[ANN][CACH] CACHeCoin released based on scrypt-jane

[Jodlar]

Hello , about hour ago , in my miner i start to see a lot of HW , they only disappear  if i reduce tc , and with reduced tc i only get 200kh. (before was 1500)

Is this normal ? or ?

The N-Factor went up today... so thats the reason. You have to reduce your intensity to get rid of HW errors.



Time: 1391047035 30 Jan 2014 N: 1024 Nfactor: 9
Time: 1393144187 23 Feb 2014 N: 2048 Nfactor: 10
Time: 1394192763  7 Mar 2014 N: 4096 Nfactor: 11
Time: 1397338491 13 Apr 2014 N: 8192 Nfactor: 12
Time: 1401532795 31 May 2014 N: 16384 Nfactor: 13

...

[1714046843]

1714046843

[1714046843]

1714046843

[thedude11]

How come difficulty has remained the same for each block after Nfactor increase? Seems to still be at ~80/block. Shouldnt the difficulty drop proportional to the net hashrate drop?

edit: oh woops doesnt change until hitting the set block change amt..cant wait lol

[bit_coin_genuis]

"api-allow" : "W:127.0.0.1",
"api-listen" : true,
"failover-only" : true,
"fix-protocol" : true,
"kernel-path" : "/usr/local/bin",
"no-submit-stale" : true,
"queue" : "0",
"gpu-threads" : "1",
"gpu-engine" : "1040",
"gpu-fan" : "70",
"gpu-platform" : "0",
"gpu-memclock" : "800",
"gpu-powertune" : "20",
"intensity" : "19",
"temp-target" : "85",
"temp-overheat" : "92",
"temp-cutoff" : "94",
"vectors" : "1",
"worksize" : "256",
"lookup-gap" : "2",
"shaders" : "0",
"thread-concurrency" : "30592",
"scrypt-jane" : true,
"sj-Nfmin" : "4",
"sj-Nfmax" : "30",
"sj-time" : "1388949883"
}

these were my setting for last N factor new yeidling HW errors !

Does anyone have Settings for R9 290's for the new N Factor ... i am currently away on business and access my miners remotely with TEAM VIEWER .   Therefore I can't really test a bunch of settings ... I am currrently getting Hardware errors .. I set my intensity as low as 13 and was still getting some ! 

Any help would be much appreciated from others mining with R9 290's

Cheers,

[vertoe]

How come difficulty has remained the same for each block after Nfactor increase? Seems to still be at ~80/block. Shouldnt the difficulty drop proportional to the net hashrate drop?

edit: oh woops doesnt change until hitting the set block change amt..cant wait lol

yeah no PoW block since nfactor...

these were my setting for last N factor new yeidling HW errors !

Does anyone have Settings for R9 290's for the new N Factor ... i am currently away on business and access my miners remotely with TEAM VIEWER .   Therefore I can't really test a bunch of settings ... I am currrently getting Hardware errors .. I set my intensity as low as 13 and was still getting some ! 

Any help would be much appreciated from others mining with R9 290's

Cheers,

tune down the nfactor until you get more accepted then rejected. not much more you can do. seems the is the HW-algorithm

[Hyacin]

This is my first jane coin, so this n factor change thing is new to me.  I was spewing hardware errors when I switched my miners over to cache this morning, so I've dropped the intensity, they're running cleanly now at about 1/4 the hashing power or something (which is making it pointless to mine cache at all), but they seem very underutilized - the temps are very low.  I'm suspect I can get more out of them if I tweak things other than intensity, but I'm not about to spend a half a day doing that when I'm sure people have seen this before and can save me a lot of time ... is there a general guide/rule of thumb for what to tweak other than just intensity when there is an n factor change on a jane coin?  I don't expect to see the same hashrate, obviously, but my GPUs sitting there bored and cool can't be optimal.


Cheers.

[Sukarti]

How come difficulty has remained the same for each block after Nfactor increase? Seems to still be at ~80/block. Shouldnt the difficulty drop proportional to the net hashrate drop?

edit: oh woops doesnt change until hitting the set block change amt..cant wait lol

Difficulty drops very slowly and only on PoW blocks.

It's very difficult to mine for at least a week after a large hashrate spike, the N factor adjustment is just icing on the cake.

[Sukarti]

This is my first jane coin, so this n factor change thing is new to me.  I was spewing hardware errors when I switched my miners over to cache this morning, so I've dropped the intensity, they're running cleanly now at about 1/4 the hashing power or something (which is making it pointless to mine cache at all), but they seem very underutilized - the temps are very low.  I'm suspect I can get more out of them if I tweak things other than intensity, but I'm not about to spend a half a day doing that when I'm sure people have seen this before and can save me a lot of time ... is there a general guide/rule of thumb for what to tweak other than just intensity when there is an n factor change on a jane coin?  I don't expect to see the same hashrate, obviously, but my GPUs sitting there bored and cool can't be optimal.


Cheers.

Personally I had to completely retune my gpus this n factor to get a decent hashrate. Huge PITA for sure!

[Hyacin]

This is my first jane coin, so this n factor change thing is new to me.  I was spewing hardware errors when I switched my miners over to cache this morning, so I've dropped the intensity, they're running cleanly now at about 1/4 the hashing power or something (which is making it pointless to mine cache at all), but they seem very underutilized - the temps are very low.  I'm suspect I can get more out of them if I tweak things other than intensity, but I'm not about to spend a half a day doing that when I'm sure people have seen this before and can save me a lot of time ... is there a general guide/rule of thumb for what to tweak other than just intensity when there is an n factor change on a jane coin?  I don't expect to see the same hashrate, obviously, but my GPUs sitting there bored and cool can't be optimal.


Cheers.

Personally I had to completely retune my gpus this n factor to get a decent hashrate. Huge PITA for sure!

 

Thanks!  Was your intensity the same in the end?

[bit_coin_genuis]

these were my setting for last N factor new yeidling HW errors !

Does anyone have Settings for R9 290's for the new N Factor ... i am currently away on business and access my miners remotely with TEAM VIEWER .   Therefore I can't really test a bunch of settings ... I am currrently getting Hardware errors .. I set my intensity as low as 13 and was still getting some ! 

Any help would be much appreciated from others mining with R9 290's

Cheers,

tune down the nfactor until you get more accepted then rejected. not much more you can do. seems the is the HW-algorithm

What do you mean tune down the nfactor ... you talking about these values ?

"sj-Nfmin" : "4",
"sj-Nfmax" : "30",
"sj-time" : "1388949883"

or did you mean intensity ?

Are HW error bad for my GPU's ?   I can get no HW errors at intensity of 12 but only 75 kh/s .. when I bump intensity to 14 .. I get 143 kh/s and one HW error for every 2 accepted shares roughly ... This yeilds better results I think ... Unless someone tell me HW errors are actually bad for my Hardware.

[bit_coin_genuis]

This is my first jane coin, so this n factor change thing is new to me.  I was spewing hardware errors when I switched my miners over to cache this morning, so I've dropped the intensity, they're running cleanly now at about 1/4 the hashing power or something (which is making it pointless to mine cache at all), but they seem very underutilized - the temps are very low.  I'm suspect I can get more out of them if I tweak things other than intensity, but I'm not about to spend a half a day doing that when I'm sure people have seen this before and can save me a lot of time ... is there a general guide/rule of thumb for what to tweak other than just intensity when there is an n factor change on a jane coin?  I don't expect to see the same hashrate, obviously, but my GPUs sitting there bored and cool can't be optimal.


Cheers.

Personally I had to completely retune my gpus this n factor to get a decent hashrate. Huge PITA for sure!

 Can you please Share with us

-by posting your config file / line arguments
-what kind of GPU's you are running
-what hashrate you are yeilding?

Much appreciated.

[Walrusbonzo]

Well, I was mining at 330kh/s on my R9 280X at 1160/1500. Using

Code:

-I 12 -g 2 -w 256 --thread-concurrency 8192 --lookup-gap 2

Now having reduced intensity to 11 I'm only getting 80kh/s. That about right?

Does anyone know where I can get a scrypt-jane miner with X-intensity or even raw intensity settings? I keep looking but I haven't found anything yet.

Cheers

[Sukarti]

Are we looking at another 5-6 days before difficulty drops to a better proportion with hashrate?

That hashrate spike really screws with the difficulty.

what wrong? mining goes on, confirmations take less than the targeted 15 minutes...... All is well as for as the protocol is concerned

Nothings wrong per se.

I was just curious how long it will take difficulty to actually reflect the current hashrate?

For example Feb 13 estimated hashrate was around 65MH/s and difficulty was 10.11 with 17 coins/block. (http://stats.cachecoin.org/charts.)

Today estimated hashrate from getmininginfo is 66MH and a difficulty of 53 with 12.8 coins/block

The difficulty has been dropping 10-12 per day since 500MH/s has dropped from the network.

Block time seems to be around 14.9 minutes in the past 16 hours so that aspect seems correct.

All is well regarding the protocol and network however, the people still mining the coin are "holding the bag" to solve blocks while difficulty adjusts back down for several days to a week. You are basically mining at ~75% the efficiency or less you were prior to the hashrate spike.

Am I incorrect with this data?

If we continually see huge increases in hash once things settle back down I feel like it hurts the long term viability of the coin somewhat.

Take this as an example. There was obviously a huge pump of the coin in terms of hash and difficulty the last few days. There is now around 1 week of less efficient or less profitable mining to be done before we're back to "normal". IF there is another huge pump what is the incentive to stick it out if the coin can be "easily" manipulated with profitability ebbing and flowing.

Isn't the purpose of this coins difficulty calculations to make mining as evenly profitable and fair at all times?

Should the difficulty deltas be adjusted slightly so the difficulty is more responsive to huge spikes in hashrate as well as huge drops hence making it more resilient to pump and dump miners?

How many prior blocks are referenced with calculating next difficulty? Or is there a cap on the net change in difficulty per block if it is recalculated every block.

Is there anywhere more information is available regarding block reward and difficulty adjustments for this coin?

Thanks for taking the time to read and answer.

I'm not trying to bash the coin, I just want to understand how it works under the hood a bit better.

The difficulty is recalculated on every block. Max diff can increase is by ~0.3% but it can drop by any % (according to how long it took the last block to be solved). You can find more info from src

I guess we'll see a massive difficulty drop if this is working as intended assuming we can solve a PoW block lol...

[Hyacin]

Well, I was mining at 330kh/s on my R9 280X at 1160/1500. Using

Code:

-I 12 -g 2 -w 256 --thread-concurrency 8192 --lookup-gap 2

Now having reduced intensity to 11 I'm only getting 80kh/s. That about right?

Does anyone know where I can get a scrypt-jane miner with X-intensity or even raw intensity settings? I keep looking but I haven't found anything yet.

Cheers

As I was updating my mining scripts and profitability calculations I stumbled on UTC and ended up with the same problem I was having with CACH, fortunately they're either more helpful in that thread, or they've been on a higher n factor for longer, so settings were easy to get from there, and made a WORLD of difference when I tried the same settings for CACH mining -

R9 280x -

Code:

-I 12 -w 256 -g 2 --thread-concurrency 16384

Got me up to about 136 kh/s per 280x from the 75 or 80 or so I was getting when I was fumbling around with settings myself.

[Hyacin]

R9 270X -

Code:

-w 256 -I 17 --thread-concurrency 20480 -g 1

Again thanks to the UTC thread.  Got me up to about 88.4 kh/s which is in the target 60-70% range that I expect my 270Xs to do compared to my 280Xs.

[Sukarti]

They adjusted on Feb 28th so they've had a little time to flesh out settings.

[Sukarti]

Looks like mining this coin is pretty pointless until difficulty drops.

9 hours in between blocks and we see a difficulty increase 

PoS blocks really need to adjust difficulty for PoW especially when they count towards the 15 minute block time.

[Hyacin]

Looks like mining this coin is pretty pointless until difficulty drops.

9 hours in between blocks and we see a difficulty increase  

PoS blocks really need to adjust difficulty for PoW especially when they count towards the 15 minute block time.

My calculations show it to still be pretty profitable (this is per day per one of my dual 280x rigs and updated with my new hashrates) -

Last 15 minutes -

Code:

+------+--------------+
| coin | dailyrevenue |
+------+--------------+
| GDC  |   0.01963306 |
| CACH |   0.01609080 |
| UTC  |   0.01463573 |
| FLAP |   0.01306514 |
| AUR  |   0.01186802 |
| DOGE |   0.01125413 |
| VTC  |   0.01074249 |
| MNC  |   0.01072226 |
| MAX  |   0.00963190 |
| DRK  |   0.00940311 |
| MOON |   0.00662355 |
| MINT |   0.00596576 |
| LEAF |   0.00573166 |
+------+--------------+

Last hour -

Code:

+------+--------------+
| coin | dailyrevenue |
+------+--------------+
| GDC  |   0.02050751 |
| CACH |   0.01579187 |
| UTC  |   0.01569500 |
| FLAP |   0.01260356 |
| DOGE |   0.01144881 |
| AUR  |   0.01142074 |
| VTC  |   0.01081866 |
| MNC  |   0.01052846 |
| MAX  |   0.00914432 |
| DRK  |   0.00892256 |
| LEAF |   0.00596694 |
| MINT |   0.00592143 |
| MOON |   0.00577686 |
+------+--------------+

The only thing that kills me about it is the confirm times on the pool(s) I use.  I'm pretty sure it takes days (minimum one for sure) for earnings to be credited.

Oh, and don't read too much into GDC from that data, it's up and down wildly all day, it's 24 hour average is much lower than the last hour (cache value here is not valid due to the hashrate change, and UTC and FLAP are only the average since they were added to my scripts a few hours ago) -

Code:

+------+--------------+
| coin | dailyrevenue |
+------+--------------+
| AUR  |   0.01753956 |
| CACH |   0.01654847 |
| UTC  |   0.01578543 |
| GDC  |   0.01458298 |
| FLAP |   0.01272020 |
| VTC  |   0.01266964 |
| DOGE |   0.01080784 |
| MAX  |   0.01042885 |
| DRK  |   0.01015363 |
| MNC  |   0.00993489 |
| LEAF |   0.00814271 |
| MOON |   0.00701192 |
| MINT |   0.00500407 |
+------+--------------+

[Sukarti]

Blocks take 3-5 days to confirm and payout.

Not sure where you're getting your numbers from.

getprofitestimate after last nights N factor adjustment has created a 75% drop in profitability for me.

I was averaging 20ish coins per day with 3.3MH/s on N=10

getprofitestimate is estimating 5 coins per day now with 1.7MH/s


I still think a majority of the problem exists with the difficulty not dropping properly after a large spike. Guess we'll see over the next week


Sapphire Vapor-X 280x 144kh/s
Sapphire 290  170kh/s
Sapphire 290x 182kh/s

[Hyacin]

pool.cachecoin.net says the network hashrate is 116 mh/s right now ...

My hashrate ~271k / net hashrate = 0.00233144039880279119666983471975 (A)

~1152 (B) blocks generated a day (ballparking it as I didn't have the reward formula initially but I'm still pretty close) ...

A * B = 2.68581933 CACH/day

Lowest CACH sell order on Cryptsy atm - 0.00666000 (C)

A*B*C = 0.01788755 BTC/day/rig

getprofitestimate after last nights N factor adjustment has created a 75% drop in profitability for me.

Are your estimates accounting for the drop in network hashrate since the change?

[Slipknot79]

As I was updating my mining scripts and profitability calculations I stumbled on UTC and ended up with the same problem I was having with CACH, fortunately they're either more helpful in that thread, or they've been on a higher n factor for longer, so settings were easy to get from there, and made a WORLD of difference when I tried the same settings for CACH mining -

R9 280x -

Code:

-I 12 -w 256 -g 2 --thread-concurrency 16384

Got me up to about 136 kh/s per 280x from the 75 or 80 or so I was getting when I was fumbling around with settings myself.

Yea, one of my rigs has 2 cards, one of them is runing with 15kh/s oO and the other one with 140kh/s. Any idea whats up?