Looks like there's still some glitches with the new difficulty settings.
I used h=64 on uswest, but right now it's at difficulty 32. I thought it wasn't supposed to go lower than the "h=xx" setting?
EDIT: Now it's back up to 64, so it didn't stay on 32 very long, but it was there for a few minutes.
h= is a hint. It only suggests the starting value. You can use d= if you want fixed difficulty with no variance. From several of the posted logs, it looks like it could be as simple as the pool publishing a difficultly which is actually half that of what it accepts.
Could it be that simple, an off-by-power-of-two error somewhere?
Very unlikely. It appears the problem occurs when vardiff change and coin change happen one after another. Kinda irrelevant now with this new awesome fixed difficulty - thanks poolwaffle! |
|
|
|
This said, my 945KH/s+ cards run at a cool 65 degrees C max (lowest is 57 degrees).
Looks like you don't use standard Sapphire Tri-X card. It's quite hot. 74'C is minimum for me  Of course I use standard Sapphire Tri-X cards... Sapphire R9 290 Tri-X OC, no mods. I just have plenty of experience running rigs. I have banks of reference Sapphire R9 290s running at <60 degrees too:  Stock unmodified reference cards. 899KH/s stable as a rock too. |
|
|
|
It's better to check Watt rate and temperature also.
Why? Power consumption is almost totally irrelevant at current scrypt profitability. You could use double the power for 10% more hashrate and probably come out ahead right now. This said, my 945KH/s+ cards run at a cool 65 degrees C max (lowest is 57 degrees). PS: I use TC: 20481 don't know where I found it, but it give me better rate, than higher values
Would be shaders * 8 + 1, which makes sense. I find shaders * 10 + 1 to be solid on these cards, but this also won't make a huge difference. PS: xI don't make much difference. I get almost same rate from 200 to 600
If by "almost the same rate" you mean several percent, sure. But yes, the key is really that using xI over I makes a decent difference. |
|
|
|
|
PoolWaffle: when you implement the ability to specify starting vardiff in the password field, can you please make it a minimum value. e.g. I say d=512, and that means my vardiff should never change (because max is also 512). This would be rather ideal for larger miners, as vardiff is quite useless to me, and is just causing rejected shares on occasion. Having a fixed diff also makes certain rig stats easier to calculate.
|
|
|
|
I have currently a mini-farm pointed to MiddleCoin with a good success rate of earnings.
What can i expect if i move my operation to this pool?
How are the earnings currently?
I have consistently had comparable or better earnings from this pool. Wafflepool has higher variance to some extent (one day a little lower, one day higher), but it has overall been a much better experience then middlecoin for me. And when I ran side-by-side tests, I was pleased with the earnings here. I just dropped intensity a bit and am seem to be getting less rejects now. But it's weird, because wafflepool used to give me under 0.5% reject rate with these same settings. Slight config changes I suspect, making lower intensity a better bet.
Looks like I spoke too soon. Poolwaffle: something is borked here. I just saw basically this scroll by: Accepted share diff xx/32 Accepted share diff xx/32 Accepted share diff xx/32 Rejected share diff xx/32 ... (Share is above target) Rejected share diff xx/32 ... (Share is above target) Accepted share diff xx/32 Accepted share diff xx/32 Rejected share diff xx/32 ... (Share is above target) Accepted share diff xx/32 Accepted share diff xx/32 Accepted share diff xx/32 Rejected share diff xx/32 ... (Share is above target) Rejected share diff xx/32 ... (Share is above target) Rejected share diff xx/32 ... (Share is above target) Accepted share diff xx/32 Accepted share diff xx/32 Accepted share diff xx/32 Accepted share diff xx/32 Accepted share diff xx/32 Accepted share diff xx/32 |
|
|
|
Maybe it's that I'm on the same network as the server...
That would definitely help, heh. My rejects all seem to be because of vardiff switches, and rare cases of 2 blocks found within seconds of each other. I just dropped intensity a bit and am seem to be getting less rejects now. But it's weird, because wafflepool used to give me under 0.5% reject rate with these same settings. Slight config changes I suspect, making lower intensity a better bet. |
|
|
|
Also, has anyone had any trouble with the new stratum endpoints? I've added them to the homepage so we can get a few more users moved over to them, but I haven't seen anything yet in this thread about problems  uswest has been rock stable for me so far. Seeing a higher reject rates then before tho - over 3% right now. |
|
|
|
Theres something odd with our block processor (might have started sometime last night). Working on fixing it now, but looks to be something odd with processing casinocoin (keeps processing the same block over and over).
I presume that means the casinocoin balances are higher then actually? Not quite clear what direction this means the balances are going to go for the day. It did seem a bit crazy when I saw almost .1BTC in unexchanged casinocoins on my account, but I thought we might just be having a good day. |
|
|
|
Yeah i wasnt sure about the PSU. So a 1200 watt would probably be better. What other mobos do you suggest? I've built plenty of gaming rigs in the past but this minning rigs are new to me.
1200W is still overkill I would say, but sure. 1000W would be what I would buy - look up the RM1000 from Corsair. Good quality gold rated PSU, pretty cheap because they skimped on the gimicky extras like pretty cabling, packaging, and velvet bags. Asus Z87-A is a decent board for a good price. Can run 4 cards without issue (never had luck getting it to detect a 5th tho). A lot of this stuff comes down to availability tho. See what you can find for a good price that is actually in stock ready to ship, and then search for info about it. |
|
|
|
|
Get a better powersupply. You don't need 1500watts, so spend more to get a gold or platinum PSU of better quality.
3x 280X should run fine on a 1000W. Or if you can find a good deal, get 2x 650W or 750W and save some $.
But, good luck actually getting an ASRock Pro BTC board... Almost nobody has stock, and you definitely won't find one for $70.
kalus: he has 3cnt 280X listed.
|
|
|
|
I've now got it set up with a .conf file and running kalroth's 3.7.2 cgminer and can't get as high of a hash as I was getting before without the xintensity. I think my card has 2816 shaders so the xintensity would be 372?
Try bumping it up a good chunk, say double. Personally I didn't have much luck with kalroth's miner, even though it has the same xintensity code as sgminer. 372 makes sense by my math. I use 1100 on my 290 (non-X). Also set your tc to a multiple of 2816, and then add or subtract 1 to see which gives more efficiency (usually adding 1 will give you a bump). Say, 28161 or something. xintensity means you don't have to push TC so high to get the same performance, because you are aligning the number of threads launched with the number of threads you can actually run at one time. |
|
|
|
Absolutely right, didn't think of that. But (newbie question alert), isn't true that whenever the total pool's rejectionrate is lower (and acceptrate higher) we as a pool have more chance of finding blocks because less hashpower is wasted?
Nope. Some hashpower is still be wasted on mining old blocks - you're just getting share credit for them. It's true. We take a chance to orphan someone else's block that other pools don't take. If it works, extra profit
Yes and no. Every pool has this chance, presuming they submit blocks to the network even if they are 1 behind the main chain. Doesn't matter if you get credited for shares 1 up the chain or not. For us to have a solid chance to orphan someone else's block as the second submitter, we would have to only build on our own chain (which I presume we don't), and that is provably only beneficial with 30%+ of the network hash rate. (Read about selfish mining) Re: Removing coins I would keep LTC only because it is a good baseline, stable block chain that will always be there. I would argue that a scrypt switching pool should always guarantee to do no worse than LTC by at least including LTC as an option.
The others I see no issue with removing them.
I would say having LTC on the profitability chart is good... but since difficulty changes so rarely, maybe pull the daemon and don't have it in the switching rotation. LTC is probably one of the heavier daemons too, since the spend volume high, so a good one to pull. |
|
|
|
Still the rejectrate is kinda sick low. I've got one GPU that has 2045160 accepted shares and 1920 rejected. That's less than 0,1%... (connected to main server wafflepool.com and living near Amsterdam).
Keep in mind, this is the same for everyone, so your profit won't likely be higher as a result of lower rejects. Wafflepool accepts shares that are 1 block old (just not two blocks old). This is unlike middlecoin, that stops accepts shares for the previous block as soon as a new block is found. This means everyone has the same advantage of lower shares on wafflepool. All this said, I like wafflepool's stale strategy a lot more, as it seems to decrease variance and makes rejects more meaningful. |
|
|
|
|
Any live block explorers right now? Both the ones listed in the OP are down.
|
|
|
|
Coin removals:
<snip>
Unless there is any major opposition, those leftover amounts would be left in the exchange account for if we start mining that coin again (at which point they'll go right back into normal exchange cycles). If we don't ever end up mining that coin again, either they'll be considered a donation to the pool, or we'll have a vote here to decide what to do with them.
Thoughts? Just keep the fractions of coins for dropped chains. They equate to barely anything when you divide them up anyway. I'm okay with you considering it a donation for your efforts. No idea why I lost shares for an hour last night eh? Was weird, but at least I noticed it so not much income lost. --- Maybe it is just being calculated incorrectly? You should give users personal BTC/mhash based on BTC paid divided by sum of submitted share difficulty.
Edit: in particular i think you might be calculating it by past 24 hours bitcoins divided by current hashrate when you need to sum it all instead of using current hashrate
So far, BTC/MH seems to be consistently a bit higher then reported on the main page. That said, the orphan rate being reported was bugged - poolwaffle noted this a few posts above you. |
|
|
|
[...]
On the R9 290, usually memclocks between 1250 and 1499 will decrease performance. If you can't hit 1499/1500 memclock, then use 1250. not on hynix cards mate..r9 290 with elpidia and with hynix meories are 2 completly different cards and require complety different tunning..what You said is completly right for elpidia models. If I go 1499 or 1501 my hashrate drops a lot, but not on hynix one's. I have both elpidia and hynix cards... |
|
|
|
[...]
On the R9 290, usually memclocks between 1250 and 1499 will decrease performance. If you can't hit 1499/1500 memclock, then use 1250. |
|
|
|
Something very fishy is going on. Main server vardiff stuck on 1024, shares not being counted, website inaccessible. Something is badly wrong or perhaps something malicious?
I hope it gets sorted out soon. I've moved everything to uswest for now, as it seems to be working reasonably well.
Website hasn't had any issues for me. Main server might be pushing the vardiff up because of load. Weird stuff about the uncounted shared tho, probably just an odd glitch during the upgrades. I highly doubt it is anything malicious. |
|
|
|
Is it possible to know how much BTC u have earned so far?
Yes... look on your miner page. Go to http://wafflepool.com/miners , enter your BTC address in the textbox and hit submit. Poolwaffle: something just borked badly on the main wafflepool.com server. I had 10MH/s of shares go uncounted for over 30 minutes until I restarted sgminer and switched the rigs to the uswest server. Looks like I lost over half of my hashing power for an hour - was mining and submitting shares, but the site wasn't counting them. This happened starting around round 4330. |
|
|
|
|
[22:03:51] Pool 0 difficulty changed to 32
[...]
[22:03:58] Rejected 0ec9617a Diff 17/16 GPU 3 Pool 0 (Share is above target)
Just FYI. 7 seconds != 15 != 30 seconds I think, heh.
---
Revelations86: so far this has been more profitable then middlecoin for me, and I've been here from close to the start.
|
|
|
|
|
Show Posts
