[ANN][DASH] Dash (dash.org) | First Self-Funding Self-Governing Crypto Currency - page 6816.

xorpheus

member

Activity: 75

Merit: 10

Quote from: DRKenstone on February 22, 2014, 12:24:38 PM

Quote from: HammerHedd on February 22, 2014, 01:00:09 AM

Quote from: raze182 on February 21, 2014, 11:58:58 PM

Quote from: DRKenstone on February 21, 2014, 11:31:48 PM

CPU DRK mining question:

I'm using the fantastic darkCoin-cpuminer-1.3-avx-aes-windows-binaries, with these settings:

Code:

minerd-corei7-avx -a X11 -o stratum+tcp://drk.lotterymining.com:4444 -u username -p password -t 8 -q

My proc: Intel Core i7-3770K, 3.50 GHz

I've been consistently getting 490+ khash/s with those settings, which I think is pretty great, but I'm not sure what others are getting with the same or similar proc.

I've been consistently getting lines like this:

Code:

[2014-02-21 22:01:06] accepted: 43/43 (100.00%), 490.02 khash/s (yay!!!)

But as of this morning and after about 50/50, I've been getting lines like this:

Code:

[2014-02-21 22:22:17] accepted: 77/78 (98.72%), 486.72 khash/s (yay!!!)

Concerning that percentage, 1) what does it mean exactly, and 2) why all of a sudden (after about 2 weeks worth of mining) am I getting less than 100%?

Thanks in advance for helping me with my n00b question! Wink

I'm also mining on a 3770K and getting more or less the same performance. If your percentage is less than 100% that means you've submitted a stale share(s) ("boooo" instead of "yay!!!"). In your case 1 stale share, hence the 77/78 instead of 78/78. Not anything to worry about unless the percentage drops significantly.

My 3770k does about the same.

And as raze said, one stale share is nothing to worry about. It's consistently stale shares that indicate a problem. A share can go "stale" if your pool issues you a hash to solve, then someone finds the block before your computed hash is submitted. this can also happen if there is some sort of network delay as well.

https://deepbit.net/stale.php

------

Many thanks to raze182, HammerHedd, and meebs for responding! I appreciate the insight.

This is frustrating, since last night I'm now down to 24 stales on average:

Code:

[2014-02-22 12:18:49] accepted: 1440/1467 (98.16%), 475.63 khash/s (yay!!!)

I'm going to try a couple of things today: 1) I have a new awesome crazy heatsink to replace my stock Intel with and then I'm going to gently overclock to see if that makes any difference and then 2) I'll try a different pool for a bit (but not blaming the current pool at all) to see if that changes things any also.

I would be very happy with 98% accepted shares. My reject rate mining darkcoin is currently 4 to 5% (95% accepted). The only coin ever been better than that for me is litecoin.

jasemoney

legendary

Activity: 1610

Merit: 1008

Forget-about-it

Someone sold 10k+ at 19:06:07 it'll rebound shortly
Edit what u get like 30btc?

Ximp

sr. member

Activity: 336

Merit: 250

CS Student - BC Logo Guy

How did you guys like that? Cool

darko84

member

Activity: 62

Merit: 10

what's happening on Poloniex!?!

fusecavator

member

Activity: 117

Merit: 10

Quote from: eduffield on February 21, 2014, 06:45:57 PM

Quote from: fusecavator on February 21, 2014, 06:39:29 PM

Quote from: eduffield on February 21, 2014, 10:22:54 AM

Other ideas?

I’m open to ideas on how to provide the best security to the network. I would love to hear what people have in mind.

How about adding an option for adding decoy transactions?

Sometimes coinjoins can be picked apart just by looking at what adds up. This was painfully obvious on the darksend testnet the other day, as I remember seeing darksend transactions like Inputs: 500, 5 Outputs: 400, 100, 4, 1 where there's only one possible solution of who sent what.

One could easily construct decoy transactions to specifically add up to the amounts of the legit transaction. For example:
Intended transaction:
Input: 100
Output: 20, 80(change)
Decoy transaction:
Input: 500
Output: 10, 90, 50, 50, 35, 65, 95, 5, 100(all sent to change addresses)

If that decoy transaction is coinjoined with the intended transaction, there becomes 5 extra possibilities of where the 100 was sent to even if no transactions by others help mask it

Also, use more inputs than necessary, and use several change addresses instead of just one to help mask the intentions, and also having more and smaller outputs causes it to be more likely that they can be useful to help mask other transactions joined with them.

The testing we did was just a test of the transaction pooling and remote signing of inputs, not a test of anonymity (good eye though checking that out). One of the next tests we should have that working.

Here’s the plan for anonymizing amounts if anyone is interested:

DarkSend anonymizes unique amounts like 15.15 by using “denominations” of currency in the different transaction pools available. These denominations come in the amounts of 5000, 1000, 500, 100, 50, 20, 5, 1, .50, .25, .10, .05, and .01. For example a payment of 15.15 would be broken down and submitted to the following transaction pools:

Submit payment for 10DRK to addr Xyz using pool 10 (in pool 10 ALL outputs are for 10DRK)
Submit payment for 5DRK to addr Xyz using pool 5 (in pool 5 ALL outputs are for 5DRK)
Submit payment for 0.10DRK to addr Xyz using pool .10
Submit payment for 0.05DRK to addr Xyz using pool .05

The four payments total 15.15, just like paying in cash, except you have no idea who paid you.

Users receiving anonymous payments will then receive separate out of order payments for various amounts adding up to the total amount they were intending on receiving.

With this methodology, payments could be as granular as 0.01, this could be changed in the future if the currency becomes more valuable by adding smaller denominations.

That sounds great, and far better than anything I thought of. Your denominations seem a little weird though. Several of them have rather limited usefulness ( 0.25 -> 0.5 -> 1 is only double each step), and the lack of pools for very low amounts seems like a strange choice. Is there a reason you have all those close amount pools rather than a pool for each 10^n? (1, 10, 100, 1000, 0.1, 0.01, 0.001, etc)

Decoys might still improve that pooling system though. As the payment is broken down, an address belonging to the sender and the output of the send will end up in a transaction together for each part. If the sender has received most or all of the coins to one address, or if someone watching knows multiple of the sender's addresses, they might be able to connect the source and destination by looking at if known addresses of the sender, and any output, are always in the same transaction with each other over the duration of the transaction. By using decoy sends, you can add in parts where the sender's addresses are seen in a block, but the intended destination is not, making analysis more difficult.

Neon001

member

Activity: 67

Merit: 10

Quote from: bigc1984 on February 22, 2014, 12:03:05 PM

You didn't read a word, did you?

Of course I did, but I'm failing to see the practical difference between simply handing a disproportionately large amount of the coins to a select few controlling entities, and awarding them the coins after a (mostly) trivial amount of mining. You can theorize about the altruistic and right-minded nature of the founders and early adopters (and I'm certainly not saying you're wrong), but those same arguments could just as easily support pre/instamined coins. At the end of the day, they control a significantly larger amount of coins than the amount of work they put into mining them relative to those that come later, regardless of what they choose to do with them.

I should note that I'm not begrudging them the right to do whatever the hell they want with their coin. I've never understood the people that bitch and moan about premined coins, like it was their God-given right to a fair shake for a particular coin. Maxcoin is a great recent example. Personally I just vote with my hash power and go elsewhere if I don't like a coin or its profitability is low. I like Darkcoin and would probably buy into it if I weren't risk averse and tend to avoid speculation.

eduffield

legendary

Activity: 1176

Merit: 1036

Dash Developer

Quote from: drakoin on February 22, 2014, 01:33:27 PM

Quote from: eduffield on January 18, 2014, 12:50:23 PM

Block reward is controlled by moore's law: (11111 / (((Difficulty+51)/6) ^ 2))

Please help me to understand.
http://explorer.darkcoin.io/block/00000000000295e2ecbbeb3b69f1ff9aaf24d2aebaefd534cfd22857e668d0a1

Block 22067:
Difficulty: 57527.019
Generation: 22 + 0.004 total fees

11111 / (((Difficulty+51)/6) ^ 2 =
11111 / (((57527.019+51)/6) ^ 2 =
= 1.207 * 10 ^ -4
!= 22

So I understand the block reward formula in the wrong way. Please explain.

Also, as 22 is a whole number, there needs to be an INT() or round(__,0) somewhere

thanks :-)

The GPU formula is a bit different, we switched to it when they became available via blockchain folk. It's 2222222/(((x+2600)/9)^2) and it's modeled to better represent the variance of difficulty that they bring to our mining pools. i.e, under CPUs a move from 1 to 10 difficulty was large, but under GPUs a move from 100 to 110 is nothing.

drakoin

hero member

Activity: 826

Merit: 1000

see my profile

Quote from: eduffield on January 18, 2014, 12:50:23 PM

Block reward is controlled by moore's law: (11111 / (((Difficulty+51)/6) ^ 2))

Please help me to understand.
http://explorer.darkcoin.io/block/00000000000295e2ecbbeb3b69f1ff9aaf24d2aebaefd534cfd22857e668d0a1

Block 22067:
Difficulty: 57527.019
Generation: 22 + 0.004 total fees

11111 / (((Difficulty+51)/6) ^ 2 =
11111 / (((57527.019+51)/6) ^ 2 =
= 1.207 * 10 ^ -4
!= 22

So I understand the block reward formula in the wrong way. Please explain.

Also, as 22 is a whole number, there needs to be an INT() or round(__,0) somewhere

thanks :-)

luke997

full member

Activity: 294

Merit: 100

Quote from: eduffield on February 22, 2014, 01:03:32 PM

Quote from: bigc1984 on February 22, 2014, 12:32:44 PM

Quote from: anonymousxx1503 on February 22, 2014, 12:04:33 PM

Quote from: Neon001 on February 22, 2014, 11:54:13 AM

lol, how is this effectively different than a premine?

Neither the devs or anyone else got a fixed % of coins at the beginning. Doge is not premined either but people who start at the very beginning get the benefit of making coins more easily. That's not how you see it?

That's the way it is supposed to be.

The block reward formula's + KGW == Whale Harpoon. It see it as being a critical part of the success we've seen thus far.

+1, best diff strategy around...

Kreativekrypto

full member

Activity: 196

Merit: 100

I just wish I knew about darkcoin earlier I'm not here to say insta or premine but it takes more hardware than I have to get anywhere. And let's be honest its not even 2 months old. I'd consider that an early adopter I just can barely get coins. I still help market dark though and hopefully will have good news for all of us soon! Some people are willing to give me some coin love.

HammerHedd

full member

Activity: 168

Merit: 100

Quote from: eduffield on February 22, 2014, 01:03:32 PM

Quote from: bigc1984 on February 22, 2014, 12:32:44 PM

Quote from: anonymousxx1503 on February 22, 2014, 12:04:33 PM

Quote from: Neon001 on February 22, 2014, 11:54:13 AM

lol, how is this effectively different than a premine?

Neither the devs or anyone else got a fixed % of coins at the beginning. Doge is not premined either but people who start at the very beginning get the benefit of making coins more easily. That's not how you see it?

That's the way it is supposed to be.

The block reward formula's + KGW == Whale Harpoon. It see it as being a critical part of the success we've seen thus far.

Something else to consider:
If (for some reason) the difficulty rate of DRK went back down because people stopped mining it, the algorithm encourage "coin CPR" because once the difficulty drops, the rewards become higher, hopefully encouraging more mining... and more interest...

buryat

newbie

Activity: 28

Merit: 0

https://i.imgur.com/Y8Jpsrq.png

I was able to push R9 270s' cores over 1200Mhz, stable and awesome.

eduffield

legendary

Activity: 1176

Merit: 1036

Dash Developer

Quote from: bigc1984 on February 22, 2014, 12:32:44 PM

Quote from: anonymousxx1503 on February 22, 2014, 12:04:33 PM

Quote from: Neon001 on February 22, 2014, 11:54:13 AM

lol, how is this effectively different than a premine?

Neither the devs or anyone else got a fixed % of coins at the beginning. Doge is not premined either but people who start at the very beginning get the benefit of making coins more easily. That's not how you see it?

That's the way it is supposed to be.

The block reward formula's + KGW == Whale Harpoon. It see it as being a critical part of the success we've seen thus far.

bigc1984

hero member

Activity: 658

Merit: 534

Quote from: anonymousxx1503 on February 22, 2014, 12:04:33 PM

Quote from: Neon001 on February 22, 2014, 11:54:13 AM

lol, how is this effectively different than a premine?

Neither the devs or anyone else got a fixed % of coins at the beginning. Doge is not premined either but people who start at the very beginning get the benefit of making coins more easily. That's not how you see it?

That's the way it is supposed to be.

child_harold

hero member

Activity: 812

Merit: 1000

Hi darkcoiners

I've overclocked my r9 290 21227-00-50G using TRIXX:

gpu core from 947 to 1000 and memory from 1250 to 1350

this gives me an extra 5% or so.

My question is whether people think this overclock is safe. i don't want to burn my card or shorten life expectancy too much.
There are no HW errors and everything SEEMS fine. Do you think it is?

Thanks in advance for the feedback.

DRKenstone

newbie

Activity: 5

Merit: 0

Quote from: HammerHedd on February 22, 2014, 01:00:09 AM

Quote from: raze182 on February 21, 2014, 11:58:58 PM

Quote from: DRKenstone on February 21, 2014, 11:31:48 PM

CPU DRK mining question:

I'm using the fantastic darkCoin-cpuminer-1.3-avx-aes-windows-binaries, with these settings:

Code:

minerd-corei7-avx -a X11 -o stratum+tcp://drk.lotterymining.com:4444 -u username -p password -t 8 -q

My proc: Intel Core i7-3770K, 3.50 GHz

I've been consistently getting 490+ khash/s with those settings, which I think is pretty great, but I'm not sure what others are getting with the same or similar proc.

I've been consistently getting lines like this:

Code:

[2014-02-21 22:01:06] accepted: 43/43 (100.00%), 490.02 khash/s (yay!!!)

But as of this morning and after about 50/50, I've been getting lines like this:

Code:

[2014-02-21 22:22:17] accepted: 77/78 (98.72%), 486.72 khash/s (yay!!!)

Concerning that percentage, 1) what does it mean exactly, and 2) why all of a sudden (after about 2 weeks worth of mining) am I getting less than 100%?

Thanks in advance for helping me with my n00b question! Wink

I'm also mining on a 3770K and getting more or less the same performance. If your percentage is less than 100% that means you've submitted a stale share(s) ("boooo" instead of "yay!!!"). In your case 1 stale share, hence the 77/78 instead of 78/78. Not anything to worry about unless the percentage drops significantly.

My 3770k does about the same.

And as raze said, one stale share is nothing to worry about. It's consistently stale shares that indicate a problem. A share can go "stale" if your pool issues you a hash to solve, then someone finds the block before your computed hash is submitted. this can also happen if there is some sort of network delay as well.

https://deepbit.net/stale.php

------

Many thanks to raze182, HammerHedd, and meebs for responding! I appreciate the insight.

This is frustrating, since last night I'm now down to 24 stales on average:

Code:

[2014-02-22 12:18:49] accepted: 1440/1467 (98.16%), 475.63 khash/s (yay!!!)

I'm going to try a couple of things today: 1) I have a new awesome crazy heatsink to replace my stock Intel with and then I'm going to gently overclock to see if that makes any difference and then 2) I'll try a different pool for a bit (but not blaming the current pool at all) to see if that changes things any also.

baka

full member

Activity: 196

Merit: 100

Quote from: Neon001 on February 22, 2014, 11:54:13 AM

lol, how is this effectively different than a premine?

It is different than a premine, it's a instamine done by the first miners/early adopters.

blueforever

member

Activity: 62

Merit: 10

Quote from: Queeq on February 22, 2014, 12:16:01 PM

Poloniex's alive.

YAY. Other pairs unfroze too. Staff probably just woke up.

Queeq

sr. member

Activity: 427

Merit: 250

Poloniex's alive.

anonymousxx1503

hero member

Activity: 532

Merit: 500

Quote from: Neon001 on February 22, 2014, 11:54:13 AM

lol, how is this effectively different than a premine?

Neither the devs or anyone else got a fixed % of coins at the beginning. Doge is not premined either but people who start at the very beginning get the benefit of making coins more easily. That's not how you see it?

Topic: [ANN][DASH] Dash (dash.org) | First Self-Funding Self-Governing Crypto Currency - page 6816. (Read 9723508 times)