Cryptocoin Mining?

Page 348 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Torn Mind

Lifer
Nov 25, 2012
11,902
2,716
136
What's up with using two threads? Aren't GPUs massively paralleled already? How does running mining in two main threads help there?

No clue, but I do get my best hashes with 2 threads on my Powercolor 7970s.

It's also doable on the 7850, but it doesn't do as well as just a single thread. I dunno, maybe there's so many more stream processors on a 7970 that there are untapped resources when using just a single thread?

I too would like to know, although the person who provides the explanation would probably be from someone similar in caliber to The Stilt in terms of GPU and mining software knowledge...not the easiest person to find...
 

Mark R

Diamond Member
Oct 9, 1999
8,513
16
81
There is a new kid on the block on the Middlecoin mining pool - Over the last 5 days, he has increased hashing power to 750 MH/s on scrypt, and his speed is still rising.

No one really knows what he's using, but a guy claiming to be him has posted some hints on the bitcointalk forum, suggesting that he's using ASIC miners. (Specifically, the Gridchip BTC/LTC miner - or more precisely about 15,000 of them).
 

IEC

Elite Member
Super Moderator
Jun 10, 2004
14,440
5,429
136
There is a new kid on the block on the Middlecoin mining pool - Over the last 5 days, he has increased hashing power to 750 MH/s on scrypt, and his speed is still rising.

No one really knows what he's using, but a guy claiming to be him has posted some hints on the bitcointalk forum, suggesting that he's using ASIC miners. (Specifically, the Gridchip BTC/LTC miner - or more precisely about 15,000 of them).

That seems highly unlikely considering those chips cost $65 each for a (supposedly working) engineering sample at 0.065 MH/s each. Not to mention he'd need to have a reference board for every 4 of them or make his own PCB to drive them.
 

njdevilsfan87

Platinum Member
Apr 19, 2007
2,331
251
126
I'm trying out this coin called "Noble" right now. It seems to have flown under the radar with COYE taking all of the attention. The only thing I don't like about it the 2% premine. But the dev seems to be handing out a lot. I figure I'll give it a quick go like I did with DOGE, 2-3 days of mining and then back to WDC.

https://bitcointalk.org/index.php?topic=402667.0

I did receive 20,000 of these for free from the dev so worst case just post and get some free coins. :thumbsup:
 
Last edited:

bob32768

Member
Feb 7, 2013
41
0
76
There is a new kid on the block on the Middlecoin mining pool - Over the last 5 days, he has increased hashing power to 750 MH/s on scrypt, and his speed is still rising.

No one really knows what he's using, but a guy claiming to be him has posted some hints on the bitcointalk forum, suggesting that he's using ASIC miners. (Specifically, the Gridchip BTC/LTC miner - or more precisely about 15,000 of them).

The one on the top of the list? Already mined 35 BTC, but how long has he been running?

That's more hashing power than the next 10 or so miners! What would that be in GPUs? Like 1,000 7970s? 280 Kilowatts? Yikes!
 

Attic

Diamond Member
Jan 9, 2010
4,282
2
76
Maybe that speed increase is from growing numbers of infected computers mining for his bitcoin wallet on middlecoin.
 

blastingcap

Diamond Member
Sep 16, 2010
6,654
5
76
That seems highly unlikely considering those chips cost $65 each for a (supposedly working) engineering sample at 0.065 MH/s each. Not to mention he'd need to have a reference board for every 4 of them or make his own PCB to drive them.

Engineering samples are not production chips. As tweaks are made and volume ramps up the cost per die should go way down.

65 kH/s for 4-5 watts is first-batch specs IIRC, 84 kH/s for second-batch Scrypt ASIC. Slap a bunch of them on a board with some RAM and you can get something like 3500 kH/s for 200 watts. Compare this to 600 kH/s for 200 watts for a video card. If you pay anything for electricity, ASIC Scrypts are worth way more than the equivalent amount of hashrate from video cards.

That said, I think any big hash farms are still on GPUs right now. Or botnets. Scrypt ASIC is in production as we speak but the finished products aren't quite here yet.
 

IEC

Elite Member
Super Moderator
Jun 10, 2004
14,440
5,429
136
Well the Chinese forums apparently believe Gridseed is real and some users claim to have gotten working samples up to 85 kH/s from the coffee shop meeting. I'll keep an eye on weibo too...
 

KlokWyze

Diamond Member
Sep 7, 2006
4,451
9
81
www.dogsonacid.com
No clue, but I do get my best hashes with 2 threads on my Powercolor 7970s.

It's also doable on the 7850, but it doesn't do as well as just a single thread. I dunno, maybe there's so many more stream processors on a 7970 that there are untapped resources when using just a single thread?

I too would like to know, although the person who provides the explanation would probably be from someone similar in caliber to The Stilt in terms of GPU and mining software knowledge...not the easiest person to find...

I get interesting results with the '-g 2' flag depending on the cards I am running.

7950 = crash
7970 = hash

 

geokilla

Platinum Member
Oct 14, 2006
2,012
3
81
Lol, I was able to get just under 600 Khashes/s on my the Powercolor 7970 I had with 850/800 and 850/900. I don't know if The Stilt's BIOS has done anything but change my default clocks to 955/1500 instead of 925/1375 though. I don't know if fglrx sucks or if my card needs different optimizations...but my best hashes are still around 680 Kh/s at 975 Mhz/1500 Mhz...

Find another BIOS. Your 7970 should be capable of at least 700Kh/s at those speeds. A lot of vBIOS had been requested before, so I suggest you just do a CTRL+F for Powercolor first before posting.

What's up with using two threads? Aren't GPUs massively paralleled already? How does running mining in two main threads help there?
It's GPU dependent I think.
 

Torn Mind

Lifer
Nov 25, 2012
11,902
2,716
136
Find another BIOS. Your 7970 should be capable of at least 700Kh/s at those speeds. A lot of vBIOS had been requested before, so I suggest you just do a CTRL+F for Powercolor first before posting.

Someone else had the exact same model as I did on page 2 at litecointalk, so I just used that one.
 

wand3r3r

Diamond Member
May 16, 2008
3,180
0
0
Have you guys gone solo?

Looking at a random coin on coinwarz:

Block Reward: 5.00
Blocks: 492,614
Block Time: 30.00 second(s)



Difficulty: 0.6282
Rate: 0.00005637

Just for a random example for some numbers, that appears to have a very low difficulty. If you would go solo and blocks are every 30 seconds do you have a chance to get any/many blocks? I assume pools can't be huge with such a low difficulty, but do they still usually get the coins that much faster due to odds?
 

Torn Mind

Lifer
Nov 25, 2012
11,902
2,716
136
But does it use your memory? The new vBIOS are further optimized as well.

I have Elpida memory, and The Stilt said that he optimized the BIOS for both. I found that a couple modules are visible without removing the heatsink.


However, I really think that I might be thermally limited as 980 Mhz does work for me right now but fans are blowing at the max right now. Might need a reapplication of thermal paste.


Hmm, after taking the card out, putting it back in, and restarting the system a few times, 1000 Mhz (and the aforementioned 980 Mhz) now works, whereas it didn't work at all in the past. I'm hovering around 700 Khashes with 1000 Mhz core and 1500 Mhz memclock.... The readings are a bit volatile. A bit more stable at 995. Can't go a lick above 1000. 1005 kills it...
 
Last edited:

geokilla

Platinum Member
Oct 14, 2006
2,012
3
81
So say if ASIC miners become mainstream for scrypt what's next for GPU mining?

You go back to Folding@Home.

However, I really think that I might be thermally limited as 980 Mhz does work for me right now but fans are blowing at the max right now. Might need a reapplication of thermal paste.

Hmm, after taking the card out, putting it back in, and restarting the system a few times, 1000 Mhz (and the aforementioned 980 Mhz) now works, whereas it didn't work at all in the past. I'm hovering around 700 Khashes with 1000 Mhz core and 1500 Mhz memclock.... The readings are a bit volatile. A bit more stable at 995. Can't go a lick above 1000. 1005 kills it...

You reapplied thermal paste? Right now, my 7950 tops out at 83C or so at 100% fan speed. It's ridiculous how bad XFX is... They used to be so good. What happened to them?
 

blastingcap

Diamond Member
Sep 16, 2010
6,654
5
76
So say if ASIC miners become mainstream for scrypt what's next for GPU mining?

There is only room for about 2 major cryptocoins which derive their value solely from network effects (the more people who use them, the more valuable they get), just like Facebook and Google+ dominate social networks. Similarly, if you invest in precious metals, you know that gold and silver dominate.

Since there is only room for 2 cryptocurrencies and we already have them (Bitcoin BTC and Litecoin LTC), the only way a third currency becomes viable is if it kills bitcoin and/or litecoin to make room for itself. Whatever kills BTC/LTC, it has to be a lot better than BTC/LTC. It will likely be radically different and may not even require "mining" for "block rewards" at all.

Even if somehow a third "mineable" coin emerges, there is no guarantee that it will work as well on current cards. As someone said recently in this thread, some algorithms work better on NV GPUs such as (allegedly) scrypt-jane. If a scrypt-jane PoW coin kills BTC and/or LTC, you will see a massive fall in AMD GPU prices and a massive rise in NV GPU prices.

But back to the main point: the world does not need yet another bitcoin clone. It needs something that gives bitcoin-like advantages (or more) and yet fixes bitcoin's drawbacks like blockchain bloat and "mining" which wastes a lot of energy. Something radically different, perhaps a Proof of Stake system unmineable by CPUs, GPUs, ASICs, or anything else, because the entire concept of mining for block rewards will be done away with as it is not necessary to ensure security in such a case.
 

wand3r3r

Diamond Member
May 16, 2008
3,180
0
0
There is only room for about 2 major cryptocoins which derive their value solely from network effects (the more people who use them, the more valuable they get), just like Facebook and Google+ dominate social networks. Similarly, if you invest in precious metals, you know that gold and silver dominate.

That very subjective statement is stated as fact. That is all.

(You could start with In my opinion, or there could be .. and this is why I think so)

I think people will hop on anything that they think will "make" money. I wouldn't look at the market today and start guessing about 5 years from now much less this years end. It's electronic tulip mania in the end and there's no telling which way the market will go much less if or when the whole trend will pass.
 

taq8ojh

Golden Member
Mar 2, 2013
1,296
1
81
Is LTC supposed to save less data on a disk or is it just so young it's only 2GB?
BTC might kill itself once we need to store 100GB of crap or so. Starting the client off classic HDD might take half a day then. It's already pretty painful right now when the data is only 15GB.
 

blastingcap

Diamond Member
Sep 16, 2010
6,654
5
76
That very subjective statement is stated as fact. That is all.

(You could start with In my opinion, or there could be .. and this is why I think so)

I think people will hop on anything that they think will "make" money. I wouldn't look at the market today and start guessing about 5 years from now much less this years end. It's electronic tulip mania in the end and there's no telling which way the market will go much less if or when the whole trend will pass.

Your dismissive "that is all" means you have nothing to refute what I wrote, imho. I said "about 2" which could mean 1, 2, 3, or maaaybe 4, but let's be real here, it probably means 1-3 (often 1-2) because cryptocurrency is in tech, and tech tends to consolidate around 1-3 (often 1-2) major networks. Economists have known about the network effect for ages, so have computer scientists. If you want to educate yourself, you can start by simple google search. But any self-respecting tech enthusiast already knows this.

Take Windows for instance. Why is Windows so freaking hard to dislodge as a desktop OS? Because it dominated OSes early so more software was written for it. More software written for it means more people gravitate towards the OS. It just kept snowballing and even today, Microsoft has a de facto monopoly over large portions of businesses' client PCs and PC gaming. Is there room for anything other than Windows? Technically yes. But Microsoft OSes utterly dominate, Apple is a distant second, and everything else like Linux is rounding error, despite Microsoft product not necessarily being the most technologically advanced, and despite how some alternatives like Linux are free to install.

If you talk about stuff outside of tech, you may find more than 1-3 major networks, for historical, cultural, political, and other reasons. For instance, there are more than 1-3 major languages (side note: the number of languages going extinct keeps growing... we're definitely consolidating around fewer and fewer languages over time). We also have more than 1-3 major currencies, though the USD is the world reserve currency by a mile, with the Euro attempting to consolidate its position as a distant second--when times get tough, people want to hold stuff like USD and Euros, just like if crypto goes down, people panic and want to either get out of crypto entirely, or to at least hold bitcoin and maybe litecoin.

But within the fast-moving world of tech, it's almost always the case that only 1-3 (often 1-2) networks dominate and that anything that replaces them has to be significantly better. So in tech you get stuff like:

Video Game Graphics API: Direct3D dominates, OpenGL a distant second
Social networks: Facebook dominates, Google+ a distant second
Client OS: Windows dominates, iOS a distant second
Mobile OS: iOS and Android dominate (BBRY was dominant pre-touchscreen; Apple knew it needed something "different" and the way they blended touchscreen with UI was that killer feature; Android did manage to catch up and overtake iOS but only because they were quick to scrap their existing plans and jump on the touchscreen UI bandwagon, and because a LOT of the industry was shut out of the iPhone deal between Apple and AT&T; since Apple did not license iOS to anyone else, that left the non-Apple hardware vendors and non-AT&T carriers desperate for a solution, which helped Android build enormous market share quickly... if Apple had licensed its technology we may have seen things happen differently, like iOS being a de facto monopoly with Android a distant second)
Mobile processors: ARM dominates, x86 a distant second
Desktop/Server processors: x86 dominates, ARM tablets attempting to make inroads
eReaders: Kindle dominates, Nook a distant second
Consoles: Playstation, Xbox, and Nintendo a distant and fading third that may pull a Sega and become a software firm only
Cameras: CaNikon dominate DSLR market share (since nobody wants to sell all their lenses just to go to a new mount; their competitors have never been able to catch up, just ask Sony how hard it's been to pry a few % market share away from CaNikon... that's why they are also trying to change the game by introducing Mirrorless cameras... it's hopeless to compete vs. CaNikon in DSLRs and doing so means you will forever be a distant third at best, but it's not so hopeless for Mirrorless which potentially offers unique advantages)

So yes, everything I write is "imho," but let's be real here, the network effect is very powerful, especially in tech, meaning each industry usually has 1-3 dominant standards, often just 1-2. If you want to make a big splash and steal market share away, you have to be sufficiently different in order to lure people to your network.

Fact of the matter is, nobody wants to buy multiple e-readers or consoles if they can help it--they would rather just buy once and run everything on that one. Nobody wants to buy 239865723468 cryptocurrencies if they can help it.

As long as bitcoin remains "good enough" with more and more companies willing to do business with it, then bitcoin will be hard to kill (meaning, hard for a rival cryptocurrency to displace). Every time a bitcoin ATM goes up, an exchange goes up, a company accepts bitcoin, etc., bitcoin becomes a little harder to kill. Litecoin doesn't fundamentally add anything new, so it's a minor miracle that it's worth anything at all. I think people just bid it up because they were late to bitcoin and hope litecoin will take off, or maybe they think it has some other small value, like as a backup to bitcoin if anything ever happens to bitcoin's blockchain.

Is LTC supposed to save less data on a disk or is it just so young it's only 2GB?
BTC might kill itself once we need to store 100GB of crap or so. Starting the client off classic HDD might take half a day then. It's already pretty painful right now when the data is only 15GB.

Exactly. Bitcoin has some flaws like bloated blockchain, Litecoin is going to have the same problem as it grows older. There have been calls to prune the blockchain but so far they've gone nowhere... it should be technically possible but so far the developers have not made it a top priority. A revolutionary new coin may be able to give bitcoin-like advantages and also address bitcoin shortcomings from the get-go. Such a coin has absolutely no guarantee of being on SHA256 or Scrypt, are possibly more efficient on NVidia hardware than AMD, and such a coin might not even have "mining" as we know it. Proof of Stake coins operate differently than Proof of Work coins like Bitcoin/Litecoin, so PoS coins don't require wasting tons of electricity on hashing power for security.
 
Last edited:

taq8ojh

Golden Member
Mar 2, 2013
1,296
1
81
Oh balls. I thought LTC was different in this regard, assuming by the name: LITEcoin. Is it basically a bitcoin ripoff with some minor aspects slightly changed then?

Bitcoin will never be widespread enough (as in between "normal" people) simply because of this. It's somewhat popular, but you still can't use it. you can make money off it, but it doesn't really serve the purpose of currency in the slightest. Sadly. And I don't think it ever will.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |