Looking into the pseudo share difficulty setting ( SomeWalletAddress+512 ), it seems meaningless to configure, especially when running a local p2pool node. So, now I'm confused as to why Bitmain says to set the share difficulty on Antminer S3 to 512 in their FAQ, their answer leads me to believe they don't actually understand what that value represents. After removing the +512 my local node is assigning difficulties in the range of 500-800, is my node calculating which difficulty to give me based on the total hashrate of my miners, or the hashrate of each individual miner, or something completely different?
If I calculate this based on a couple other sites rule of thumb, I get a difficulty around 3016 for my total hashing power, and 522 for per miner. I don't see why the p2pool node should care about a setting based on per miner, I mean what's the difference from the view of the p2pool node if I have one 2TH miner, or four 500GH miners? It seems like the the difficulty assigned should be for the whole of the hashing power.
From Bitmain FAQ:
" AntMiners support p2pool. Static difficulty configuration recommended.
Since p2pool adjusts the difficulty frequently, therefore would bring hash rate loss when the difficulty changes at any time. As a result is that AntMiners' hash rate may be a little lower than that released officially when they are mining in p2pool.
We suggest you configure a static difficulty for AntMiner, when you mine in p2pool. Recommended difficulty is 256 or 512.
2. How to configure a static diff when AntMiner mines in p2pool?
Through ‘Status->Miner Configuration‘, configure Pool Worker as 'BTC address/256+256' or 'BTC address+256'.
"
From my testing of my S1s and S2s on my home grown proxy, Ants have two significant problems with p2pool: 1 - They can't restart work fast enough. Supposedly the latest S2 firmware addresses this, but I have not tried it. I can't imagine them fixing it entirely, but I could see it being much better than it was. That means every 30 seconds when p2pool tells your miners "hey, stop what you're doing and start over with this data instead!", it takes a few seconds for the Ant to do so. In those few seconds there's a good chance it'll submit a share that is no longer valid. In the S2s case, it was a lot more than a few seconds, and if the work restarted back to back in less than 30 seconds, it quickly gets out of hand. 2 - This is what pertains to your question. It also can't respond to share size changes that quickly. If you don't fix the share size, p2pool will at some point increase the minimum share size required, and your Ant can't stop what's it's doing fast enough to switch and then ends up submitting a share that's undersized. This is what generates the somewhat confusing "share submitted larger than difficulty" message in p2pool. That said, this is all pseudo share stuff, missing this share really doesn't matter. It'll just mess with your stats. It may also decrease your throughput, as your Ant has to switch gears, although I would suspect it's not as bad as the "stop and restart" message I refer to in part 1. M |
|
|
|
I see, no cgminer is only running on the antminers and bitcoind is running on a xeon based desktop. I will definitely check the "solo" pool. Thanks.
so: antminer-->cgminer <-- local network --> bitcoind<--desktop.
Maybe you're missing the http:// part on the -o parm? From the readme.txt in cgminer: cgminer -o http://localhost:8332 -u username -p password --btc-address 15qSxP1SQcUX3o4nhkfdbgyoWEFMomJ4rZ
Note the http:// is mandatory for solo mining.
M |
|
|
|
Problem with antminer s3+ solo. My miners work fine for a pool. But when I try to run solo I get either a very, very slow hash rate and super high error rate(s), or the miners may also refuse to connect to the machine running bitcoin-qt/bitcoind and also may incorrectly report connections as being alive or dead. I have tried upgrading cgminer to the most recent version, to no avail. Often, cgminer quits after a few minutes. here is my cgminer command line (<> indicate substitutions to protect the innocent): --bitmain-options 115200:32:8:18:225:0882 -o http://192.168.0.2:8332 -O <somename>:<some password> --verbose --btc-address 1D7CFcY3gYMgNrns6ttRmomyGTLNzZ6XRA --api-listen --api-network --bitmain-checkn2diff --bitmain-hwerror --version-file /usr/bin/compile_time --queue 4096 here is my bitcoin-qt/bitcoind command line: -datadir=/home/<some_user_path>.bitcoin -rpcuser=<somename> -rpcpassword=<some password> -blocknotify=%s -alertnotify=%s The verbose flag did not seem to give any more info, I can find an open port :8332 on the bitcoind machine and netstat will report "ESTABLISHED" then "CLOSE_WAIT". Any suggestions on how to debug the rpc connection or cgminer would be greatly appreciated. I may have no hair left soon. The way you describe this, it sounds like you are running cgminer somewhere outside of your Ants. I'm pretty sure that's not going to work for you. Try editing the cgminer config inside the Ant to have the solo mining options to point it to your Bitcoin instance: cgminer -o http://<bitcoin IP>:<bitcoin port> -u <username> -p <password> --btc-address <your address> Also, CK (author of cgminer) has a public "solo mining pool" up.. if you don't mind the 0.5% fee: https://bitcointalk.org/index.php?topic=763510.0M |
|
|
|
Problem with antminer s3+ solo. My miners work fine for a pool. But when I try to run solo I get either a very, very slow hash rate and super high error rate(s), or the miners may also refuse to connect to the machine running bitcoin-qt/bitcoind and also may incorrectly report connections as being alive or dead. I have tried upgrading cgminer to the most recent version, to no avail. Often, cgminer quits after a few minutes. here is my cgminer command line (<> indicate substitutions to protect the innocent): --bitmain-options 115200:32:8:18:225:0882 -o http://192.168.0.2:8332 -O <somename>:<some password> --verbose --btc-address 1D7CFcY3gYMgNrns6ttRmomyGTLNzZ6XRA --api-listen --api-network --bitmain-checkn2diff --bitmain-hwerror --version-file /usr/bin/compile_time --queue 4096 here is my bitcoin-qt/bitcoind command line: -datadir=/home/<some_user_path>.bitcoin -rpcuser=<somename> -rpcpassword=<some password> -blocknotify=%s -alertnotify=%s The verbose flag did not seem to give any more info, I can find an open port :8332 on the bitcoind machine and netstat will report "ESTABLISHED" then "CLOSE_WAIT". Any suggestions on how to debug the rpc connection or cgminer would be greatly appreciated. I may have no hair left soon. I'm pretty sure Bitcoin is not capable of providing work fast enough to any of the recent ASICs. Furthermore, I'm pretty sure Bitcoin only supports the "getwork" protocol, not stratum. So essentially your S3 is starving for work. I think, but I'm not sure, that there's a way to run cgminer as a "proxy" that your S3 can point to for work. cgminer takes care of converting the getwork data from Bitcoin to stratum for your S3 to use. I don't think you should try to be using cgminer to mine against your S3 as you have it written. M |
|
|
|
|
I've got a bunch of coupons up for grabs. I'd like to keep one of each for myself, just in case ... but the rest I'm happy to give away. PM if you want one, and of what type if interested. First come first serve.
M
|
|
|
|
The statement regarding not getting a share is the same as saying:
I picked a card 20 times and didn't get a Jack - there's something wrong with the deck of cards.
It's more like saying, I'm using one of those card dealing machines from Vegas to deal me 3 cards a second. For the last two weeks I've gotten 1-3 four of kinds every 24 hours, then for 35 hours, I get zero, and I think a card got stuck in the machine. I did finally get a share today but, then it got orphaned. Is it normal with 2TH/s to have a stretch of 35 hours with no share? Is 2.5TH/s big enough to compete in p2pool? If we're talking about random odds, it seems less likely that I would go that long without getting a share than it does that I would get three shares in a day. I'm fine if this is normal, just trying to learn what to expect, and trying to make sure my systems are configured correctly. Shares are all about luck. Your miner(s) process a hunk of data, then looks at the result. If the result is of "high" enough value, it's submitted. p2pool then looks at the work submitted, and if it's of "high" enough value for the share chain, it's submitted to the rest the pool. The rest of the pool either accepts it, or rejects it (orphan/dead). That's what mining is all about. Higher hash rates simply means your miners are doing more of what I stated above more often. That means it's more likely to find a "higher" value share, but it's still all about luck. BTW, the same logic applies to blocks. If the share is REALLY "high", then it's considered a block and submitted to the Bitcoin network. If it's accepted, payout occurs based on the content of the block. If it's not.. it's called an orphan. M |
|
|
|
...
Why do you keep saying "forked code"... unless you consider BTCGuild, Ghash and all the other major pools 'forked'.
Because that is what a "forked" version of software on git is called  forrestv wrote p2pool. Bitmain forked it and have their own version of the code. Nothing at all to do with BTCGuild or ... Right. We're still talking about two different things. Antpool is not p2pool. Any claims to work with the existing p2pool are BS. If, however, they are claiming to rewrite it and fork p2pool to improve upon p2pool's problems, that's an entirely different story altogether. M |
|
|
|
Here's version 3.6. This adds scheduled reboot functionality, either by time or by uptime.  Download link: MAntMonitor36.zipM |
|
|
|
i would be patient , this pool has done really well considering network difficulty and the pool hash rate,
It is just luck ... and past results do not guarantee future returns. M Yes I just said that in the post above yours ... Sorry, I posted the condensed version.  M |
|
|
|
Who just added 2Ph?  AntPool? Edit: We had a similar spike back in October, the ~2PH/s did not stick around long, could be p2pool is a fail-over for a large mine?? Looks like you're right - whoever it was has gone again...... Edit: Has anyone sifted through the changes made by Bitmain in their p2pool git repo yet? Any thoughts? I'm shite with python code, so it's over my head I'm afraid..... I don't think their changes mean squat. If they change any of the fundamental pieces of p2pool, the rest of the network will reject the work, essentially forking the chain. So it's won't be p2pool.. maybe antpool? But definitely not p2pool. M EDIT: It'd be like someone "improving" Bitcoin by tweaking the protocol. That's fine if you want a new coin, but don't expect the Bitcoin network to accept it. |
|
|
|
i would be patient , this pool has done really well considering network difficulty and the pool hash rate,
It is just luck ... and past results do not guarantee future returns. M |
|
|
|
So I recently acquired some Prismas, and now have the unfortunate fact that I am now mining with some of my hashrate on GHash.io (I'd never even created an account with them before).
Would it be possible, since ASICMiner apparently is clueless on how to properly handle vardiff, to add a setting to an Eligius account that allows the miner to set a default minimum difficulty? I set the GHash.io difficulty to 1024 and it's working smoothly (other than the very frequent hardware issues). However the gear does nothing when pointed at Eligius.
I plan on setting up a bfgminer stratum proxy in the hopes that by manually setting -it's- difficulty, I can sanitize the traffic so I can mine on Eligius again. In the meantime, and for users who aren't quite as well equiped to do stuff like that, having the option to use a fixed minimum difficulty instead of vardiff would be nice.
I know ASICMiner needs to fix their miners and their controllers, but I can't control that, so I wanted to address it here in the hopes that it might be an easy(ish) fix, so I don't have to mine on GHash any longer than is absolutely necessary!
Slush's stratum miner will work too. But be mindful how you do it. If your proxy only passes on shares of size 1024, and the pool you mining at is expecting 128, you won't get proper credit. If you have other non ASICminer hardware that works properly at lower difficulties, point them at your proxy to get the difficulty ramped up, then point your Prismas at your proxy. M |
|
|
|
From what I saw, platinum isn't that much more efficient than gold.
M
So are you on a 110-120V or 220-240V power source? 115v. However I'm not posting measurements, as I don't have a C1. M |
|
|
|
There could be some missing info. needed to help explain the difference. First, list if you are in a 120v (U.S.) or 220v(E.U.) area. Higher volt. countries will have a little better efficiency. Also, the closer you are to the 50% load on the power supply, the more efficient it usually is (so a 1000w power supply running 850w will probably run more wattage than a 1500w power supply running the same 850w). Also, a Titanium, Platinum, Gold, Silver, Bronze or none supply will have different efficiencies.
From what I saw, platinum isn't that much more efficient than gold. M |
|
|
|
I got my C1 coupons. Thanks! Now the checkout is messed up:  I hope their SHA256 math is better than this. M Looks like the shipping cost is showing artificially low, can you PM me an address to test please? I think it happens by clicking refresh on the checkout page. M |
|
|
|
I got my C1 coupons. Thanks! Now the checkout is messed up: I hope their SHA256 math is better than this. M |
|
|
|
Both the C1 and S2's are in the same room, right next to each other, so the ambient temp is the same. The C1 clock speeds are the default of 250. The S2's are overclocked to 250. So, both comparison have 4 hashing boards running at 250Mhz. I'm using the exact same power supply in the tests. I just swapped the power supplies. I had the 1300W connected to one, and the 1000W connected to the other, then swapped. Don't forget the PSU for the C1 config is also powering the water cooling unit, which is the pump plus 3 fans.
I thought of the pump + 3 fans idea ... but two S3+'s have four fans altogether. So I'd think it's roughly the same, unless there's something about those C1 fans and pump that is using more power than 4 fans on the S3+'s. M |
|
|
|
I have 2x S3+ that run well at 250Mhz, and I have some C1's. So I did a power comparison and am sharing my results.
Configuration MHz Setting Avg Gh/s Watts Efficiency W/Gh PSU Model
2x S3+ 250 1000 903 0.90 1000W Coolmax 80 Plus Bronze
C1 250 1000 984 0.98 1000W Coolmax 80 Plus Bronze
2x S3+ 250 1000 821 0.82 1300W Rosewill Lightening 80 Plus Gold
C1 250 1000 881 0.88 1300W Rosewill Lightening 80 Plus Gold
Notes: Actual GH/s is close to 1000. 2x S3+ averaged 504Gh/s each. The power usage for the C1 includes the "ANT C1 Mining Cooling Kit" from syscooling.com.
I wonder how far an OC'd C1 can be pushed though .... seeing it is stock rated for ~1Gh/s @ 250 (and the S3+'s were never rated that high, albeit for bitmain's assertion that most will hit 500 Gh/s), that liquid cooling may have some mileage still, what with ramping up that solo fan via the blue wire hack in tandem ........ just a thought. I was thinking that too. However, at what cost? Just because it can run at say, 1.2 TH/s, would you want to? Would the additional power usage really not make it worth while? If an S3+ overclocked to the highest uses less power than a C1 at base speed, then C1 power usage might be off the chart for OCing. M I don't think I agree with you there. Essentially a C1 is four S3 boards and the only way it can theoretically (I know we have empirical evidence above!) consume more power than 2 S3+'s at the same freq has to be something that can be rectified by a user, possibly heat handling (aka loose paste or the liquid cooling being inferior to the S3+'s, which I doubt) or voltage setting (the new firmware has this, do both the S3+ and C1 compared above have the same setting, if at all?). Am I reading the above chart wrong? M |
|
|
|
I have 2x S3+ that run well at 250Mhz, and I have some C1's. So I did a power comparison and am sharing my results.
Configuration MHz Setting Avg Gh/s Watts Efficiency W/Gh PSU Model
2x S3+ 250 1000 903 0.90 1000W Coolmax 80 Plus Bronze
C1 250 1000 984 0.98 1000W Coolmax 80 Plus Bronze
2x S3+ 250 1000 821 0.82 1300W Rosewill Lightening 80 Plus Gold
C1 250 1000 881 0.88 1300W Rosewill Lightening 80 Plus Gold
Notes: Actual GH/s is close to 1000. 2x S3+ averaged 504Gh/s each. The power usage for the C1 includes the "ANT C1 Mining Cooling Kit" from syscooling.com.
I wonder how far an OC'd C1 can be pushed though .... seeing it is stock rated for ~1Gh/s @ 250 (and the S3+'s were never rated that high, albeit for bitmain's assertion that most will hit 500 Gh/s), that liquid cooling may have some mileage still, what with ramping up that solo fan via the blue wire hack in tandem ........ just a thought. I was thinking that too. However, at what cost? Just because it can run at say, 1.2 TH/s, would you want to? Would the additional power usage really not make it worth while? If an S3+ overclocked to the highest uses less power than a C1 at base speed, then C1 power usage might be off the chart for OCing. M |
|
|
|
I have 2x S3+ that run well at 250Mhz, and I have some C1's. So I did a power comparison and am sharing my results.
Configuration MHz Setting Avg Gh/s Watts Efficiency W/Gh PSU Model
2x S3+ 250 1000 903 0.90 1000W Coolmax 80 Plus Bronze
C1 250 1000 984 0.98 1000W Coolmax 80 Plus Bronze
2x S3+ 250 1000 821 0.82 1300W Rosewill Lightening 80 Plus Gold
C1 250 1000 881 0.88 1300W Rosewill Lightening 80 Plus Gold
Notes: Actual GH/s is close to 1000. 2x S3+ averaged 504Gh/s each. The power usage for the C1 includes the "ANT C1 Mining Cooling Kit" from syscooling.com.
That's interesting. That means I should be OC'ing my S3+ units as I'd still come out better than a C1 will! Not what I expected. M |
|
|
|
|
Show Posts
