Bitcoin Forum
May 01, 2024, 11:32:50 PM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
  Home Help Search Login Register More  
  Show Posts
Pages: « 1 ... 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 [199] 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 »
3961  Other / Archival / Re: Pictures of your mining rigs! on: August 22, 2011, 06:51:54 PM

Clocks:
5870 1: 985/300 458mhs 72C
5870 2: 940/300 437mhs 65C
5870 3: 940/300 437mhs 60C
5870 4: 940/300 437mhs 63C
6990 1: 920/600 413mhs 72C
6990 2: 920/600 413mhs 71C
6990 3: 920/600 413mhs 73C
6990 4: 920/600 413mhs 69C
All fans are at 85% and ambient room is around 25C

how did you get the 6990 memory to 600? flash bios and running linux?
Bios is flashed on both the 6990's & 5870's, I did have the 6990's sitting at 150hmz mem clock but I was worried it would cause damage so I upped them just in case.

do you run them under linux?
3962  Other / Archival / Re: Pictures of your mining rigs! on: August 22, 2011, 06:44:22 PM

Clocks:
5870 1: 985/300 458mhs 72C
5870 2: 940/300 437mhs 65C
5870 3: 940/300 437mhs 60C
5870 4: 940/300 437mhs 63C
6990 1: 920/600 413mhs 72C
6990 2: 920/600 413mhs 71C
6990 3: 920/600 413mhs 73C
6990 4: 920/600 413mhs 69C
All fans are at 85% and ambient room is around 25C

how did you get the 6990 memory to 600? flash bios and running linux?
3963  Bitcoin / Mining / Re: Pool hopping... ethical or not? on: August 22, 2011, 06:34:16 PM
I see there's still plenty of pontification here...

all these arguments against hoping are totally useless.  they are your "hurt" feeling.  that's all they are.

If you want to start having a real and useful debate, let's see some Terms Of Service or agreements from pools.

If a pool ain't got one, then you all just wasting your breath.

words like believe, belief, faith, common, implied, un-implied etc... etc.. don't mean squat without set rules that miners agree to.

no agreement = no rules.

Wait a minute, why is it that pools don't have them? Hmmm.....
3964  Bitcoin / Pools / Re: [~5000 Gh/s] DeepBit.net PPS+Prop,instant payouts, we pay for INVALID BLOCKS too on: August 22, 2011, 06:19:59 PM
What is a "low enough number of false positives" - meaning: How many % of legitimate users are you willing to screw over?!
No legitimate users are planned to be screwed at all.
So how is anyone who submits valid shares during a round not legitimate?


It is NOT "evil" to start and stop mining whenever you please! It is "evil" however to have a payout system that is mathematically proven to be flawed(!) and taking counter measures that will always also impact users that don't even know of pool hopping at all.
Bank robbery is just "an optimized approach to withdraw funds", yeah ? Smiley
It's not evil to start and stop mining whenever one pleases, but it's evil to deliberately use some automated techniques that are known to hurt other users and the pool itself.
Currently we don't have any counter-measures that can impact legitimate users.
Pool hopping is like using something like http://geizhals.at/ to compare online shops for best prices - why on earth do you compare it to bank robbery? Noone gets paid anything less or more than the proportional payout for the shares he/she/it submits. Pool hoppers ONLY gain more from this because other miners are "loyal" enough to mine knowingly at a loss during longer blocks, which is a bad economic decision. They are hurt by these decisions, not by pool hoppers. It is not even possible for anyone except you as the operator to determine exactly how far into a round you are, so for an average user it is not even possible to know whether he is mining at a loss or not! And THIS is called "protection" of your users?

Also @CanaryInTheMine:
Please watch your tone.

It is NOT trivial to identify hoppers without the possibility that some (un)fortunate users who did not inted to hop but show the same patterns get targeted too. It is actually impossible to do so unless you can identify hoppers via a user agent - something that would have been possible, but since another pool already tried that it's no longer an option.

Also as you said - creating an account is easy... and as long as CAPTCHAs can be solved cheaper than the potential gain from pool hopping, any counter measures on an account basis are useless.

Oh, btw. anti hopping measures are exactly what you don't want: Some exotic variation that will fuck you over if your miner goes down.

Do explain "Please watch your tone."  what is this? some sort of censorship?

I don't agree with your assessment of triviality to detect hopping.  Are you familiar with probabilities and statistics? Confidence intervals etc...?  I guess not...

counter measures are not useless.  they make you a harder target.

You misunderstood my exotic variation comment.  it's a direct reference to specific, currently in place payout algos on other pools.  On some pools that employ these exotic payout schemes, if your miner goes down (say half way into the round) you are SOL (more so depending on length of time).

my exotic variation was not about detection algorithms, but existing payout algos that attempt to discourage hoppers.

But ultimately, it is the pool's policy that has to drive this.  As long as the Terms Of Service of a pool do not prohibit hopping explicitly, I hate to break it to you all: hopping is OK.  If you want it out of your pool, set the rules first.

Lack of spelled out rules only creates debates and apparently the need to watch my tone.
3965  Other / CPU/GPU Bitcoin mining hardware / Re: What's better, a 5970 or 6990? on: August 22, 2011, 05:52:57 PM
I think I might get a dual GPU card next and I'm wondering what's better for overall use (mining, gaming, resale): a 5970 or 6990?  I'm looking at the HW comparison chart and it looks like the 5970 has more stream processors... 3200 vs. 3072, so theoretically it would be marginally better for mining.  They're both around the same price, when you can find them.  Resale looks to be about the same too, more or less.  Or should I just wait for the dual GPU HD 7XXX? Help me decide!

5970

or this 5830 for 1 BTC:
https://bitcointalk.org/index.php?topic=38000.0
3966  Other / CPU/GPU Bitcoin mining hardware / Re: only 315MHash/s with 950 core on 5870s on: August 22, 2011, 02:57:22 AM
your cards are bad.  just give them to me instead.
3967  Other / CPU/GPU Bitcoin mining hardware / Re: Powercolor AX6870X2 @ Newegg for $450 on: August 22, 2011, 02:56:43 AM
TO clarify the people asking about flashing 6870's bios and linux: it works fine. All 11 of my 6870's (Stupid hard to get 5830's here in soviet canukistan) are bios flashed and chomp along fine in linux, to increase the volts and fan curve and max OC. I did most of this before the new catalyst came out, letting you OC past bios limits.

So, the problem with the bios flash (at least for single 6870's, so I assume the 6870X2), is very much the windows driver, not the flash itself.

Side note, I have never dealt with NCIX US, but the canadian side I have a few times, and they are legit. No problems with hardware to know how good they are about RMA, though.
Thanks for the info!  I might be forced into linux after all...
3968  Other / CPU/GPU Bitcoin mining hardware / Re: Mining troubles 2 BTC bounty to help me solve on: August 22, 2011, 02:55:41 AM
I guess I am looking for some help from someone who has 7 GPU working on a windows 7 32 bit machine. Most know that certain versions of ccc work with up to 8 gpu, MSI 890fxa-gd70 can handle 4 video cards, but I would like an answer from someone with the same motherboard with6 -8 GPU running...if possible.for the time being, i am running linuxcoin and other than heat issues, all is going extremely well, but I want windows 7.
i have same mobo.
2 6990s + 6870 x2, so 6 GPUs.

win7 64 bit.
not sempron, i have some dual core chip, forget which one...
running 11.8, no issues.

it doesn't sound like power is an issue...  maybe it is the CPU having issues handling you6 6 GPUs...  it's pretty far fetched though...
linux and windows drivers will handle this CPU differently.

BTW, you should not have CCC installed at all. it interferes sometimes.  you only need the driver and SDK. nothing else.

another thing... are you running something like GPU-Z at the same time as mining and witnessing this behavior?
3969  Other / CPU/GPU Bitcoin mining hardware / Re: I think my PSU just died :( on: August 22, 2011, 02:47:19 AM
Unplug all cables from psu to PC.take a paper clip, and bend it into a U Put one end into a black wire (ground) and the other into the green wire.

WTF??? are you trying to electrocute our beloved mousepotato???

lol actually I think it was decent advice.  I remember reading about this method to daisy-chain multiple power supplies this way without having to plug them into the mobo. 

Anyway, I'm back from the store and ~$200 later (I bought some other stuff with the new PSU) I have a Corsair AX850 running smoothly and my miners mining away Smiley  I guess I'm going to try to RMA this Thermaltake and maybe when I get it back I'll keep it around as a spare.  Ooorrrr maybe build a new system for mining. 

Thanks for all the advice guys Smiley

Yeah, that makes sense for connecting multiple PSUs, even then, you shouldn't be using exposed wires/paper-clips... that's just asking for trouble. serious trouble.

Glad you solved your issue!  but I bet you ended up spending more money on that new PSU than it would cost online.  but in time of need, you gotta do what you gotta do... Smiley

I highly recommend to have a spare PSU (even a crappy one like 650W diabloteck for $30) that can be used for a week or two while you RMA the busted one.  The better thing to do would be to buy a good PSU when it's on sale online.  a bunch come up pretty often and with rebate you can get a good deal...

You have no idea how tempted I am to just bring up another miner using all the spare components I have... but you have to draw the line somewhere.... it's a vicious circle! Smiley
3970  Other / CPU/GPU Bitcoin mining hardware / Re: I think my PSU just died :( on: August 21, 2011, 09:47:36 PM
for mining in general, it's always good to have a spare PSU laying around...

Even picking up a cheap, used one on eBay or something is a good idea to have it available for situations like these...

then assuming you had a good quality PSU, rma it to the manufacturer for fixing
3971  Other / CPU/GPU Bitcoin mining hardware / Re: I think my PSU just died :( on: August 21, 2011, 09:44:42 PM
Unplug all cables from psu to PC.take a paper clip, and bend it into a U Put one end into a black wire (ground) and the other into the green wire.

WTF??? are you trying to electrocute our beloved mousepotato???
3972  Other / CPU/GPU Bitcoin mining hardware / Re: I think my PSU just died :( on: August 21, 2011, 07:49:29 PM
How do I test for a bad PSU?

you try another PSU

Yep.

You could also move the suspect PSU into your WoW machine and see if it dies too.

You can sniff the PSU and see if there is an odor of burnt electronics.

You can unplug everything from the PSU except the motherboard and remove all the addon cards from the motherboard and try again.



I plays WoW on a laptop Sad  I was hoping I could short some pins on my PSU and then maybe plug in an optical drive and see if it opens/closes the tray.

mousie...  there are no universal, set in stone, EASY steps to troubleshooting PSUs
the fastest, simplest way is to use a good PSU instead of your suspect.  then go from there.
3973  Other / CPU/GPU Bitcoin mining hardware / Re: I think my PSU just died :( on: August 21, 2011, 06:23:22 PM
How do I test for a bad PSU?

you try another PSU
3974  Bitcoin / Pools / Re: [~5000 Gh/s] DeepBit.net PPS+Prop,instant payouts, we pay for INVALID BLOCKS too on: August 21, 2011, 04:12:44 PM
What is this pool's policy on hopping?
Hoppers are not welcome here.

This policy is not yet added to the pool rules because my hopping-detection routine is not ready. When I'll be able to detect poolhoppers with low enough number of false positives, this policy may become forced.
I know that some people call proportional method "flawed", but since there are lots of users that like this method, I have to protect them from evil ones.
What is a "low enough number of false positives" - meaning: How many % of legitimate users are you willing to screw over?!

It is NOT "evil" to start and stop mining whenever you please! It is "evil" however to have a payout system that is mathematically proven to be flawed(!) and taking counter measures that will always also impact users that don't even know of pool hopping at all.

Many of your users are completely new and unaware of any 50% attacks, that other pools are also existing (because many forum threads that were started some time ago only mention deepbit, as it was nearly the only pool around) or that there are alternatives to prop + PPS. Please reconsider delaying stats + forcing policies that won't be able to only hit pool hoppers (and which most likely will be circumventable) and switching to a proven secure payout system instead.
PPLNS for example still lets users earn more if the pool has a luck streak (no crazy 0.07 BTC shares most likely though...), is dead simple to explain and is actually more "proportional" than prop., as it takes the hash rate into account more.


fuck this rant!  this is all bullshit.  i want a simple payout system, which we have here.  i don't want some exotic variations that will fuck me over if my miner goes down.  nor am i interested in debating this.

the "solution/s" , as I read about them on different post, are too frikin complicated.  thus never ready for prime time.
All in all, the simplest solution is the best one.

Simple solution:

- post the rules.  now anyone still hopping on deepbit is violating TOS.

- identify hopping behavior (very easy to do) and suspend the account.

- when the person logs in, tell them that their account has been flagged as hopping.

- give them another opportunity to Agree to TOS that spell out: "not going to hop"

- if they violate again. suspend the account.

since this would be automated, let them open a million accounts. and get a million accounts banned.  an automated system won't get tired.  a hopper, will on the other hand.
 
3975  Economy / Gambling / Re: Xtreme HD5830 Sapphire raffle on: August 21, 2011, 02:56:45 AM
You want more than $200 for a card worth less than $100? Good luck.

it's a raffle where 1 ticket is 1 BTC.
3976  Other / CPU/GPU Bitcoin mining hardware / Re: Cheap 5850s on: August 20, 2011, 10:50:12 PM
    Just found a website that states Radeon 5850s are under 140 Canadian.Don't forget to read my sig,lol.

it doesn't look like they're in stock.
3977  Other / CPU/GPU Bitcoin mining hardware / Re: Powercolor AX6870X2 @ Newegg for $450 on: August 20, 2011, 10:46:32 PM
Yikes!!

I've been keeping an eye on this:

http://www.newegg.com/Product/Product.aspx?Item=N82E16814131383R

it was pricey... I last checked a few hours ago and it was still over 200 dollars...

just checked now and wow! someone snatched it for 127... nice deal...
3978  Other / CPU/GPU Bitcoin mining hardware / Re: Mining troubles 2 BTC bounty to help me solve on: August 20, 2011, 08:34:54 PM
sempron 140


ohhhh.... unless you have a bit more powerfull chip to at least try to rule out if cpu is an issue i'm not sure what else to suggest...

the CPU bug is nasty....
3979  Other / CPU/GPU Bitcoin mining hardware / Re: Mining troubles 2 BTC bounty to help me solve on: August 20, 2011, 08:24:32 PM
uninstalling ccc did not help. I am going to install 11.6b.

if your CPU is multicore, try setting affinity of miners to different cores.  iff this is related to CPU issues.
3980  Other / CPU/GPU Bitcoin mining hardware / Re: Regarding the theoretical maximum Performance of GPUs on: August 20, 2011, 07:35:07 PM
You realize a bitcoinhash is *2* sha256 blocks operations, right?
Well, not exactly 2 thanks to some optimizations possible
you can drop the last 3 rounds completely (they don't change H), and lose part of the previous round (you only need the E output of the 4th-to-last)
Initial rounds can be optimized as well, as the last DWORD of hMerkleRoot and nTime/nBits don't change between loops, so you can drop the equivalent of ~3 rounds there as well.
Same thing goes for optimizing/precalculating parts of the W mangling, as we're feeding in quite a bit of constants.
Register access is basically free on GPUs (they mask reg r/w by pipelining 4 "threads" on the shader pipeline).
Ch() can be done in 1 cycle, and Maj() in 2.
Also, what LUT accesses? just hardcode the K constants in the instruction stream.

So while you came up with a somewhat reasonable result, you did so by pure chance using invalid assumptions and numbers.

ArtForz,  you are quite the legend on these forums...  Glad to see you here! Smiley
Pages: « 1 ... 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 [199] 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 »
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!