Topic: [ANN][BLC] Blakecoin Blake-256 for GPU/FPGA With Merged Mined Pools Stable Net - page 161. (Read 409641 times)

If we can't get connect and mine, FAIL!

Connected to the pool and all I got was "rejected .... (high hash)". Nice

so if we compare the $ value using a speed based matrix then we could assume this:

if BTC is the gold standard and LTC is silver standard we can see that the speed of the hash rate goes from fast to slow therefore if we have a faster BLC is this the Platinum standard  

if so then from this we can deduce that BLC is worth more than BTC if we analysed it from the speed H/s or from a hash per watt perspective  

until Blakecoin has been around for at least 4 years we won't be able to scientifically prove this  

Elementary my dear Watson. Simply compare the blockchain securing effort requirements against similar efforts for other blockchains and compare reward parity. Have I came across a false conclusion?

I think kramble described the issue best


is vardiff on?
issue with cron jobs?

+1

I'm back, I will be looking into the pool issue today.

How can you affirm this ?

would be silly to sell at those prices  

but Bitcoin was also worth very little $ per BTC at one point so time will tell  

Anyone want a pizza only 10,000 BLC  

Why Blakecoin beats other coins in the long run:

• The reward for mining Blakecoin does Not decrease over time it only increases with block height and difficulty
• No restriction on any platform as Blakecoin does not include artificial *Security* that slows down possible mining hash rate and reduces power efficiency
• Already has planned use for Blakecoin as a currency storage between MMO systems with prototypes in development
• The modified blake-256 algorithm hash rate is just under 3x faster on the GPU and just over 2x on the FPGA compared with Bitcoin

this is going to take time but here is the early schedule:

MOOC - due for testing ~ Q1/Q2 2014 (some courses will be free and some will be Blakecoin/£/€/$)
virtual learning so you can lean how to make games and see how we make games, new courses will be added on a ongoing basis

MMOFPS - due for closed beta ~ Q2 2014: 60% done

bit too early for these to be correct!:

MMOFPS - due for open beta ~ Q3 2014

MMORPG - due for closed beta ~ Q3/Q4 2014: 30% done

so by the end of 2014 Blakecoin should be well known at least to gamers and students


Edit:

working on Blakecoin version 0.8.9 atm porting the changes from Bitcoin 0.8.6 and updating the checkpoints so should be out soon

I can report the pool works correctly at least 50% of the time and that at this time each blakecoin is worth about 0.01 USD or so...

for the pool this is my cgminer.conf:

{
"pools" : [
   {
      "url" : "stratum+tcp://stratum.blakecoinpool.org:3333",
      "user" : "UserName.Worker",
      "pass" : "WorkerPassword"
   }
],

"intensity" : "9",
"auto-gpu" : true,
"expiry" : "120",
"failover-only" : true,
"gpu-threads" : "2",
"log" : "5",
"no-restart" : true,
"queue" : "5",
"scan-time" : "10",
"worksize" : "128",
"temp-hysteresis" : "4",
"blake256" : true,
"vectors" : "1",
"no-submit-stale": true,
"kernel-path" : "/"
}

I only run cgminer no option switches as it is all in the cgminer.conf

downloaded and compiled from https://github.com/kR105/cgminer

run cgminer -k blake256



Why WU = 0 

I will back to work on this soon, I'm out on a business trip for a few days and will be back very soon!

its great to see a working pool but the high-hash error needs to be fixed by kr105 please its so close to working perfect  

the modified blake-256 algorithm is just under 3x hash rate in my testing compared with Bitcoin on the GPU and just over 2x on the FPGA, Nvidia cards are slow needs a Cuda Miner  

Thanks for making A POOL i have a feeling this coin will hit the exchanges now and  in a big way!  too bad this pool just came out    ive found with a little tuning i was able to double my mhs from the default conf file. is that the blake hashing formula? (im about three times faster than btc mining) now 1 issue  high hash reject wtf what do  i need to do? i lose about35%





0
 

yeah I noticed the diff 4 on the dashboard before it gets the high-hash error then dashboard drops back to diff 2 but the pool still gets the high-hash error until a new block is found on the network?

I do get less on the pool vs solo but it should be more consistent, but due to the high-hash issue on the pool the average per hour is lower  

I do get the correct payout for my round share ratio on the pool

Yep, but in the long run (evening out the variation) the earnings should be identical (less the 3% pool fee). My concern comes from the fact that the pool is accepting shares at diff=2 (even though my worker sometimes shows as diff=4 on the dashboard), and given that the earnings to date are roughly half, I'm wondering if only half the shares are being credited? It could just be that my FPGA cgminer is not submitting all the shares, though I tried modifying it to sumbit diff=1 shares and they were all rejected high-hash (which explains what that message means). Anyway if you're seeing the correct earnings/blocks ratio then it must just be variation as you've got a lot more blocks than anyone else  

EDIT yep, its high-hashing right now. Unfortunately I didn't save the overnight log (just grepped it for BLOCK!) so I can't check how much was actually rejected, actually, belay that, its in my terminal history..

 [2013-12-16 11:52:07] ZTX0 0100401300-1 | 212.0MHz | (5s):422.5M (avg):426.0Mh/s | A:1754 R:1016 HW:440 U:0.0/m
 [2013-12-16 11:52:07] ZTX1 0100401300-2 | 212.0MHz | (5s):422.7M (avg):423.6Mh/s | A:1823 R:1045 HW:58 U:0.0/m
 [2013-12-16 11:52:07] ZTX2 0100401300-3 | 208.0MHz | (5s):414.7M (avg):418.5Mh/s | A:1757 R:946 HW:418 U:0.0/m
 [2013-12-16 11:52:07] ZTX3 0100401300-4 | 204.0MHz | (5s):406.3M (avg):407.6Mh/s | A:1792 R:963 HW:99 U:0.0/m

pi@tvpi ~/blakemine/cgminer-ztex $ grep BLOCK log.txt
 [2013-12-15 20:27:15] Rejected 0001f401 Diff 33.6K/2 BLOCK! ZTX 0  (high-hash)
 [2013-12-15 20:32:48] Accepted 000f79c7 Diff 4.23K/2 BLOCK! ZTX 0
 [2013-12-15 22:20:11] Accepted 000ed150 Diff 4.42K/2 BLOCK! ZTX 1
 [2013-12-15 22:21:13] Accepted 00085b65 Diff 7.84K/2 BLOCK! ZTX 2
 [2013-12-16 06:19:00] Accepted 0014d800 Diff 3.14K/2 BLOCK! ZTX 0
 [2013-12-16 07:23:40] Accepted 0007393c Diff 9.07K/2 BLOCK! ZTX 0
 [2013-12-16 07:30:00] Accepted 000e03e4 Diff 4.67K/2 BLOCK! ZTX 1
 [2013-12-16 07:36:11] Accepted 0000a9a6 Diff 98.9K/2 BLOCK! ZTX 1
 [2013-12-16 09:37:23] Accepted 00113296 Diff 3.81K/2 BLOCK! ZTX 0


So it was actually bad for just under half the time. Strange that there was a eight hour gap in the block submissions though. Next time I'll save the log rather than overwriting it.

EDIT2. Its possible that the ztex's just went SICK->DEAD, then they restarted at around 06:19 ... I've just had that happen right now (running cgminer interactively rather than in background). I'm going to have to do a work around for this as the SICK/DEAD logic is more appropriate to the GPU cards than FPGA. Perhaps I'll just exit cgminer on detecting SICK and run it in a loop in a bash script so it restarts (and reloads the bitstreams) automatically?

I think pool mining comes down to amount of share per round rather than blocks found?

solo is better if you find the blocks quickly and don't get any orphans or wallet issues  

Edit:
still getting the (high-hash) issue on the pool periodically 

It'll be the higher diff. The changes I made to cgminer 3.1.1 won't have affected solo mining at all.
Specifically:
gen_hash() lines 5392-94 is only used in the stratum pool not for solo mining.
diff1targ line 5607 just affects the final check before submitting a share. I should revert this as its just plain wrong (in the sense that it allows sub diff 1 shares to be submitted, but these will never be generated by the FPGA except as HW errors).

I'm not sure that the (unimplemented) changes to calc_diff() to are actually required as I think this just affects the displayed diff and as that is currently correct, it seems pointless to change it (especially as I don't fully understand the code anyway).

The pool seems to have been behaving itself overnight as I didn't see any rejected blocks in the log. I'm a little concerned that to date I've found 17 blocks (425 BLC) on the pool but my earnings (both confirmed and unconfirmed) are just 225 BLC (almost exactly half). It could just be Lady Luck doing her hot/cold thing again, but it would be useful to cross-check with other users' experience.

Well I got orphan blocks again.

Not sure if less often because of Diff reporting is better or if Higher Diff means I'm finding few blocks to notice.

Fascinating, that prompted a little google-fu. Apparently Latinum cannot be "replicated" (though I assume Gold can be), so it has some similarity to cryptocurrencies. Of course currency would be an anathema in a post-scarcity culture. Anyway this is getting seriously off topic, so I'll stop there.