Bitcoin Forum
June 19, 2024, 08:54:22 PM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
  Home Help Search Login Register More  
  Show Posts
Pages: « 1 ... 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 [98] 99 100 101 102 103 104 105 106 »
1941  Alternate cryptocurrencies / Announcements (Altcoins) / Re: Myriad [1st Multi-PoW] Prypto Partnership --- Donate to Exhumed Race game on: July 16, 2014, 08:43:30 PM
Guys, we have uptrend now Smiley
If we reinforce it buy some good news like PolyMYR launch, we will attract a lot of new investors to MYR! Think about it.

Lookin' good. A backtest of 100 would present a very nice buying opportunity.
1942  Alternate cryptocurrencies / Announcements (Altcoins) / Re: [ANN] Diamond (DMD) Evolution v 2.0 | NEW wallet, 50% POS Active on: July 15, 2014, 11:58:15 PM
HR
I sent you a PM did you get it?
Poke

I was reading around and came across this, and, the answer now is, YES! Wink

The PM notifications just don't seem to get my attention. lol

Thanks again for the info. I'm sure it will help.


1943  Alternate cryptocurrencies / Announcements (Altcoins) / Re: Myriad [1st Multi-PoW] Prypto Partnership --- Donate to Exhumed Race game on: July 15, 2014, 11:36:11 PM


We've got a Myriad clone in the delivery room.


Here is the mining information you will need to help us test multi-algo DigiByte mining. Please feel free to ask questions & help us iron out all the issues before we roll out the multi-algo update (will be called version 3.0). We will release the addresses when the pool is properly configured. We are shooting for 11 PM UTC now. Stay tuned!


Mining Algorithm #1: DigiBitcoin (Sha256)
This is the original Bitcoin mining algorithm. We have chosen this algo as one of the 5 to tap into the market of used ASIC miners.
Test Pool Address:
.bat example:
Miner Download: Any CGMiner / ASIC mining software
Configuration Settings: https://en.bitcoin.it/wiki/Mining_hardware_comparison



Mining Algorithm #2: DigiLitecoin (Scrypt)
This is the original DigiByte mining algorithm and can be mined just like before.
Test Pool Address:
.bat example:
Miner Download: Any CGMiner v2.7.2 or lower
Configuration Settings: https://litecoin.info/Mining_hardware_comparison



Mining Algorithm #3: DigiGPU (Skein)
This is the most efficient GPU mining algo of all 5 and can be mined with a specialized cgminer download.
Test Pool Address:
.bat example:
GPU Miner Download: https://www.dropbox.com/s/uhbp9r6m8zzqnke/cgminer_skein.zip
GPU Miner Source: https://github.com/reorder/cgminer_skein
Configuration Settings: http://myriad.theblockexplorer.com/help.php


Mining Algorithm #4: Digi3to1 (Myr-Groestl)
Groestl can be mined either by a CPU or a GPU in roughly a 3:1 performance ratio. Also, it is noted some people have used Nvidia graphics cards to mine this algo, although we have not tested this out yet.
Test Pool Address:
.bat example:
CPU Miner Download: https://drive.google.com/file/d/0B9cvOfoOekSdUkwyQmhfWGhJMDg/edit?usp=sharing
CPU Miner Source: https://github.com/ig0tik3d/myriadcoin-groestl-cpuminer-v1.0b
GPU Miner Download: https://bitcointalk.org/index.php?topic=475795.0
GPU Miner Source: https://github.com/prettyhatemachine/sph-sgminer
CC Miner Download: https://github.com/cbuchner1/ccminer/releases
Configuration Settings: http://myriad.theblockexplorer.com/help.php



Mining Algorithm #5: DigiCPU (Qubit)
Qubit can be mined by CPU or GPU but it is primarily a CPU mined algorithm. Please report your feedback or thoughts on Qubit and how it performs for you.
Test Pool Address:
.bat example:
CPU Miner Download: https://docs.google.com/file/d/0B9cvOfoOekSdRHl4QW5wSHJsLU0/
CPU Miner Source: https://github.com/qubitcoin/QubitCoin-cpuminer
GPU Miner Download: https://docs.google.com/file/d/0B9cvOfoOekSdRHl4QW5wSHJsLU0/
GPU Miner Source:
Configuration Settings: http://myriad.theblockexplorer.com/help.php
1944  Alternate cryptocurrencies / Mining (Altcoins) / Re: [ANN][Groestl] grs-sgminer Optimized OpenCL kernel Groestlcoin/Diamondcoin miner on: July 15, 2014, 11:23:44 PM
What about the power consumption of the optimized and unoptimized kernels?

Does anyone have some numbers for us? Tongue

Without modifying any settings except the kernel this is what I get:

Standard kernel: 4 x 280x = 695 Watts >> 14.5 Mh per card
Optimized kernel: 4 x 280x = 918 Watts >> 5 Mh per card

This is the total wattage reading from the watt meter at the wall so it also includes the rest (mobo, ram, cpu etc) but what is interesting is the relative difference.

Also note that in order to get the best of the standard kernel I would need to modify certain parameter in order to "optimize" the hashing.

EDIT:

After I optimize parameters with the stantard kernel I get:

Standard kernel: 4 x 280x = 790 Watts >> 7.7 Mh per card

EDIT 2:

Noticed I inverted hashing values above. It was supposed to be:

Standard kernel: 4 x 280x = 695 Watts >> 5 Mh per card
Optimized kernel: 4 x 280x = 918 Watts >> 14.5 Mh per card



polanskiman,

I see you were monitoring this. How about your previous readings using sph-sgminer (both energy consumption and hashrates)? Much appreciated.


1945  Alternate cryptocurrencies / Mining (Altcoins) / Re: [ANN][Groestl] grs-sgminer Optimized OpenCL kernel Groestlcoin/Diamondcoin miner on: July 15, 2014, 10:44:28 PM
7850 @ 1100/100 and drivers 13.12 = 5 Mh/s
7950 @ 1080/100 and drivers 13.12 = 11 Mh/s

7850 @ 1100/100 and drivers 14.6b = 12.5 Mh/s
7950 @ 1080/100 and drivers 14.6b = 21 Mh/s

I heard people with 280x have approx 30 Mh/s...

EDIT : Myriadcoin-groestl

Yes, this seem quite low for the optimized kernel. I'm not sure about 30MH/s on R9 280X (I have 33MH/s on R9 280), but 21MH/s on 7950 seem to be low.

Please send your command line.

You are using -k myriadcoin-optimized, right?

Because if you are using just the normal -k myriadcoin-groestl, that's pretty much what you should be getting.

Thanks
That's my scrypt, i have 1x7850 and 3x7950

export DISPLAY=:0
export GPU_MAX_ALLOC_PERCENT=100
export GPU_USE_SYNC_OBJECTS=1

# 7850
aticonfig --adapter=0 --pplib-cmd "set fanspeed 0 34"
aticonfig --adapter=0 --odsc 1100,100 <-- 12.5 Mh/s optimized, 8 Mh/s normal kernel
# 7950
aticonfig --adapter=1 --odsc 1080,100 <-- 21 Mh/s optimized, 13 Mh/s normal kernel
aticonfig --adapter=2 --odsc 1050,100
aticonfig --adapter=3 --odsc 1050,100

/home/user/Downloads/grs-sgminer-master/sgminer -k myriadcoin-optimized -o stratum+tcp://cryptohunger.com:60002 -u Sbh4EmQz15hY4nSPvXVP9DszxBpinpUstv -p x -I 22 -g 1 -w 256

On the last miner i was using i had 17 Mh/s instead of 21 Mh/s for my 7950.


Have you documented the difference in AC consumption and temperatures?

1946  Alternate cryptocurrencies / Announcements (Altcoins) / Re: [ANN] Diamond (DMD) Evolution v 2.0 | NEW wallet, 50% POS Active on: July 15, 2014, 07:43:41 AM
Woo Hoo!  I got my DMD's in my wallet today!  It took a couple phone calls to them but they finally pulled through.  Once I had them on the phone they quickly figured out what happened.  I am a happy camper again!  I am going to test some more small purchases on Cryptsy and send to my wallet to see how it goes.  I did inform the gentleman I spoke with that we are now on 2.0.3.  He was not aware of the new wallet.

Pokeytex


Pokeytex, I'm still trying to get mine (they send me a periodic/idiotic e-mail saying they appreciate my patience, etc., but do nothing, and that after having told me over 6 weeks ago that a re-send had been approved); what phone number did you use, and who did you speak with?

PM me if you want to keep this below radar.

TIA

1947  Alternate cryptocurrencies / Announcements (Altcoins) / Re: ★★ DigiByte ★★ [DGB] ✈ DigiShield ✔ v2.9.1 ✔ Multi-Algo mining coming soon! on: July 15, 2014, 07:27:41 AM
We have talked with the Myriadcoin devs as well as other devs and it appears to not be possible to include groestl with NOMP. No one to date has got groestl running on a NOMP pool.

We are now in the process of getting Groestl set up on a second MPOS pool. We will try and get everything set up under the same front end but no guarantees.

We are able to solo mine all 5 algos, but for testing purposes we want to get everyone to link to a single pool so we have some very accurate network data when we run the 48 hour mining test. Will post another update tonight or tomorrow morning as we get closer to the test.


http://groestlcoin.biz/

1948  Alternate cryptocurrencies / Announcements (Altcoins) / Re: Myriad [1st Multi-PoW] Prypto Partnership --- Donate to Exhumed Race game on: July 15, 2014, 06:44:40 AM


What do they say in real estate? Location, location, location?

Would we say coin, coin, coin?

Probably not.

We'd probably say, coin, wallet, network.

That's our foundation, and without it, everything else is wasted time and will eventually fail.




Make sure you've got a good location before putting your sign up.


(My 'Chinese proverb' of the week. Smiley )
1949  Alternate cryptocurrencies / Announcements (Altcoins) / Re: Myriad [1st Multi-PoW] Prypto Partnership --- Donate to Exhumed Race game on: July 14, 2014, 07:42:47 AM
What would stop a botnet operator from making his own wallet implementation that bypassed said throttles? I feel like this would have to be apart of the coin itself.

You're probably right. Wallet hacking would be a huge vulnerability, but I don't know enough about the technical aspects to be able to intelligently comment on this. Hopefully we'll hear from others with more specific knowledge on the subject.

But the general idea would be to do as much as possible with the wallet, and only the absolutely necessary with the coin.



You don't even need to do any hacking. Just set up multiple headless linux VMs and install have multiple wallets. With a 32gb system you could have about 15 VMs (assuming 2gb per machine) and 15 wallets. So 15x what normal people would be getting...

Would tying the wallet to the users IP address help, as in the blockchain woulk only accept 1 miner from any IP?

That's an idea, but if people have multiple machines (laptop for work, old machine dedicated to downloads, the kids' computer, etc.) they should be able to use them all . . .

I wouldn't have too much of a problem with multiple VMs - the real 'horse power' of the 'rig' can't be broken into more parts than the whole so, well, if it's got some powerful graphic cards to break up into CPU equivalents (remember there's a throttling threshold that caps each VM), then maybe . . . but would it really be worth it? The reward differences aren't really equivalents - not by a long shot - the GPU still has a handsome advantage over the CPU (and ASIC over GPU) and would most likely be better off mining Skein or Groestl anyway in the vast majority of cases, and the operator breaking things up into VMs would probably be shooting him/herself in the foot by trying to get cute. In the best of cases it would be a close call, but the margin of error would always go to legitimate use.



A compromise then, say up to 5 devices from the same IP?  This way the 'family' is catered for but the geeks like us would find it harder to stand up 500 bots (although there would probably be a way round it).

Sounds good.


I also like e-mail registration/confirmation for each install. (I know that would kill anonymity, but that's a pipe dream that I think we'd better get out of our heads sooner than later, because sooner or later we know the IRS will have its ugly nose in the middle of everything and that sooner or later we'll be looking at forced transparency. You know what they say about negatives? Turn them into positives! MYR would also become the first truly transparent coin - my coins are my coins by golly, and I want my identity time stamped on each and every one of them!  Smiley )

1950  Alternate cryptocurrencies / Announcements (Altcoins) / Re: Vertcoin - First Scrypt N | First Stealth Address - Privacy without mixer on: July 14, 2014, 07:21:16 AM


Does FUD and flame cause price to fall?

or

Does a falling price cause FUD and flame?



It really isn't as simple as either one. It's a chicken or the egg argument. They are intimately connected, very symbiotic. When things are going good, everything builds together, and when things are going bad, we get reinforcing downward spirals - what is sometimes called the "snowball effect".



Oh, if life were only so simple that we could really believe in the "don't worry, be happy" meme.  Sad


1951  Alternate cryptocurrencies / Announcements (Altcoins) / Re: [ANN] Diamond (DMD) Evolution v 2.0 | NEW wallet, 50% POS Active on: July 13, 2014, 11:06:25 AM


Happy Birthday, keep up the good work, and best wishes for continued success!

1952  Alternate cryptocurrencies / Announcements (Altcoins) / Re: Myriad [1st Multi-PoW] Prypto Partnership --- Donate to Exhumed Race game on: July 12, 2014, 09:54:47 PM
What would stop a botnet operator from making his own wallet implementation that bypassed said throttles? I feel like this would have to be apart of the coin itself.

You're probably right. Wallet hacking would be a huge vulnerability, but I don't know enough about the technical aspects to be able to intelligently comment on this. Hopefully we'll hear from others with more specific knowledge on the subject.

But the general idea would be to do as much as possible with the wallet, and only the absolutely necessary with the coin.



You don't even need to do any hacking. Just set up multiple headless linux VMs and install have multiple wallets. With a 32gb system you could have about 15 VMs (assuming 2gb per machine) and 15 wallets. So 15x what normal people would be getting...

Would tying the wallet to the users IP address help, as in the blockchain woulk only accept 1 miner from any IP?

That's an idea, but if people have multiple machines (laptop for work, old machine dedicated to downloads, the kids' computer, etc.) they should be able to use them all . . .

I wouldn't have too much of a problem with multiple VMs - the real 'horse power' of the 'rig' can't be broken into more parts than the whole so, well, if it's got some powerful graphic cards to break up into CPU equivalents (remember there's a throttling threshold that caps each VM), then maybe . . . but would it really be worth it? The reward differences aren't really equivalents - not by a long shot - the GPU still has a handsome advantage over the CPU (and ASIC over GPU) and would most likely be better off mining Skein or Groestl anyway in the vast majority of cases, and the operator breaking things up into VMs would probably be shooting him/herself in the foot by trying to get cute. In the best of cases it would be a close call, but the margin of error would always go to legitimate use.

1953  Alternate cryptocurrencies / Announcements (Altcoins) / Re: Myriad [1st Multi-PoW] Prypto Partnership --- Donate to Exhumed Race game on: July 12, 2014, 04:48:03 PM
Do you check out the reddit much? There's a project underway you'd probably be interested in. It's basically a wallet with a built in p2p miner that mines directly to the wallet. Supposed to auto-config, etc, so you just open the wallet and click mine.

Read it more and more everyday. I'll check that out. Sounds very interesting, and very 'parallel'. Wink  Thanks.

Have you got a link?

1954  Alternate cryptocurrencies / Announcements (Altcoins) / Re: Myriad [1st Multi-PoW] Prypto Partnership --- Donate to Exhumed Race game on: July 12, 2014, 03:38:44 PM
What would stop a botnet operator from making his own wallet implementation that bypassed said throttles? I feel like this would have to be apart of the coin itself.

You're probably right. Wallet hacking would be a huge vulnerability, but I don't know enough about the technical aspects to be able to intelligently comment on this. Hopefully we'll hear from others with more specific knowledge on the subject.

But the general idea would be to do as much as possible with the wallet, and only the absolutely necessary with the coin.

1955  Alternate cryptocurrencies / Announcements (Altcoins) / Re: Myriad [1st Multi-PoW] Prypto Partnership --- Donate to Exhumed Race game on: July 12, 2014, 03:12:31 PM
I'm not a coder but I'm having a hard time imagining how you'd make it so the throttling would only be applied to one of the algorithms.  

I'm not a "coder" either, but I do have basic IT knowledge having worked with hardware, system and network administration, and web design/creation/administration for over 20 years (and I'm slowly learning C++ since that is the bedrock that everything is built on and will continue to be built on . . . so, someday, I might be able to at least tell you what's under the hood.  Cheesy ).

In-so-far as this idea goes, the basic idea would be to do everything you possibly can with the wallet - the wallet would control the throttling of the hashrate input. Theoretically, you could throw TH/s at it, but it would simply ignore everything over the threshold and only report/submit everything up to that limit. Everything else would simply be ignored, or thrown away as it were. The other algos would mine unrestrictedly. You would essentially leave current functions in place, and add a new, restricted mining function for Qubit (along with its sub-functions).

The only change I would envision for the coin itself would be the reward amounts. Everything else would be done with the wallet.

Add: you might have to put some sort of dynamic checksum into the coin code to protect against wallet hacking . . .
1956  Alternate cryptocurrencies / Announcements (Altcoins) / Re: Myriad [1st Multi-PoW] Prypto Partnership --- Donate to Exhumed Race game on: July 12, 2014, 02:36:03 PM
I'm not an expert but I imagine the botnet operator would have one wallet and use the botnet basically as a mining pool.

That's my understanding as well, and while it would still be possible, it wouldn't make sense since that one wallet would have the same throttling threshold as every wallet, meaning that the botnet operator's total hashrate could never exceed that threshold either - as for the coins potentially mined, they would number the same as what would be mined with his/her own computer.

One wallet, one address, and your CPU (or the throttling threshold in the case of someone who wants to try and pipe more through, like with a GPU for example - keep in mind that older graphic cards may actually be slightly more efficient than an older CPU, and not hit the throttling threshold, or barely exceed it).

1957  Alternate cryptocurrencies / Announcements (Altcoins) / Re: Myriad [1st Multi-PoW] Prypto Partnership --- Donate to Exhumed Race game on: July 12, 2014, 11:55:44 AM
I wouldn't say it's not impossible to address. Nothing is impossible. But so far, other than perhaps the coin primer linked to, coins that are easily cpu mined get raped by botnets literally dumping coins until there's no value left.

But, is it to be implied that a bot would download and install the wallet and then configure it to automatically start without the end user being aware??? Would this really be botnet vulnerable to begin with?

1958  Alternate cryptocurrencies / Announcements (Altcoins) / Re: Myriad [1st Multi-PoW] Prypto Partnership --- Donate to Exhumed Race game on: July 12, 2014, 11:35:40 AM
You ever hear of botnets? a zombie network of compromised computers would mine the majority of the coins leaving your average person getting almost nil.  

Yes I have (and usually in the same off handed manner).

Could you give everyone a quick run down on the theoretical vulnerabilities given my loosely defined specs (one miner per address, wallet restricted, throttled mining), and perhaps also on why such vulnerabilities would be impossible to address?

1959  Alternate cryptocurrencies / Announcements (Altcoins) / Re: Myriad [1st Multi-PoW] Prypto Partnership --- Donate to Exhumed Race game on: July 12, 2014, 08:32:06 AM


Here’s a Radical [Not So Novel] Idea (please don’t laugh)

Would you be interested hearing an idea that could take MYR ballistic and worldwide (in the truest sense of the word)? Well, sit back and read on if you answered yes.

This is an idea that could really cause MYR to go 'viral'. Can you imagine people from all over the world automatically mining MYR on startup of their PC? What could possibly get millions of people all over the world (including the underdeveloped world) doing that?

Let’s step back for some background and start out by remembering and taking into account some of my logical anti-ASIC arguments (I haven’t been too bad with my predictions in recent months either) and then move forward going one step further by looking at the opposite end of the spectrum.

Just like at one end of the spectrum we have an ever more concentrated and centralized network in combination with price under serious pressure, at the other end we also have an authentic widely distributed network with a stable and generally rising price, and if hardware specialization leads to the first, then, the opposite, namely, hardware generalization, should logically lead to the second.

Economies of scale are generally a good thing when referring to competitive markets; the only problem exists when those markets become concentrated (centralized) in a few hands, and then economies of scale can work against the common good. We need look no further than the fiat banking system to see what eventually happens.

One of the most important ideas at the heart of the crypto ethic was the idea that decentralization was key for creating a viable 'alternative' storage of wealth that could compete with the traditional fiat system.

Have we by some chance become sidetracked with our obsessions for short term profit? What with, hey, we’ve even got a website dedicated to this extreme distraction. What’s the name? Oh, yeah, here it is, Crypto Pump of course (how could it be named anything else?). http://cryptopump.com Was this what we all had in mind when starting out with cryptocurrencies?

Decentralization is all about making something as available as possible to as many people as possible. Are we really going in the right direction to accomplish that by favoring specialized and centralized economies of scale? Okay, a large market cap would be great for PR, but how long lived would it be if it’s not built on a real foundation. Selling out to short term specialization and centralization is nothing more than assured short term gain in exchange for long term pain.

So, if centralized specialization leads to ruin, then generalization, or decentralization as it were, should lead to system health and vitality – exactly what is desired for a solid and authentic vehicle for storage of wealth. And just what do I mean by “generalization”? I’m referring to making something as “generally” available as possible, and not just in words, but in real deeds as well. It’s one thing to say that even CPUs can mine, and it’s quite another altogether to say (and make real) that they can do so competitively (and we all know that the little guy can only be competitive in a “little guy world” where the rules are fair and designed to protect the little guy, and not one where major competitors are swallowing up what little guys are still left).

Have you guessed where I’m going with this?

CPU mining that is just as profitable as GPU mining, and also with ASIC?

Had you guessed?

Yes, CPU on par with ASIC!

If you do that, if you make MYR the cryptocurrency that everyone who owns a computer could competitively mine (seriously now, everyone), you’d have millions of users all over the world configuring their MYR wallets to launch on system startup!

Okay, there are some details to work out, like having a launch on boot setting in the wallet, and, of course, the automatic mine on launch 'plug and play' coding as well.

The wallet would need pre-configured mining parameters so that the new user could mine with a simple click, either solo, or better yet, with an official MYR worldwide P2P (worldwide really meaning worldwide with scores of active nodes all over the world).

The pre-configured, in wallet, miner would only mine using the CPU, or CPU equivalent that would be limited by a throttling mechanism that either matched the most powerful end-user CPU, or an average of the most common CPUs in active use (meaning that it would be possible to mine with a GPU miner, but never at a hashrate that exceeded the CPU throttling level per address, and never independent or outside of the wallet).

Rewards would look something like this:

Throttled Qubit CPU pre-configured wallet restricted mining = 4000 coins per block
Groestl = 450 coins per block
Skein = 450 coins per block
SHA-256 = 50 coins per block
Scrypt = 50 coins per block

And, with that, you'd put MYR on the world map as the first truly fair cryptocurrency in the world where everyone who owns a computer is looked at funny if their MYR wallet isn’t running 'in the background'.

How’s that for an idea for taking MYR mainstream?

You’re right. This would be even beyond mainstream!

Isn’t that what we really had in mind when getting involved in cryptos? Wink



Now, let all those specialized, short term profitability obsessed, miners tell us why this is a bad idea (and GPU miners as well, for that matter) . . . something which should also tell us a lot.


1960  Alternate cryptocurrencies / Announcements (Altcoins) / Re: Myriad [1st Multi-PoW] Prypto Partnership --- Donate to Exhumed Race game on: July 11, 2014, 04:45:51 PM
I would like to make this clear here for anyone interested. ASIC miners have the least dump risk of all. Right now the biggest dumping is being done by two entities:
1. multipools because that's what they do.
2. gpu miners because of the high costs of running a gpu mining operation.

Asic miners have the least power consumption and best $/mh ratio so thei have the smallest incentive to mine and dump at these prices.


This is just my reasoning if you feel I'm wrong please comment.


PS: I'm a gpu miner too so there's no need for me to defend asics but I just think that the current schema is the best possible solution to include everyone.

I would take issue with your assumption that ASIC miners have the smallest incentive to mine and dump at these prices. In fact, they are the only ones who can do so at a profit at these prices. I would suggest that the only long term holders among miners at this time would be the GPU miners, or how else could we justify their mining at a loss (short term loss in their minds since they are thinking long term). It would just make no rational economic sense at all to be mining and immediately selling at a loss. That's tantamount to giving your money away to strangers. The GPU miner who is looking to support GPU mining operations currently does so at a guaranteed loss and would be better advised to shut down if not holding for the long term.

My assumptions are based on classical economic theory: if you can produce something at a lower cost, you can also profitably sell it at a lower cost, meaning the price of the product adjusts lower accordingly.

And a 20:1 differential is nothing to shake a stick at.  Wink

(An adjusted block reward would also solve the multipool issue as well.)


As someone who mines both gpu and asic, I must say that my asics are currently making a KILLING with this coin in comparison to my gpu's.  From an investment standpoint, they've cost me about the same (~$2500 usd for asics, ~$2500 worth of gpus), but my asics clear ~25k myr a day while my gpus get around 7000 myr (roughed in numbers, just illustrating my point).

I agree with foodies' notion that we shouldn't really shut the door on any type of mining gear.  
I also think that we shouldn't change myriads block reward system (my prior post was talking about offset block rewards for MERGED coins, big distinction there).
But what about adjusting the difficulties?

edited numers



Difficulties seem to be joined at the hip. It's roughly a .0349:1 ratio per MH/s. Still think an adjusted reward for each algo is the easiest way to go . . .

Thanks for sharing your personal mining numbers. It's good to have corroborative data. Hopefully others will also post theirs.  Wink

Pages: « 1 ... 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 [98] 99 100 101 102 103 104 105 106 »
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!