could previously do (whole 18ghs pool).  But even at 1ghs, certain coins are still too small to reasonably mine. 

Ideally to mine coins with difficulties down near 1.0, we would need to be able to switch something like 200-300MHs at a time, which is just something we can't do yet.  This is a huge step in the right direction, and if the switcher performs well here, we can likely take it steps further, but for the time being, some coins are just too small.

All of those mentioned are bitcoin miners (sha256), all of the coins supported on wafflepool are scrypt (completely different algorithms).

The majority of those I think we already support, and those that we don't have very specific reasons (note, this is just when I'm pulling it):
Grandcoin: 0.58 difficulty (way too low, even switching only a fraction of a pool will cause huge numbers of orphans / wasted power)
Mintcoin: supported
Reddcoin: supported
Diamond: 1.23 difficulty (same as grand)
Flappycoin: just added today to coinwarz, has a total of 4.8 btc of exchange depth before being 50% reduced in price (about 15 minutes of mining time)
Coino: also rediculously low difficulty (1.44)
Lucky: rediculously low difficulty (1.2)
Neocoin: tons of changes needed to be supported (tons of extra rules in block parsing)
Nyancoin: a total of 1btc depth (about 3 minutes of mining until it is 100% worthless)
Catcoin: should probably add
Penguin: supported
Doge: supported

Did I miss any?

So I'm new to this so excuse my stupid questions. I've decided to do some mining for fun with a few VPSes I have access to and people seem to be happy with this pool so I joined.
I'm pulling between 110 - 330 kH/s which amounts to a few cents a day. Not much of a return but this is mostly for fun and is free.

I see some used miners on Craiglist for sale, for example a 60 GH/s BFL Single for 800 bucks. Or Red Fury 2.5Gh/s units for $100. Or a 38 GH/s Block Erupter ASIC Cube for $500

Now my question, what is the expected GH/s if I picked up a unit like that? The whale you guys are talking about has like 8 GH/s and the whole pool is under 30. Are these mining units measured differently somehow or am I missing something?
Are those rates for mining Bitcoins only and cannot be applied to alt coins?

how can we find out if new system works, if we mine same coin list as before?

Lol... one step at a time...  wait to see how the new switching system works first.

Coinwarz is also pretty useless, because it doesn't consider market depth or variable block rewards.

@poolwaffle :

I think now you can add a lot of small coins (maybe top 10 in coinwarz) to include them in mine system.

Very cool. Congrats on the big milestone in WP history, poolwaffle.  Looking forward to see how the new system fairs over the next few days.

Nice. 

As some of you might notice on the stats page, I've removed the pie-charts showing what we're mining.  This is for two major reasons.

1) It was already moderately inaccurate (it was based on switch timings, and not on shares submitted)

2) I've replaced our entire switching system with a new version just now for testing over the next few hours (needs probably 6-12 hours of testing to get any real results), and the new switching system doesn't work nicely with the old way of calculating what we were mining essentially.

On to the exciting part.  We've got a new switcher in place.  It allows us to switch segments of the pool to different coins based on a variety of factors.  It _ideally_ lets us get back to mining coins under 10-20 difficulty without the huge increase in orphans (the main reason we couldn't before).  What this means for you, is ideally our profitability increases (very difficult to tell by how much), assuming everything works.

I'll obviously be monitoring the profitability stats over the new few hours and throughout tomorrow.  Its unfortunately difficult to tell what effect it has over a short period of time (any change in profitability can be attributed to market [price changes], luck, or switcher), so as long as we're not plus or minus more than 10% or so, I'll try to let it ride out to try to rule out factors other than the switcher.

Was a long day of code, I'm stoked to see it in action

Just targeting a round every 10 minutes.  I've been adjusting every so often (today was the bump from 100 --> 150, as the round time had gotten close to 5min).  Really no hard reason for it other than as round time decreases, the variance between rounds increases (we can only process shares so quickly).  And there is a minor increase in load on the server with shorter rounds (caches need cleared)

150mill represents a 50% increase (not 150%) from 100mill unless I'm missing something here.

Why are the rounds 150m shares now? That's an increase of 150% from the normal 100 million.

Something is definitely wrong there.  All of those are valid locations (wafflepool.com soon to not be).  Can you ping them to make sure your DNS is resolving properly?