| 00:01 | ByteCoin | May I post your code to the forum. I believe it may suffer from a similar flaw to appamatto's |
| 00:01 | ByteCoin | please? |
| 00:01 | ByteCoin | I will post a reply with my findings... |
| 00:01 | ByteCoin | Will take some time though... |
| 00:01 | ByteCoin | Happy to discuss |
| 00:02 | ByteCoin | now |
| 00:02 | ArtForz | sure, whatever |
| 00:02 | ByteCoin | ish |
| 00:02 | ArtForz | I knwo it's inaccurate |
| 00:02 | ByteCoin | Cool. Thx |
| 00:02 | ArtForz | doesnt properly model race situations |
| 00:03 | ByteCoin | I'm suspicious that because you "generate blocks" in two places in the code it's not counting things properly... |
| 00:03 | ArtForz | huh? |
| 00:04 | ByteCoin | It's hard to explain at this time of night... |
| 00:05 | ByteCoin | Suffice to say, that if I submit a string of ACCAACCAAACC data in place of your random function, the results should be essentially the same regardless of whether I miss off one character from the start of the string. Do you agree? |
| 00:06 | ArtForz | errr. not quite |
| 00:06 | ByteCoin | The choice of whether the next block was generated by A or C has a different effect depending on where it occurs in the code. |
| 00:06 | ByteCoin | Artforz explain why not quite? |
| 00:07 | ArtForz | you're missing the case of next block found by A and C |
| 00:07 | ByteCoin | Ok. I thought you might mention that,... |
| 00:08 | ArtForz | = A and C both finding a block in one timestep without knowing about the other side |
| 00:08 | ByteCoin | So if I submit a string of ACCAABAACCCB etc where B indicates that the next block is found by A and C then you would be happy?? |
| 00:08 | ArtForz | thats also why increasing the timestep causes more races |
| 00:08 | ArtForz | yup |
| 00:09 | ArtForz | I'm completely restructuring it anyways |
| 00:09 | ByteCoin | If one symbol is missed off the start of any arbitrary sequence of plausible ACB characters then it should have a negligable effect on the overall outcome? |
| 00:09 | ArtForz | sure |
| 00:10 | ArtForz | 2 "A" populations for tiebreaking |
| 00:10 | ByteCoin | I will aim to show that your program is defective by finding a string where missing off one initial character produces very different results. |
| 00:10 | ArtForz | thats easy |
| 00:11 | ArtForz | "AC" |
| 00:11 | ByteCoin | You would accept that such a string would demonstrate a flaw in your program. We're talking LOOOONG strings |
| 00:11 | ArtForz | errr... no |
| 00:11 | ByteCoin | Strings millions of characters long. |
| 00:12 | ArtForz | lets see |
| 00:12 | ByteCoin | I thought you agreed to "if one symbol is missed off the start of any arbitrary sequence of plausible ACB characters then it should have a negligable effect on the outcome" |
| 00:13 | cosurgi | tell me: I need to patch bitcoind in order to allow communication between it and poclbm. Which version of bitcoind I have to patch? Where is the patch? |
| 00:13 | cosurgi | (OpenCL 2.1 on my ATI seems to work fine) |
| 00:13 | ArtForz | btw, who wins for a B ? |
| 00:13 | tcatm | cosurgi: no patch needed :) |
| 00:14 | cosurgi | oh! |
| 00:14 | ArtForz | and of course you need to have the proper weighting of events |
| 00:15 | ByteCoin | B means that both A_found_block and C_found_block are set to true at the same time |
| 00:15 | ByteCoin | B= both |
| 00:15 | ArtForz | well, as long as the relative occurences of ABC arent skewed, it should work |
| 00:16 | tcatm | what about just trying it on a custom testnet? |
| 00:16 | cosurgi | tcatm: is it going to work if both poclbm and bitcoind will be both calculating hashes? Or I need to disable it for bitcoind ? |
| 00:16 | ByteCoin | Artforz: Fair enough. I will ser to work tomorrow. |
| 00:16 | ByteCoin | set |
| 00:16 | tcatm | cosurgi: better disable it for bitcoind. It's only a waste of energy. |
| 00:17 | cosurgi | ok |
| 00:17 | cosurgi | -gen=0 |
| 00:18 | ByteCoin | Artforz: Just explain a little about what A_chain and C_chain represent? |
| 00:18 | ByteCoin | plz |
| 00:18 | ArtForz | current block count in the respective chains |
| 00:18 | ArtForz | total public chain is A_won_blocks + C_won_blocks |
| 00:20 | ByteCoin | Ok. I'm getting there.... |
| 00:21 | cosurgi | wow. |
| 00:22 | cosurgi | 205 Mhs/sec on ATI 5830 |
| 00:22 | cosurgi | ;;bc,calc 205820 |
| 00:22 | gribble | The average time to generate a block at 205820 Khps, given current difficulty of 12252.03471156 , is 2 days, 23 hours, 1 minute, and 10 seconds |
| 00:22 | cosurgi | is that a speed that you'd expect from ATI 5830 ? |
| 00:23 | tcatm | cosurgi: looks right |
| 00:23 | cosurgi | good :) |
| 00:23 | ArtForz | yep |
| 00:24 | cosurgi | thanks :) |
| 00:24 | ArtForz | http://pastebin.com/AvymGnMJ |
| 00:24 | ByteCoin | I take it back about your program suffering the same flaw.... |
| 00:25 | ByteCoin | Funny how everyone visualizes the same thing in different ways.... |
| 00:27 | cosurgi | --frames=1 gives 209 Mhs/s |
| 00:27 | cosurgi | anything else I could finetune ? |
| 00:27 | ArtForz | hurrm, thats actually pretty damn good |
| 00:27 | cosurgi | wow, now it's even 210 MHs/c |
| 00:28 | ArtForz | using m0s miner? |
| 00:28 | cosurgi | ...oscillating between 209 and 210 |
| 00:28 | cosurgi | yes |
| 00:28 | tcatm | overclocked? |
| 00:28 | cosurgi | ~/m0/m0mchil-poclbm-1e3b163$ |
| 00:28 | cosurgi | no. |
| 00:29 | ArtForz | thats... weird |
| 00:29 | cosurgi | I just bought that ATI today. I only: inserted that into box (had some problems with cables, etc..) upgraded to debian squeeze, installed ati drivers and opencl, and I've just run that miner. nothing else. |
| 00:29 | cosurgi | I don't even knopw how to overclock this :) |
| 00:30 | ArtForz | my table is 95% of theoretical peak |
| 00:30 | cosurgi | I can paste you some specs of this card |
| 00:30 | ArtForz | diablos miner is around there, m0s is usually a tad slower |
| 00:30 | cosurgi | just give me the commands you want me to run |
| 00:30 | cosurgi | it's all remote terminal, so no fancy GUI fglrc-stuff allowed |
| 00:30 | cosurgi | *fglrx-stuff |
| 00:31 | ArtForz | so m0s suddenly getting 97.3% theoretical peak is... unexpected |
| 00:31 | cosurgi | 01:00.0 VGA compatible controller: ATI Technologies Inc Radeon HD 5800 Series (Cypress LE) |
| 00:31 | cosurgi | ArtForz: where's diablo's miner? :) |
| 00:32 | ArtForz | http://www.bitcoin.org/smf/index.php?topic=1721.0 |
| 00:32 | cosurgi | thanks, checking.... :) |
| 00:32 | ArtForz | openCL miner written in java |
| 00:41 | fabianhjr | I got 0.10 BTC to spare. Anyone have a suggestion on what to spend them on? |
| 00:41 | AAA_awright | Donation? |
| 00:41 | AAA_awright | Give it back to the faucet you stole it from |
| 00:41 | donpdonp | 2oz of Four Loco |
| 00:41 | AAA_awright | ;) |
| 00:43 | fabianhjr | WTF are you talking about? I got them to spare, so I was thinking on getting some kiba's art. |
| 00:43 | AAA_awright | Woa |
| 00:43 | AAA_awright | ...kiba has art? |
| 00:43 | cosurgi | thats 2 cents actually :) |
| 00:44 | fabianhjr | http://www.bitcoin.org/smf/index.php?topic=1929.0 |
| 00:45 | kiba | a long running thread of kiba's art project |
| 00:45 | kiba | I barely colored the girl's head lol |
| 00:49 | fabianhjr | kiba: which one is your favourite.(I can only buy one atm) |
| 00:49 | doublec | !pool |
| 00:51 | kiba | fabianhjr: they are just the same art piece |
| 00:51 | kiba | and it's a work in progress |
| 00:52 | kiba | the latest image is the latest revision |
| 00:52 | kiba | and oldest image is the ugliest |
| 00:52 | fabianhjr | Oh! Do you have any finished work I could use to promote BitCoins(Planing on selling posters and tshirts) |
| 00:52 | doublec | ;;bc,calc 200000 |
| 00:52 | gribble | The average time to generate a block at 200000 Khps, given current difficulty of 12252.03471156 , is 3 days, 1 hour, 5 minutes, and 10 seconds |
| 00:53 | da2ce7 | Proposal Two: Named Addresses is done http://domainchain.org/wiki/doku.php?id=bitname 2/3rds complete. |
| 00:53 | AAA_awright | I wonder if I could sell music like that... |
| 00:55 | fabianhjr | well, I am actually setting up a pledge for an easy to use PrestaShop module to accept BitCoins as payments. You are welcomed to claim them anytime(Currently mining them) |
| 00:56 | fabianhjr | That could help many actually offer products with BitCoins |
| 00:57 | kiba | fabianhjr: no. If I was finished, I would already have a t-shirt in print production :P |
| 01:02 | fabianhjr | Ok. |
| 01:05 | xelister | USA payed people to troyan/backdoor opensource kernels, reports OpenBSD developer. This then probably is true also for Linux. Damn USA spies everywhere, what the hell. |
| 01:07 | da2ce7 | http://domainchain.org/wiki/doku.php?id=bitname |
| 01:07 | da2ce7 | done :) well i'm going to flesh it out a bit... what do you think? |
| 01:11 | xelister | http://bsd.slashdot.org/story/10/12/15/004235/FBI-Alleged-To-Have-Backdoored-OpenBSDs-IPSEC-Stack |
| 01:11 | fabianhjr | xelister, just sniff the traffic from any given box for a week after a kernel upgrade. Search for anything out of ordinary and boom. You got them. |
| 01:12 | fabianhjr | Tough, a commit passes trough a lot of revision before it gets pushed to the mainline, 0_o |
| 01:15 | cosurgi | hmm.. problems with diablo miner |
| 01:15 | cosurgi | Exception in thread "main" java.lang.UnsatisfiedLinkError: /home/bitcoin/diablo-miner/DiabloMiner/target/libs/natives/linux/liblwjgl.so: /usr/lib/jvm/java-6-sun-1.6.0.22/jre/lib/i386/libjawt.so: symbol awt_FreeDrawingSurface, version SUNWprivate_1.1 not defined in file libmawt.so with link time reference |
| 01:16 | cosurgi | export DISPLAY=:0 seems to have fixed this |
| 01:16 | cosurgi | [15.12.10 03:16:07] Started |
| 01:16 | cosurgi | [15.12.10 03:16:17] Added Cypress (#1) (14 CU, local work size of 256) |
| 01:16 | cosurgi | Waiting... |
| 01:16 | cosurgi | huh?? |
| 01:17 | cosurgi | 213624 khash/sec\n105517 |
| 01:18 | cosurgi | it is now oscillating between 212 and 214 |
| 01:18 | ArtForz | guess you got a fast card |
| 01:18 | cosurgi | :-D |
| 01:19 | ArtForz | happens sometimes, one of my 5970s is also doing this |
| 01:19 | cosurgi | is there a C++ OpenCL miner somewhere? |
| 01:19 | ArtForz | I dont think so |
| 01:20 | cosurgi | ok. so maybe I'll look at that java code and rewrite it into C++ |
| 01:20 | ArtForz | would be nice, needing JRE on miners is inconvenient |
| 01:20 | xelister | fabianhjr: what? |
| 01:21 | fabianhjr | ? |
| 01:21 | cosurgi | but I still don't get what miner is doing. I only know that it wants to find hashes that start with zeros. |
| 01:21 | xelister | fabianhjr: I dont get you ;) how is tracking devel box traffic going to help any shit? |
| 01:21 | cosurgi | care to say shortly exact procedure? |
| 01:21 | fabianhjr | xelister: if you sniff for a week the keylogger must send the log at least once. Then you will know for sure. |
| 01:21 | xelister | United States of American fucking government (allegedly) paid developer to 'donate' troyaned code to open source kernels like openbsd |
| 01:21 | fabianhjr | If you develop a Keylogger you want a way to access the logs. |
| 01:22 | xelister | fabianhjr: also, what you say is fucking stupid |
| 01:22 | fabianhjr | Why? |
| 01:22 | xelister | fabianhjr: I could even hide the keylogger report (if that would be the use of the troyan) in timing informations of regular trafic |
| 01:22 | xelister | no "watching" would detect such stenography |
| 01:22 | cosurgi | wow. it has just hit 216 MHs/s |
| 01:23 | cosurgi | now back to 213 |
| 01:23 | fabianhjr | xelister: if you stego something you still have to send it. |
| 01:23 | xelister | fabianhjr: no |
| 01:23 | xelister | fabianhjr: just variate timing of the NORMAL traffic a bit, and use this tiny time drifts as bits of your steno |
| 01:24 | fabianhjr | Yeah, you neeed a way to get the logs from the victim to the attacker. |
| 01:24 | xelister | <xelister> fabianhjr: just variate timing of the NORMAL traffic a bit, and use this tiny time drifts as bits of your steno |
| 01:24 | ArtForz | yes, but hiding something like that in open source code... good luck |
| 01:24 | xelister | perhaps this attack is too complex for you to understand though, fabianhjr |
| 01:25 | xelister | ArtForz: it happened. Attacker: USA gov, target: OpenBSD kernel. Already done (allagedly, they are now verifying how bad the backdoor was and how much that code is still in use today) |
| 01:25 | fabianhjr | What you are saying is that you can send data from point A to point B by sneding this data to a random point C and hope it routes to you? |
| 01:25 | fabianhjr | Oh! |
| 01:25 | xelister | ArtForz: http://bsd.slashdot.org/story/10/12/15/004235/FBI-Alleged-To-Have-Backdoored-OpenBSDs-IPSEC-Stack... |
| 01:25 | ArtForz | backdoor != keylogger |
| 01:26 | xelister | fabianhjr: duuh yeah. Just bribe/mob the ISP of either point A or C. That is what thugs and criminals like usa gov can do well |
| 01:26 | fabianhjr | You mean like them reading the actual routing? That would cause a massive lag and would be clearly notisable. |
| 01:26 | xelister | fabianhjr: boy you are so naive its even not funny, this FBI is doing for many years |
| 01:26 | ArtForz | hiding something like messing up a DH key exchange so it accepts certain "invalid" keys is a LOT easier than hiding a full blown keylogger |
| 01:27 | xelister | fabianhjr: actually, for example USA ISPs are REQUIRED BY LAW to prepare equipment for the FBI/other thugs to easly listen on your traffic, on ISP level |
| 01:27 | fabianhjr | ArtForz got a point |
| 01:27 | fabianhjr | xelister: I am not aware of USA regulations, I apologies. |
| 01:27 | xelister | also, for ISP logging all packages times/src/dst is triviall. Even I do it right now |
| 01:28 | xelister | and recording it is trivial, and of course introduces no lag |
| 01:28 | fabianhjr | Doesn't it cause a lot of lag? |
| 01:28 | xelister | I do it RIGHT NOW, of course it makes NO LAG |
| 01:28 | xelister | ok perhaps a 0.00001 sec lag |
| 01:28 | xelister | if the router would be a pentium I |
| 01:28 | fabianhjr | So, let me guess, you cannot filter, however, you make a copy of each packet and send it to another machine for loggging? |
| 01:29 | xelister | if computer A have backdoor in firewall stack, then I could steal any information from this computer |
| 01:29 | fabianhjr | Sorry, I am newb and I am catching up. |
| 01:30 | xelister | computer A sends the packets with a tiny time variance. or special way of 'random' afair cookies or port numbers and so on |
| 01:30 | xelister | and this tiny bits can be used as stenography |
| 01:30 | fabianhjr | Ok |
| 01:30 | fabianhjr | Yeah, I agree. |
| 01:30 | xelister | computer A sends to A's ISP, it send to backbone, it sends to other backbone etc |
| 01:30 | noagendamarket | I guess since its open source the governemnent can contribute code :)- |
| 01:31 | xelister | if A's ISP is for example in USA then it is required to allow fully logging. Then get logs, compare times, filter out noise, and you have the data |
| 01:31 | xelister | and that is just one of ideas |
| 01:31 | xelister | anyway, that are the tech details |
| 01:31 | xelister | the poins was, USA are fucking dick faggots, |
| 01:31 | noagendamarket | lol |
| 01:31 | xelister | for hiring coders to plant backdoors even in open source kernels |
| 01:31 | noagendamarket | You can do that when you can print all the cash you need |
| 01:31 | xelister | imho this is even bigger as USA idiots gathering DNA of world leaders and NATO leaders |
| 01:32 | xelister | anyway this stroy is developing now, on http://bsd.slashdot.org/story/10/12/15/004235/FBI-Alleged-To-Have-Backdoored-OpenBSDs-IPSEC-Stack and http://www.reddit.com/r/programming/comments/elw0x/allegations_regarding_openbsd_ipsec_fbi_backdo... |
| 01:32 | xelister | so if you think your computer is quite secure - think again - uncle sam may be screwing you in the ass right now with such criminal activities |
| 01:33 | noagendamarket | ffs |
| 01:33 | kiba | you believe everything you read, xelister ? |
| 01:33 | kiba | you need some mild skepticalism |
| 01:34 | xelister | kiba: I beliver the developer of OpenBSD kernel. And as he said, the story is unfolding right now, we will see where it leads |
| 01:34 | midnightmagic | ArtForz: interesting book; seek "Malicious Cryptography: Exposing Cryptovirology" by A. Young, M. Yung. |
| 01:35 | midnightmagic | could be as simple as a keygen width reduction. couple bits might do the trick. |
| 01:35 | xelister | kiba: but since we knew USA are ok with stuff, like, from killing natives, via posioning watter supply in France with LSD (resulting in few deaths too), via attacking many countries, to smear campaign against wikileaks and to war on own citizens... I think it is well within normal behaviour of this criminakls |
| 01:35 | xelister | criminals |
| 01:35 | Granttt | is this really any "shock", what took them so long, they announced war on internets a while ago: http://news.bbc.co.uk/2/hi/8073654.stm |
| 01:35 | noagendamarket | *rages |
| 01:35 | kiba | there's criminal behavior and feasialbity of doing said criminal behavior |
| 01:36 | kiba | national security is such a murky concept ;) |
| 01:36 | midnightmagic | feasibility of a keyspace usage reduction > small. :) |
| 01:36 | xelister | kiba: paying some coder to hide a backdoor? it happens all the time even by accident, why it should be not possible to do it on purpose |
| 01:36 | noagendamarket | "cyber tsar" |
| 01:36 | midnightmagic | "National Security" is a weapon by which the prole is oppressed. |
| 01:36 | xelister | Granttt: yeah, what that nigger is doing |
| 01:37 | kiba | if you want security, abolish all gun control laws |
| 01:37 | noagendamarket | xelister thats a bit much |
| 01:37 | midnightmagic | xelister: yeah, now you're just batshit foaming.. |
| 01:37 | xelister | Granttt: what are the people thinking now? "oh yea we voted the liberal, free, internet friendly guy hehe <3... oh, no... wait..... oh." |
| 01:37 | kiba | than if somebody want to invade you, they better be prepared for a long war against civilians |
| 01:37 | Granttt | he's just doing his job, the president is decentralized p2p engine of the lobbyists who sponsored him ;) |
| 01:37 | xelister | kiba: what do you mean? |
| 01:37 | midnightmagic | perhaps trollboy has outlived his useful presence in the channel. |
| 01:37 | kiba | xelister: the proliberation of arms amongst civilians.... |
| 01:37 | xelister | kiba: in USA guns are allowed though |
| 01:38 | noagendamarket | not everywhere |
| 01:38 | kiba | xelister: some people want to control who get to have guns... |
| 01:38 | midnightmagic | racist trolls aren't appreciated. |
| 01:38 | xelister | kiba: well I think allowing all guns and allowing much more things is good - in a sense that it makes it harder for grup of thugs (like gov) to get exclusive controll over power |
| 01:38 | kiba | xelister: that's the whole point |
| 01:39 | kiba | and you can't get something like a bunch of terrorists take over whole planes with a bunch of box cutter or something |
| 01:39 | xelister | midnightmagic: nigger please. I have nothing against black people. But I laugh at idiots that put so called "politicall correctness" above anything else. Example: it is o.k. to kill people based on their nationality or race (in example 'sand niggers' in Iraq), but it is bad to say the N- word. Wtf is up with that. |
| 01:40 | xelister | but that is just small digression from the topic of - how usa gov are dicks and can we trust then any computing system to be secure |
| 01:40 | kiba | real security is not about sending an occupation force to some faraway country |
| 01:40 | midnightmagic | xelister: nice pause before escalating. You are a troll. Your timing gives you away. |
| 01:40 | midnightmagic | Hey.. OPS.. wtf guys? |
| 01:40 | noagendamarket | fed |
| 01:40 | xelister | midnightmagic: there is a magic /ignore command |
| 01:40 | kiba | xelister is not a troll, just a very passionate person |
| 01:41 | xelister | I mean, even hardware CPUs are being backdoored |
| 01:41 | ArtForz | yep |
| 01:41 | xelister | now the supposedly most secure os, openbsd is (allagedly still) backdoored since 'ever' |
| 01:41 | ArtForz | at least xe, diablo and me like to rant about random crazy stuff from time to time |
| 01:41 | midnightmagic | sounds trolly to me. that's deliberate antagonism. |
| 01:41 | xelister | .me highfives ArtForz |
| 01:42 | xelister | I think nowdays man can trust a computing platform, if he builds it from scratch, from logical gates from raw materials |
| 01:42 | xelister | this is sad |
| 01:42 | noagendamarket | for the 90 per cent of people who dont compile all their own shit its bad |
| 01:43 | noagendamarket | thats all that matters |
| 01:43 | ArtForz | well, at least obsd has more than 65536 possible rsa keys :P |
| 01:43 | xelister | midnightmagic: btw if you fell strongly about problems of racial (or any other) hate, then act on that case where USA police beaten up a black girl 'because', or a Pollock wemen recently. Write your senator or soemthing |
| 01:43 | midnightmagic | okay then. xelister, douchebags claim anti-PC as an umbrella for their douchebaggery. That doesn't make them any less douchebaggy. |
| 01:43 | xelister | have to go now |
| 01:43 | midnightmagic | Also, I'm not American. |
| 01:44 | xelister | everyone should tweet/tell friends/or smth about this crap from USA (and other govs;) and about bitcoin by the way o/ bbl o/ |
| 01:44 | nanotube | that said... you guys might want to take the offtopic discussion to -discussion or something. :) |
| 01:44 | nanotube | though i guess it's fine in absence of actual dev-related talk happening at the same time... |
| 01:45 | ArtForz | this channel isn't #bitcoin-usually-ot ? |
| 01:45 | ArtForz | ;) |
| 01:46 | MT`AwAy | (if anyone wants to spend time reading php, the gawker/gizmodo hackers who released the user passwords to the world also included some sourcecodes) |
| 01:48 | nanotube | ArtForz: heh yea, de-facto it is ;) |
| 01:49 | Zarutian | xelister: that (build computer up from raw materials) or use mutually private computing on two machines http://unenumerated.blogspot.com/2007/08/mutually-private-computation.html |
| 01:50 | xelister | Zarutian: seems fun |
| 01:50 | xelister | Zarutian: randomized things like even say private key generator are problem though etc et |
| 01:50 | xelister | but in fact this is one solution |
| 01:50 | xelister | ok g2g cya |
| 01:57 | xelister | btw if it didnt worked with wikileaks dotations by BTC, then perhaps #openbsd would like.. some1 should ask |
| 02:02 | kiba | donation |
| 02:07 | MT`AwAy | lol |
| 02:11 | noagendamarket | I like that someone wants to send a satellite into space |
| 02:12 | noagendamarket | it will be cool if they run a bitcoin client from there |
| 02:12 | ArtForz | where else would you send a satellite? |
| 02:12 | noagendamarket | haha |
| 02:16 | noagendamarket | http://www.bitcoin.org/smf/index.php?topic=2282.0 i was referencing that thread |
| 02:17 | aximilation_ | hey all |
| 02:20 | Auctus | won't bitcoin eventually fill your hard drive with transaction history/blocks/whatever? |
| 02:22 | Lysacor | Well tonight I am officially one more to blame for the difficulty going up :P |
| 02:23 | Lysacor | ;;bc,calc 334000 |
| 02:23 | gribble | The average time to generate a block at 334000 Khps, given current difficulty of 12252.03471156 , is 1 day, 19 hours, 45 minutes, and 51 seconds |
| 02:23 | Lysacor | ;;bc,estimate |
| 02:23 | gribble | 13735.55418392 |
| 02:24 | noagendamarket | Auctus most of the transactions will be forgotten in future |
| 02:24 | ArtForz | doesn't look like we'll be doing x1.4 this time |
| 02:25 | ArtForz | we're already > 800 blocks into current difficulty, if rate of growth stays the same we're headed for ~x1.3 |
| 02:27 | aximilation_ | hmm, so if I leave it on for a couple days how much space would I be expecting to see it use? |
| 02:27 | Lysacor | well at least as long as we don't get another ~4000 increase like we got last round |
| 02:27 | ArtForz | which would nicely balance out the ~ x1.5 from 8078 to 12252 |
| 02:28 | Lysacor | agreed, at least I won't feel like I wasted money on the investment in this card |
| 02:28 | ArtForz | as before we were doing pretty close to x1.4 per difficulty |
| 02:29 | Lysacor | We shall see what happens in the coming days. You think there could be a significant swing? |
| 02:29 | ArtForz | kinda |
| 02:29 | ArtForz | looks like prices are rising again |
| 02:30 | Lysacor | yep, I should have had my sell price higher on mtgox, I got a bit excited on the last price spike earlier last week |
| 02:30 | noagendamarket | I think they should after all the positive publicity |
| 02:30 | ArtForz | yep |
| 02:30 | Lysacor | absolutely |
| 02:30 | noagendamarket | the EFF just posted another article including bitcoin |
| 02:31 | kiba | yeah, we know |
| 02:31 | aximilation_ | I *just* started running it tonight |
| 02:31 | ArtForz | which means more miners expecting a continued uptrend, which makes investing in mining equipment more likely |
| 02:31 | Lysacor | welcome to the fun aximilation_ :) |
| 02:31 | midnightmagic | sweet.. |
| 02:32 | aximilation_ | thanks :-D |
| 02:32 | aximilation_ | I'm not sure what to expect with it, but my computer sits idle most offff the time, so hey why not? |
| 02:32 | noagendamarket | newegg would make a bit if they sold gpus for bitcoins... |
| 02:32 | Lysacor | oh heck yeah they would... |
| 02:33 | noagendamarket | probably more than the people buying them lol |
| 02:33 | Lysacor | aximilation_: could be worse ways to burn electricity |
| 02:33 | aximilation_ | haha |
| 02:33 | ArtForz | if bitcoin becomes more widespread a few people at are gonna scratch their heads why people keep buying 5970s :P |
| 02:33 | noagendamarket | lol |
| 02:33 | Lysacor | and then the price will go up perhaps because of demand? |
| 02:33 | Lysacor | especially if there is a shortage |
| 02:33 | aximilation_ | ...and say hello to a bunch of incoming connections to 8333, lolllll |
| 02:34 | ArtForz | well, I hope lots of gamers are going for 6970/6990 |
| 02:34 | ArtForz | = more decent used 5970s on ebay |
| 02:34 | Lysacor | mmmm that is true.... 500 USD is still a bit high for me :P |
| 02:34 | Lysacor | even though it is 2x my Mhash... |
| 02:34 | noagendamarket | what is the difference in hash rates? |
| 02:34 | aximilation_ | what's this, using video cards to help crunch numbers? |
| 02:35 | ArtForz | yup |
| 02:35 | Lysacor | hehe GPU mining |
| 02:35 | ArtForz | thing is, it doesnt look like 6970 will be good for mining |
| 02:35 | kiba | noagendamarket: did ya try my http://www.soulplaying.com todo RPG? |
| 02:35 | ArtForz | my calc says about 244Mhash/s |
| 02:35 | Lysacor | oh really, I haven't seen the specs on it |
| 02:35 | ArtForz | well, rumored specs are 1536 ALUs, 880MHz |
| 02:35 | midnightmagic | the GPU is sometimes two orders of magnitude stronger than a moderately-fast CPU in mining. |
| 02:35 | ArtForz | release is... today |
| 02:35 | aximilation_ | oh nice |
| 02:36 | midnightmagic | and it's possible to stuff four (or more, I guess) into a single system. |
| 02:36 | ArtForz | well, or tommorow, depends on timezone |
| 02:36 | noagendamarket | well that would increase difficulty |
| 02:36 | kiba | wonders why nobody bother to try out |
| 02:36 | ArtForz | afaict NDA lift is 15. 12AM EST |
| 02:36 | aximilation_ | how does an AMDII x64 2.6ghz rank in the whole scheme of things? |
| 02:36 | Lysacor | well, we will see who gets to do the sucker.... err beta testing :P |
| 02:36 | kiba | it's a 15 days trial |
| 02:37 | ArtForz | but if the leaked specs are true, a 6970 is barely faster than a 5850 for mining |
| 02:37 | Lysacor | aximilation_ my wife has one... but won't let me test it on her box... but I estimate a decent amount... I think that those cpu's can do 4-way if I reacll |
| 02:38 | Lysacor | recall |
| 02:38 | Lysacor | the AMD Phenom II x4 right? |
| 02:38 | ArtForz | not too surprising, 58xx was exceptionally ALU-heavy for a GPU |
| 02:38 | aximilation_ | erm, not sure |
| 02:38 | aximilation_ | AMD Athlon(tm) II X4 620 Processor |
| 02:38 | midnightmagic | I can tell you what an AMD 1090T Black Edition can do.. |
| 02:38 | Lysacor | aaah, well it is on the lower end |
| 02:39 | ArtForz | Ahtlon II is still K10 arch, right? |
| 02:39 | Lysacor | I believe so |
| 02:39 | ArtForz | should get close to the same hash/MHz/core as a PhenomII |
| 02:39 | aximilation_ | not sure about it, I just know it runs well :-D |
| 02:39 | noagendamarket | damn 5970's are all over $700 AUD |
| 02:39 | ArtForz | as the miner loop should be small enough to fit in L1 |
| 02:39 | ArtForz | australians gets fucked when it comes to tech |
| 02:40 | midnightmagic | 3.2GHz, each core using the cpuminer can do about 2.3Mhash/sec; total is about 14Mhash/sec, or (coincidentally) about as fast as a GeForce 9600GT. |
| 02:40 | noagendamarket | I dont think its worth it just to mine |
| 02:40 | midnightmagic | 5970 here is $500 CAD. |
| 02:40 | Lysacor | anyone in the southern hemisphere from what I can tell, Chile is the same way... almost 100% markup |
| 02:41 | noagendamarket | probably cost heaps in postage from the states |
| 02:41 | aximilation_ | so when I run bitcoind is it automatically mining? |
| 02:41 | Lysacor | midnightmagic: good lord... decent cpu hash rate indeed |
| 02:41 | ArtForz | 5970 here WAS 380EUR incl. 19% VAT, now it's back up to 410 |
| 02:41 | noagendamarket | I blame you for pushing up demand lol |
| 02:42 | Lysacor | didn't you get some ungodly amount of 5970's art? |
| 02:42 | ArtForz | 24 |
| 02:42 | midnightmagic | Lysacor: that's with the 4way core in the cpuminer from.. hrm.. git://github.com/jgarzik/cpuminer.git |
| 02:42 | Lysacor | wow... yeah... I don't wanna see that electric bill |
| 02:42 | ArtForz | cancelled my order for 12 more, they pushed the shipping date out to january |
| 02:43 | Lysacor | there goes the difficulty, thanks for giving me at least a week to TRY to get some generated coins! :D |
| 02:43 | ArtForz | guess I made a e-tailer happy (ordered em back when they were 380 a pop) |
| 02:43 | Lysacor | damn... I can't find them for less than 499 anywhere, you must have done some research |
| 02:44 | ArtForz | thats 380 euros |
| 02:44 | Lysacor | aaah not USD |
| 02:44 | noagendamarket | I think artforz is plugged directly into a power station |
| 02:45 | noagendamarket | with an extension lead lol |
| 02:45 | Lysacor | ok sorry, gotta get my shit fiat currency out of my head by default... |
| 02:45 | ArtForz | I should write a FAQ |
| 02:45 | noagendamarket | haha |
| 02:45 | noagendamarket | an Arti FAQ ? |
| 02:45 | ArtForz | yup |
| 02:45 | Lysacor | who needs an extension cord, I got some 8 gauge copper waiting to whip up there! :P |
| 02:46 | noagendamarket | I have 3 phase power in my shed :)- |
| 02:46 | ArtForz | I have my own micro cogeneration plant (aka diesel gen replacing oil burner for central heating) |
| 02:46 | noagendamarket | is diesel generators efficient? |
| 02:47 | Lysacor | nice, helps to keep the grid usage lower, though is the diesel fuel producing a bit less expensive energy, compared to the grid at large? |
| 02:47 | ArtForz | if you use it to heat your house... yes |
| 02:47 | Lysacor | true |
| 02:47 | ArtForz | thats the whole idea about micro-cogen |
| 02:47 | ArtForz | instead of turning oil into heat, you turn oil into heat+power, and the power into heat, too |
| 02:47 | noagendamarket | move to antarctica and you could generate more |
| 02:47 | Lysacor | a lot more efficient |
| 02:47 | ArtForz | so you basically get free power out of the deal |
| 02:48 | ArtForz | well, except for the initial investment and higher maintenance of course |
| 02:48 | Lysacor | probably go from 20-30 percent efficiency into the 50-70 range? |
| 02:48 | midnightmagic | ArtForz: what's the power grid there cost per kWh? |
| 02:48 | ArtForz | about $0.30 |
| 02:48 | Lysacor | yeowch |
| 02:48 | ArtForz | a modern central heating oil burner is >90% efficient, so is a micro-cogen |
| 02:49 | Lysacor | very nice! |
| 02:49 | ArtForz | of course only about 30% of the cogens output is electric |
| 02:49 | midnightmagic | holy crap. it's $0.0875 here for tier-2 power (so, more than X kWh per month, which a farm like yours would definitely put me into) |
| 02:49 | Lysacor | I wish I could 1. afford one and 2. install one in such a way that my landlord wouldn't evict me on the spot |
| 02:50 | midnightmagic | and that's CAD, too.. |
| 02:50 | gribble | {"ticker":{"high":0.2468,"low":0.21,"vol":10188,"buy":0.2152,"sell":0.2459,"last":0.2467}} |
| 02:50 | Lysacor | ;;bc,mtgox |
| 02:50 | ArtForz | benefits of owning your own house ;) |
| 02:50 | ArtForz | yep, canada has about the cheapest power in the world |
| 02:51 | Lysacor | average here where I live is $0.10 |
| 02:51 | aximilation_ | funny you mention that, my workplace just relocated some web servers to canada partly due to the power reasons |
| 02:52 | midnightmagic | monthly cost for one of your 1.3KW at-the-plug machines is ~$82/month.. |
| 02:52 | midnightmagic | ArtForz: did you order the new machines because you want to stay ahead of the influx of GPU miners? |
| 02:52 | ArtForz | not really |
| 02:52 | midnightmagic | you're doing it for fun aren't you |
| 02:53 | ArtForz | yup, pretty much |
| 02:53 | kiba | for fun and profit! |
| 02:53 | ArtForz | well, is there a better reason? I like challenges :P |
| 02:53 | Lysacor | imagine if you could Crossfire all those things into one big array of cards... wow... |
| 02:53 | noagendamarket | lol |
| 02:53 | midnightmagic | No, no better reason. :) |
| 02:53 | noagendamarket | cloud server? |
| 02:54 | ArtForz | hrrrm, looks like theres another option |
| 02:54 | jb55 | I have been trying to compile bitcoin on ec2 gpu cluster instances with no avail |
| 02:54 | jb55 | damn centos |
| 02:54 | ArtForz | older FPGAs arent worth it, but xilinx spartan6 looks like it could be a winner for hash/W |
| 02:55 | Lysacor | jb55:wonder how quick they would boot you off of EC2 if you managed it? |
| 02:55 | ArtForz | still sucks for hash/$ though |
| 02:55 | midnightmagic | I would love to see a bunch of machines pay for themselves while I do R&D on distributed computing.. it would mean I could extend out my reach.. to.. well beyond what I thought would be within my reach as a hobbiest. |
| 02:55 | kiba | I make more money drawing than coding :( |
| 02:55 | ArtForz | so... copacobana with spartan6 anyone? |
| 02:55 | jb55 | Lysacor: I dont see why they would, that's what they're for. number crunching. |
| 02:55 | kiba | nobody want to play with my wwww.soulplaying.com :( |
| 02:56 | kiba | errr wwww.soulplaying.com |
| 02:56 | kiba | err..damn |
| 02:56 | kiba | www.soulplaying.com |
| 02:56 | noagendamarket | kiba im too lazy for that lol |
| 02:57 | kiba | don't play an MMO, play a real life RPG! |
| 02:57 | kiba | play your very soul! |
| 02:57 | Lysacor | so I am not selling my soul for a couple of trinkets right? :P |
| 02:57 | Lysacor | not that I had one to begin with :P |
| 02:57 | kiba | no, you're playing as yourself |
| 02:57 | midnightmagic | well the virtex almost certainly aren't worth it. |
| 02:57 | kiba | in improving your life skill |
| 02:58 | kiba | it's only like 3 bucks a month... |
| 02:58 | ArtForz | 120 spartan6-sx45 should be able to get about 4.5Ghash/s using ~400W |
| 02:59 | ArtForz | beats 5970s by about a factor of 6 on hash/W |
| 02:59 | jb55 | are there FPGA hobby kits? :D |
| 02:59 | ArtForz | yup |
| 02:59 | ArtForz | and horribly loses on hash/$ |
| 02:59 | ArtForz | the chips alone are ~$6.5k |
| 02:59 | aximilation_ | kiba do you run your own server there? |
| 03:00 | midnightmagic | you can get cheap spartan 6..! |
| 03:00 | ArtForz | add custom PCBs, PSU, custom case ... 9 grand |
| 03:00 | jb55 | ouch |
| 03:00 | ArtForz | well, it's also about as fast as 8 5970s |
| 03:00 | Lysacor | whew... a bit too rich for me |
| 03:00 | kiba | run what server? |
| 03:01 | kiba | mining is just a side income |
| 03:01 | ArtForz | so roughly half the hash/$ of 5970s |
| 03:01 | ArtForz | but 5x better hash/W |
| 03:01 | aximilation_ | soulplaying |
| 03:01 | kiba | does that mean you will sell 5970s for bitcoins? |
| 03:01 | ArtForz | but at least you dont need to order crazy amounts of em to get a decent price |
| 03:01 | kiba | aximilation_: no, I used heroku |
| 03:01 | midnightmagic | $14,280 for 120 spartan-6 XA6SLX75-2FGG484I from digikey. what part number are you looking at art? |
| 03:02 | aximilation_ | ah, I was going to say you should really catch the www and direct it to the non-www |
| 03:02 | ArtForz | you're checking out the -75s |
| 03:02 | ArtForz | XC6SLX45-2CSG324C |
| 03:02 | kiba | don't you mean non-www to www? |
| 03:02 | ArtForz | digikey 122-1673-ND |
| 03:03 | aximilation_ | yea, or that |
| 03:03 | Lysacor | thinking about the mtgox market... with all of those standing ASK orders for .32+ in 1000 coin blocks... that might potentially stunt the market |
| 03:03 | Lysacor | but, not knowing the historical market depth, it is an uneducated opinion at best |
| 03:03 | midnightmagic | that's half the onboard logic units.. but cheaper overall. you always seem to find the maximum economic wins. LOL how long have you been looking for fpga anyway? :) |
| 03:03 | ArtForz | not too long |
| 03:04 | ArtForz | but I was looking at FPGA to prototype my ASIC design in hardware |
| 03:04 | ArtForz | turns out my VHDL synthesizes to something with decent performance on virtex5 and spartan6 |
| 03:04 | Lysacor | kiba: next paycheck if I am looking in a good position, I may try your site, couldn't hurt |
| 03:05 | midnightmagic | can I be your disciple? |
| 03:06 | Lysacor | Art, be careful, looks like you are inadvertantly starting your own religion here :P |
| 03:06 | EvanR | i just heard about bitcoin today. how long has this operation been going on? |
| 03:06 | kiba | operation? what operation? |
| 03:06 | EvanR | bitcoin |
| 03:06 | kiba | bitcoin is not an operation |
| 03:07 | midnightmagic | it's a religion! |
| 03:07 | EvanR | lol |
| 03:07 | EvanR | or is it dead, in the final stages and i just came in at the end |
| 03:07 | mndrix_ | EvanR: first block was in Jan 2009, I think: http://blockexplorer.com/b/0 |
| 03:07 | ArtForz | it's a movement! |
| 03:07 | Lysacor | From what I know, it has been going on for some time |
| 03:07 | Lysacor | nowhere near dead |
| 03:07 | Lysacor | very much growing |
| 03:07 | kiba | it's been growing |
| 03:07 | kiba | December is going to smash record! |
| 03:08 | midnightmagic | And not the kind that you flush! |
| 03:08 | kiba | oh wait, we already did |
| 03:08 | EvanR | i tried to invent this a few months ago |
| 03:08 | ArtForz | merely pining for the fjords *ducks* |
| 03:08 | kiba | but we will smash more record! |
| 03:08 | EvanR | when reading about digital signatures |
| 03:08 | EvanR | i didnt get anywhere |
| 03:08 | kiba | not suprising |
| 03:08 | EvanR | ive been looking for a place to store my money |
| 03:08 | EvanR | i dont trust banks |
| 03:08 | ArtForz | the real genius of bitcoin is the distributed timestamping system |
| 03:09 | kiba | oh yeah |
| 03:09 | midnightmagic | not really a great economy to store large volumes of cash in.. :) |
| 03:09 | kiba | Satoshi is really a girl |
| 03:09 | kiba | and her name is not Satoshi |
| 03:09 | midnightmagic | Shhh.. don't tell him all the secrets right away. |
| 03:09 | ArtForz | and using mining/transaction fees as an incentive to keep enough CPU power on it |
| 03:10 | jb55 | EvanR: same, right now I have my wallet inside of a truecrypt volume sitting in my dropbox :D |
| 03:10 | EvanR | lol |
| 03:10 | midnightmagic | First he has to punch his fingers through the gravel for six months! |
| 03:10 | Lysacor | No, no, it is walk through glass covered in margarita salt... |
| 03:10 | EvanR | i looked at the server, looks like an iphone app waiting to happen |
| 03:10 | midnightmagic | Err.. I mean mine fruitlessly on his GPU for six months! |
| 03:10 | Lysacor | broken glass |
| 03:11 | kiba | what server, EvanR? |
| 03:11 | EvanR | bitcoind |
| 03:11 | Lysacor | possible, but I doubt the carriers would like 8333 traffic trickling into the network :P |
| 03:11 | midnightmagic | maybe as a controller via the JSON interface! that would be kind of neat. just make sure that other people with the app can move bitcoins to one another without manually typing in long strings of bitcoin addresses and you're good! |
| 03:11 | EvanR | Lysacor: it would be port 80 |
| 03:12 | EvanR | er |
| 03:12 | EvanR | https whatever it is |
| 03:12 | Lysacor | 443 |
| 03:12 | Lysacor | :) |
| 03:12 | EvanR | midnightmagic: nicknames to the rescue? |
| 03:12 | aximilation_ | EvanR: I just set up bitcoind about half an hour or so ago myself |
| 03:13 | kiba | newbies. |
| 03:13 | EvanR | i was about to, but realized i didnt have a dedicated server |
| 03:13 | EvanR | and the server i was on wasnt really mine |
| 03:13 | EvanR | so i stopped |
| 03:13 | aximilation_ | run it on a home computer :-D |
| 03:13 | EvanR | i actually dont have interne |
| 03:13 | EvanR | im stealing wireless |
| 03:13 | aximilation_ | that works |
| 03:13 | kiba | I need to do something about backup and encrypting my wallet |
| 03:13 | kiba | really paranoid |
| 03:13 | EvanR | bitcoin still works |
| 03:14 | Lysacor | you are "borrowing", stealing indicates malicious intent :P |
| 03:14 | aximilation_ | might be more difficult to forward 8333, but hey |
| 03:14 | Lysacor | borrowing without intent of return... |
| 03:14 | aximilation_ | "using" |
| 03:14 | EvanR | stealing indicates ironic abuse of loaded words |
| 03:14 | EvanR | sharing! |
| 03:14 | aximilation_ | neither borrowing nor stealing could apply, as there is no way you could give the internet back |
| 03:14 | Lysacor | nice! |
| 03:14 | ArtForz | how can you steal wireless? |
| 03:14 | kiba | ArtForz: you torrent through it |
| 03:15 | kiba | depriving the owner of his bandwidth! |
| 03:15 | ArtForz | aka "hey, I didn't ask for my neighbor to broadcast his radio waves all over my place" |
| 03:15 | Lysacor | yeah, saturate the pipe... that is where QoS comes in handy :P |
| 03:15 | aximilation_ | it's really not much different than "stealing" music from an open air concert by listening from outside the grounds |
| 03:15 | Lysacor | true |
| 03:15 | EvanR | so am i reading this dashboard correctly, 'we are on block 93691' ? |
| 03:15 | ArtForz | ;;bc,blocks |
| 03:15 | gribble | 97613 |
| 03:15 | EvanR | i am out of date? |
| 03:15 | Lysacor | actually, you are a few blocks behind, a bit more to download |
| 03:15 | EvanR | or this isnt a global number |
| 03:15 | ArtForz | no, it's a global number |
| 03:16 | Lysacor | nope, globally it is 97613 |
| 03:16 | Lysacor | every client shares the same block chain database |
| 03:16 | kiba | grow, bitcoin, grow! |
| 03:16 | EvanR | is 100000 a special milestone or something? |
| 03:16 | Lysacor | it is shared between clients |
| 03:16 | kiba | no |
| 03:16 | ArtForz | if you don't want people "stealing" your WiFi, turn on encryption or get a lot of tin foil. |
| 03:16 | EvanR | ok |
| 03:16 | aximilation_ | "Unauthorized Use" |
| 03:16 | Lysacor | tinfoil hats!! |
| 03:17 | ArtForz | well, *someone* set it up as a free-for-all |
| 03:17 | EvanR | also ironic, the guys password is "money money" |
| 03:17 | ArtForz | lol |
| 03:17 | aximilation_ | password for what? |
| 03:17 | EvanR | wireless |
| 03:17 | aximilation_ | how'd you come across the password? |
| 03:17 | EvanR | not hard to guess |
| 03:17 | aximilation_ | lol |
| 03:17 | kiba | he love money? |
| 03:18 | ArtForz | btw, huge fuckup of a german ISP, their router/APs have WPA2 turned on by default, turns out the default WPA2 key contains a whopping 24 bits of entropy |
| 03:18 | aximilation_ | in that case you're in a slightly "risier" position as it's not completely open to the public |
| 03:18 | EvanR | he told me the password |
| 03:18 | aximilation_ | oh |
| 03:18 | ArtForz | well, then its not stealing now, is it? |
| 03:18 | aximilation_ | nvm then, telling is permission then |
| 03:18 | EvanR | never was |
| 03:18 | aximilation_ | or close enough |
| 03:19 | Lysacor | backtrack is a wonderful tool... |
| 03:19 | EvanR | it says 1khash/s |
| 03:19 | aximilation_ | you know this whole discussion started because you said you were stealing wireless... |
| 03:19 | Lysacor | great for wifi key discovery and breaking |
| 03:19 | ArtForz | btw, 24dBi parabolics are fun |
| 03:19 | Lysacor | and attacking other people's AP's |
| 03:20 | Lysacor | what range you manage to get on it Art? |
| 03:20 | ArtForz | we got about 5 miles out of it with direct LOS |
| 03:20 | EvanR | aximilation_: not my fault |
| 03:20 | EvanR | i think i got all the blocks |
| 03:20 | Lysacor | very nice! |
| 03:20 | EvanR | 97614 |
| 03:21 | Lysacor | yep, you are caught up now EvanR |
| 03:21 | EvanR | 474 khash/s |
| 03:21 | ArtForz | and the transmitters turned down to lowest output |
| 03:21 | ArtForz | stupid ETSI EIRP rules |
| 03:21 | EvanR | its the code of the radio operator, use lowest power necessary to carry out communication ;) |
| 03:22 | Lysacor | true, but ETSI... FCC... they all over-regulate the spectrum to some degree... |
| 03:22 | EvanR | ignore numbers stations and best of hip hop and r&b FM stations |
| 03:22 | ArtForz | yeah, but seriously, 1W EIRP? |
| 03:23 | Lysacor | imagine if you could go 50 percent higher, what the results would be |
| 03:23 | EvanR | what is this generate coins this, does it randomly give you 1 coin? or some other amount? how often? |
| 03:23 | Lysacor | without risk of fines... |
| 03:23 | EvanR | this->thing |
| 03:23 | ArtForz | 50 percent? why not an order of magnitude |
| 03:23 | Lysacor | mmmm true |
| 03:24 | EvanR | 10W is insane |
| 03:24 | ArtForz | 10W EIRP on a 23dBi isn't exactly a lot of output |
| 03:24 | ArtForz | take line and connector losses into account and you're looking at 100mW tx output |
| 03:24 | Lysacor | so, bitcoin generates coins when you solve a given hash of a block, as time goes on the number of coins earned by solving blocks are less and less, (bitcoin.org details it all) |
| 03:25 | Lysacor | as the number of blocks in the collective block chain increases |
| 03:25 | ArtForz | yep, first halving should be late 2012/early 2013 |
| 03:25 | Lysacor | right now you get 50 per solved block |
| 03:25 | EvanR | oh |
| 03:25 | Lysacor | BUT, you can join a pooled mining effort |
| 03:26 | Lysacor | http://www.bitcoin.org/smf/index.php?topic=2027.0 |
| 03:26 | EvanR | all nodes are solving simultaneously, and whoever gets there first gets the cash? |
| 03:26 | EvanR | then you start over? |
| 03:26 | Lysacor | yep |
| 03:26 | EvanR | cant they solve in parallel? or would that be defeating the purpose somehow |
| 03:27 | Lysacor | the pooled mining effort combines the hashing power of multiple systems, to increase the likelihood of the group solving the block. As that one is setup, you get a number of coins (and they can be in decimal values) based on the amount of hash work contributed to the pool |
| 03:27 | Lysacor | the more you do compared to the rest, the larger your share of the block solved |
| 03:28 | Lysacor | the pooled mining effort has solved two blocks thus far |
| 03:28 | MT`AwAy | and your bitcoin client will throw away as soon as it can any sub-cent coin you've earned |
| 03:28 | EvanR | thats a question i was having, so 0.30 is a valid number of coins? |
| 03:28 | Lysacor | exactly |
| 03:28 | Lysacor | it certainly is |
| 03:28 | Lysacor | there are other ways to get bitcoins |
| 03:28 | EvanR | i think its also valid in json, but in a javascript engine it wont be represented as that |
| 03:28 | Lysacor | namely through trade, through a site like bitfaucet |
| 03:28 | doublec | MT`AwAy, the bitcoin client doesn't throw away sub-cent coins |
| 03:28 | Lysacor | etc... etc... |
| 03:29 | ArtForz | doublec: it does |
| 03:29 | MT`AwAy | doublec: it sure does, sends them as "fee" |
| 03:29 | aximilation_ | I got some from bitfaucet :-D |
| 03:29 | doublec | So if I have 0.001 sitting in my account it will vanish without me doing anything? |
| 03:30 | MT`AwAy | doublec: if you send coins to someone, the 0.001 part will be added to "in" automatically, but not to "out" |
| 03:30 | doublec | why does it do that? |
| 03:30 | MT`AwAy | so anyone who takes your transaction in and solves the block will get your 0.001 |
| 03:30 | EvanR | lol free bitcoins |
| 03:30 | MT`AwAy | doublec: ask Satoshi, but I think it may have to do with the inital sub-cent transaction floods? |
| 03:31 | doublec | so as long as I never send anything in my wallet then the subcent generated transactions can add up? |
| 03:31 | MT`AwAy | doublec: yep |
| 03:31 | Lysacor | www.mtgox.com is one of a few bitcoin investment sites |
| 03:31 | doublec | That's more of an argument that puddinpop should change the server to round to 0.01 |
| 03:31 | ArtForz | well, if your tx had a sub-cent output, it'd cost 0.01 in f ees |
| 03:31 | MT`AwAy | and once you reach a cent, it'll stay |
| 03:31 | ArtForz | whats better? paying < 0.01 or paying 0.01 ? |
| 03:32 | MT`AwAy | ArtForz: he can just keep accumulating sub cents until it gets to 0.01 |
| 03:32 | EvanR | sub cent makes sense |
| 03:32 | MT`AwAy | or until someone modifies the client to not throw away subcents |
| 03:32 | ArtForz | well, if your client uses it as a input, it has reason to |
| 03:32 | EvanR | how hard is it to make your own client? |
| 03:33 | MT`AwAy | EvanR: I'm doing it, and it's not that simple |
| 03:33 | ArtForz | if you have 0.075 and send 0.05 you keep the 0.025 |
| 03:33 | MT`AwAy | ArtForz: actually, you don't |
| 03:33 | MT`AwAy | if you have 0.075 and send 0.05, you'll keep 0.02 |
| 03:33 | ArtForz | err, the throw-away-subcent code only kicks in if a output is < 0.01 |
| 03:33 | EvanR | i got 0.05 coins! |
| 03:33 | nanotube | looks like there's need for a patch to undo the throwing away of subcent bitcoins... |
| 03:33 | MT`AwAy | and 0.005 are added as "fee" to the transaction |
| 03:34 | aximilation_ | is this the gui client, or does it apply to bitcoind as well? |
| 03:34 | aximilation_ | since my balance shows 0.05000000 |
| 03:35 | Lysacor | brb gonna put my computer on a kill-a-watt meter and see what the damage to my electric bill is gonna be... |
| 03:35 | EvanR | i got coins from 'someone' it says unconfirmed. im 'pretty sure' its bitfaucet. how do i tell the client to identify that address as that? |
| 03:36 | MT`AwAy | found one |
| 03:36 | MT`AwAy | http://blockexplorer.com/tx/19daf081d3139e8fa11a2ae6467098312f1e86fe3a9e9e8b81a26e68b0f4f145 <- 50 btc sent, the input also includes subcents |
| 03:36 | EvanR | oh thats not what unconfirmed means |
| 03:36 | ArtForz | yes |
| 03:36 | MT`AwAy | EvanR: you need to wait, the transaction will get confirmed |
| 03:36 | ArtForz | and the 2nd output would'Ve been < 0.01 |
| 03:37 | MT`AwAy | the bitcoin client could have just kept the sub-cent part on the initial wallet |
| 03:37 | ArtForz | if he had sent 45, he'd have gotten 5.0049... in change back |
| 03:37 | MT`AwAy | since it doesn't systematically empty addresses |
| 03:38 | ArtForz | no it couldnt have |
| 03:38 | ArtForz | that was from a 50.0049... generation |
| 03:38 | ArtForz | if it only has that 1 tx, and you want to send 50, whats it suppsoed to do? |
| 03:38 | ArtForz | a) send 50, send 0.0049 back to yourself, needs a 0.01 fee |
| 03:39 | MT`AwAy | just send 50.00000000, and let the 0.0049... part on the address |
| 03:39 | ArtForz | b) send 50, throw away the 0.0049 as fee |
| 03:39 | MT`AwAy | s/let/keep/ |
| 03:39 | ArtForz | it's a SINGLE TRANSACTION |
| 03:39 | ArtForz | you can't KEEP part of an output |
| 03:39 | ArtForz | you have to use the whole output |
| 03:39 | MT`AwAy | you can keep part of the input on the address |
| 03:39 | MT`AwAy | nothing requires the "in" address to be emptied each time |
| 03:39 | EvanR | what does 'throw away' mean, lost to oblivion? or sender keeps it |
| 03:39 | ArtForz | ARGH |
| 03:39 | ArtForz | the input was this: http://blockexplorer.com/tx/7ff761d4749f64c41d1ba7018841b863a0f1b56fe1d62a8369ed452216c421ab#o0 |
| 03:39 | MT`AwAy | EvanR: whoever solves the block gets it |
| 03:40 | ArtForz | SINGLE OUTPUT, 50.0049... |
| 03:40 | ArtForz | user requests a send of 50 to some address |
| 03:40 | EvanR | thats strange |
| 03:40 | ArtForz | so whats his client supposed to do then? |
| 03:40 | MT`AwAy | ArtForz: what I say is to /not/ inclue the 0.0049.. part in the input |
| 03:40 | MT`AwAy | include* |
| 03:40 | ArtForz | you can't do that |
| 03:40 | MT`AwAy | you can |
| 03:41 | ArtForz | no you can't |
| 03:41 | ArtForz | once a output is used, it's marked as spent |
| 03:41 | ArtForz | read the fine code |
| 03:41 | nanotube | ArtForz: so if even one of the outputs of a single tx is <0.01, the whole tx requires a fee? |
| 03:41 | ArtForz | yep |
| 03:41 | MT`AwAy | ArtForz: I'm talking about the input |
| 03:42 | MT`AwAy | if you got 500 bitcoins on address X, you can either put the whole thing as input, or only a part |
| 03:42 | EvanR | bitfaucet is funny. TANSTAAFL was almost violated. until i realized you cant get lunch on 0.05 coins |
| 03:42 | nanotube | ArtForz: i guess in that case it's fine. since it doesn't make sense to spend .01 in order to save <0.1 |
| 03:42 | MT`AwAy | the remaining balance stays on X |
| 03:42 | nanotube | <.01 i mean |
| 03:42 | nanotube | EvanR: haha yea |
| 03:42 | ArtForz | the INPUT to the 50+0.0049fee TX was a SINGLE 50.0049 OUTPUT |
| 03:42 | MT`AwAy | oops |
| 03:43 | ArtForz | if you have multiple outputs to the same pubkey available, you can obviously use only one of them |
| 03:43 | MT`AwAy | forgot input reference tx |
| 03:43 | EvanR | how long does it take to confirm? |
| 03:43 | MT`AwAy | EvanR: ~1 hour |
| 03:43 | EvanR | >_< |
| 03:44 | ArtForz | one or more I mean |
| 03:44 | EvanR | it says 3/unconfirmed now |
| 03:44 | MT`AwAy | but you can use your coins with only one confirmation |
| 03:44 | MT`AwAy | EvanR: you can already use the coins then |
| 03:44 | EvanR | what does the number mean |
| 03:44 | MT`AwAy | EvanR: it's the number of blocks after your transaction |
| 03:44 | ArtForz | and iirc the current client code only throws away < 0.01 amounts if it'd otherwise have to send them back to itself as change, incurring the 0.01 fee |
| 03:44 | MT`AwAy | the more blocks there are, the more you know the transaction is legit |
| 03:45 | ArtForz | so you can have a tx of 50.0049 -> 45 + 5.0049 change |
| 03:45 | EvanR | is there an arbitrary threshold where it 'gives up' and says confirmed |
| 03:45 | ArtForz | I dont think it's smart enough to combine multiple sub-cent containign inputs to make a whole cent |
| 03:45 | aximilation_ | anyone work with the market like mtgox? |
| 03:46 | aximilation_ | or use it to buy/sell? |
| 03:46 | MT`AwAy | ArtForz: I guess it's still better to keep subcent until bitcoin becomes smart enough :) |
| 03:46 | ArtForz | so if you have 50.006 and 50.005 and send 50, it uses one of them and throws away the 0.005 or 6 |
| 03:46 | aximilation_ | or do you do most of your saving my mining? |
| 03:46 | EvanR | what happens if you 'open' your wallet on multiple computers at once |
| 03:46 | aximilation_ | s/my/by |
| 03:46 | ArtForz | instead of doing (50.005 + 50.006) -> 50 + 50.011 change |
| 03:46 | MT`AwAy | EvanR: you see your balance multiple times, but can use it only once |
| 03:47 | EvanR | then what |
| 03:47 | MT`AwAy | EvanR: then you can't send more coins |
| 03:47 | EvanR | on either? |
| 03:47 | MT`AwAy | because in reality you don't have more |
| 03:48 | EvanR | will it segfault, signal alert ERROR, or explain thoroughly the problem ;) |
| 03:48 | MT`AwAy | I think the transaction will just disappear or never get confirmed, I'm not sure (didn't look into that yet) |
| 03:49 | MT`AwAy | one thing sure is the fact the network will not accept that |
| 03:49 | ArtForz | the balance on the other nodes shoudl disappear when they see the tx in a block |
| 03:49 | ArtForz | if they try to send a tx before, network will just drop it as a double.-spend |
| 03:49 | EvanR | really? thats cool |
| 03:50 | annodomini | So I've solved my first block. Now what to spend my ?50 on? |
| 03:50 | Lysacor | ok, so my wallet won't be burning, very nice! there are plenty of places to go :P |
| 03:50 | EvanR | when i got the bitfauct coins, it changed my address. im confused |
| 03:50 | MT`AwAy | annodomini: you can send sms on https://smsz.net/ for example :D |
| 03:50 | EvanR | now i have two receiving addresses |
| 03:50 | MT`AwAy | EvanR: you can have as many addresses as you want |
| 03:51 | Lysacor | your address book allows you to create new addresses, and your address book let's you choose what address you want to receive on |
| 03:51 | EvanR | are they all for one balance? |
| 03:51 | Lysacor | it allows you to be anonymous |
| 03:51 | MT`AwAy | annodomini: you can also donate for the EFF |
| 03:51 | MT`AwAy | EvanR: yes |
| 03:51 | EvanR | ok |
| 03:51 | EvanR | but i can also reuse addresses |
| 03:51 | nanotube | ArtForz: well, sounds relatively reasonable. |
| 03:51 | MT`AwAy | EvanR: yes, but not reusing addresses is better |
| 03:51 | EvanR | why? |
| 03:51 | annodomini | MT`AwAy: True. Just donated to them in $ a couple days ago, and might again tonight via the Humble Indie Bundle. |
| 03:51 | MT`AwAy | EvanR: it protects your privacy |
| 03:51 | EvanR | hmm. |
| 03:52 | EvanR | yes |
| 03:52 | EvanR | but |
| 03:52 | MT`AwAy | and whoever's giving bitcoins to you's privacy too |
| 03:52 | nanotube | EvanR: there's a thread on the forum about 'pros and cons of reusing addresses'. read it. |
| 03:52 | EvanR | ok |
| 03:52 | MT`AwAy | also you can create addresses specifically for something (for example on my site, each order gets its own bitcoin address so when people pay, I know which order they paid for) |
| 03:53 | EvanR | seems like id lose track without labels |
| 03:54 | EvanR | i guess i could name each one as they are used |
| 03:55 | MT`AwAy | I guess when you generate coins you end using one address per generated block too |
| 03:55 | EvanR | when address? |
| 03:55 | EvanR | erm |
| 03:55 | EvanR | which address |
| 03:56 | EvanR | the one in the box? |
| 03:56 | ArtForz | extra credit for anyone who got the Titan AE reference |
| 03:57 | nanotube | not me... no extra credit for nanotube :( |
| 03:58 | EvanR | is there a mastercard/paypal blessed website where people can send or donate money to someones bitcoin ? |
| 03:58 | nanotube | MT`AwAy: yes, every block gen has an address in it, which by default is a new unused address. (though it doesn't have to be) |
| 03:58 | nanotube | EvanR: what does mastercard have to do with donating bitcoins? |
| 03:58 | EvanR | most people dont have bitcoins to send or donate |
| 03:59 | EvanR | they have usd or euros or something |
| 03:59 | noagendamarket | bitcoingateway for credit cards |
| 03:59 | nanotube | EvanR: --^ bitcoingateway.com you can use cc to buy btc |
| 03:59 | nanotube | EvanR: or you can deposit to mtgox.com via bank wire.... |
| 04:00 | nanotube | or you can trade for paypal and other various and sundry things on #bitcoin-otc |
| 04:00 | EvanR | otc sounds more like it |
| 04:00 | EvanR | hmm |
| 04:01 | nanotube | well, join the channel, and have fun. :) |
| 04:01 | EvanR | so someone paying with bitcoins must first set it up and exchange for it first |
| 04:01 | MT`AwAy | oh, btc is at 0.247 on mtgox :o |
| 04:01 | EvanR | first |
| 04:01 | gribble | {"ticker":{"high":0.2468,"low":0.21,"vol":10149,"buy":0.2152,"sell":0.2459,"last":0.2467}} |
| 04:01 | nanotube | ;;bc,mtgox |
| 04:01 | EvanR | what is "vol" ? |
| 04:01 | nanotube | MT`AwAy: yea looks like someone bought some. |
| 04:01 | nanotube | volume |
| 04:01 | EvanR | whats volume |
| 04:01 | nanotube | last 24 hours |
| 04:01 | nanotube | how many bitcoins were exchanged on mtgox |
| 04:01 | EvanR | ah |
| 04:01 | EvanR | not even google could explain that |
| 04:01 | EvanR | too simple |
| 04:02 | nanotube | heh... well it could be either volume or volatility |
| 04:02 | nanotube | and the number is way too large to be volatility |
| 04:02 | nanotube | so it must be volume. :) |
| 04:02 | nanotube | so you /could/ have guessed. :P |
| 04:02 | MT`AwAy | only people used to markets will know what a normal volatility is |
| 04:03 | MT`AwAy | :p |
| 04:03 | EvanR | heres whats funny, digital data in principle can be copied exactly and be reproduced indefinitely right? yet storing money on a hard drive seems orders of magnitude more risky :S |
| 04:03 | nanotube | well... or people who know what a standard deviation for numbers in the range of .22 looks like :P |
| 04:03 | nanotube | EvanR: how so? just make backups. |
| 04:03 | MT`AwAy | so volatility is stddev |
| 04:03 | nanotube | EvanR: after all... your bank stores money on their hard drives. :) |
| 04:03 | EvanR | you dont make backups of your wallet? |
| 04:04 | noagendamarket | the fed prints money wsith a button |
| 04:04 | noagendamarket | lol |
| 04:04 | EvanR | no!!! |
| 04:04 | nanotube | EvanR: i mean, when you store money on the hd, you make backups, so you don't lose it if your hd goes boom. |
| 04:04 | nanotube | noagendamarket: haha |
| 04:05 | EvanR | yes |
| 04:05 | nanotube | EvanR: it is in fact a very smart move. as long as you encrypt your wallet first. :) |
| 04:06 | EvanR | ah |
| 04:06 | noagendamarket | you can email it to me and Ill store it for you lol |
| 04:06 | noagendamarket | :)- |
| 04:06 | EvanR | ok! |
| 04:06 | MT`AwAy | xD |
| 04:06 | EvanR | exchange rate around 0.250, does that mean a bitcoin is about a quarter USD or the other way around |
| 04:07 | MT`AwAy | 1 BTC ~= 0.25 USD |
| 04:07 | MT`AwAy | right now |
| 04:07 | MT`AwAy | tommorow might be 1 BTC = 0.01 USD, you can't tell :p |
| 04:07 | EvanR | haha |
| 04:08 | EvanR | you can tell that from a volume of 10000 ? |
| 04:08 | MT`AwAy | EvanR: there are too many unknowns |
| 04:08 | EvanR | sounds like room for speculation! |
| 04:08 | MT`AwAy | if someone finds a problem in bitcoin and proves that the whole concept doesn't hold, btc will suddently have no value at all |
| 04:09 | nanotube | or, if someone with deep pockets decides to diversify into bitcoins and puts like 20k usd in... it may be 50c/btc tomorrow. |
| 04:09 | EvanR | is it possible to buy that many bitcoins? |
| 04:10 | MT`AwAy | EvanR: probably going to take a bit of time |
| 04:10 | nanotube | EvanR: yes... see the outstanding orders on mtgox |
| 04:10 | EvanR | ill have to checkout thissite out extensively |
| 04:10 | nanotube | EvanR: go to bitcoincharts.com - it has a nice display of the open order book |
| 04:10 | nanotube | if you're willing to pay up to 50c per btc, you can buy >70 thousand btc. |
| 04:11 | nanotube | (and that's not counting the dark pool orders) |
| 04:11 | EvanR | but theres only like 1 million total right now? |
| 04:12 | EvanR | youd own a tenth of them ;) |
| 04:12 | EvanR | should i be buying low? xD |
| 04:12 | ArtForz | more like 5M total |
| 04:13 | noagendamarket | someone is selling a house for btc |
| 04:13 | annodomini | 4.88 M according to http://www.bitcoinwatch.com/ |
| 04:13 | EvanR | damn theres a lot of bitcoin*.foos |
| 04:14 | kiba | look like the price for bitcoin is finally going up |
| 04:14 | annodomini | It seems like the thing to do these days; take internet business idea, add "bitcoin" to the beginning ... profit! |
| 04:15 | EvanR | annodomini: and thats why the price is going up |
| 04:15 | sgornick | ArtForz: I think EvanR was referring to $1 million (US) now. |
| 04:15 | kiba | 1.2 million dollars |
| 04:15 | EvanR | today was a special day, i think bitcoin somehow got publicity or something |
| 04:15 | kiba | not a special day |
| 04:15 | kiba | but a special month |
| 04:16 | ArtForz | oh |
| 04:16 | kiba | we got on Wikipedia, EFF promoted us, We got Wikileak and bitcoin linked |
| 04:16 | EvanR | that would suggest independent parties decided 'the concept' was pretty good |
| 04:17 | kiba | we're moving out of the ignorance stage |
| 04:17 | sgornick | EvanR: But lets say you were one of those learned about it because of Wikipedia, Wikileak, EFF, PC World, etc. .... how would you, three or four days later, come into having Bitcoin in your wallet? |
| 04:17 | kiba | to violent opposition |
| 04:17 | kiba | once we complete violent opposite |
| 04:17 | kiba | then it's acceptance |
| 04:17 | EvanR | sgornick: too late i already do! |
| 04:17 | EvanR | 0.05 ;) |
| 04:17 | sgornick | (s) plural |
| 04:17 | EvanR | i guess continue to generate blocks |
| 04:18 | EvanR | or become inspired and start a .com |
| 04:18 | sgornick | Got GPUs? |
| 04:18 | Diablo-D3 | http://www.engadget.com/2010/12/15/amd-radeon-hd-6970-and-hd-6950-launch-assault-on-enthusiast-gami/ |
| 04:18 | Diablo-D3 | official reviews are out |
| 04:19 | ArtForz | SHINY! |
| 04:20 | Diablo-D3 | http://www.hardocp.com/article/2010/12/14/amd_radeon_hd_6970_6950_video_card_review/8 |
| 04:20 | Diablo-D3 | fuckers need to learn to do math |
| 04:20 | ArtForz | 1536SP @ 880MHz, useless for mining |
| 04:20 | EvanR | how does a market work, i never understood it. we set our own prices? |
| 04:20 | Diablo-D3 | if they know the total system wattage is 190, SUBTRACT IT |
| 04:20 | nanotube | sgornick: you buy some on -otc or mtgox or bitcoingateway |
| 04:20 | Diablo-D3 | 6970 uses 238 watts max |
| 04:21 | Diablo-D3 | slightly more than the 5870 |
| 04:21 | ArtForz | 6970 needs more power than 5870, barely faster than a 5850 |
| 04:21 | nanotube | EvanR: same way any market works. there's a range of buy orders, and a range of sell orders... when they come together, trading occurs. |
| 04:21 | Diablo-D3 | but apparently rapes it on 3D |
| 04:21 | Diablo-D3 | LOL 12 idle watts |
| 04:21 | EvanR | nanotube: occurs? |
| 04:21 | Diablo-D3 | jesus |
| 04:21 | ArtForz | still FAIL for mining |
| 04:21 | EvanR | nanotube: what i dont understand is who is deciding what |
| 04:22 | ArtForz | and seems to run hot as hell |
| 04:22 | Diablo-D3 | yuck, full load reaches 89c |
| 04:22 | nanotube | EvanR: simple case: say i want to buy 100 bitcoins, and i'm willing to pay .20 usd/btc. someone else is looking to sell, and is willing to sell at .25/btc |
| 04:22 | nanotube | EvanR: if there's nothing else happening, then no trading occurs. |
| 04:22 | ArtForz | I bet the fan never goes >60% on auto |
| 04:23 | EvanR | nanotube: ok, then what? |
| 04:23 | nanotube | EvanR: now, some other guy comes in, and he really wantsn some btc... so he's willing to buy at .25. so he comes in and snaps up some of the seller's coins at .25/btc |
| 04:23 | nanotube | EvanR: and voila, a trade has occurred. it's just like when you go to a farmer's market and want to buy some tomatoes. there are guys selling... at various price points... |
| 04:23 | ArtForz | wait a sec |
| 04:23 | ArtForz | the slide changed |
| 04:23 | EvanR | so its like ham radio, people have to agree on a frequency or they cant hear each other |
| 04:23 | ArtForz | we got 4 32-bit adds per clock again |
| 04:24 | nanotube | EvanR: you go around, and if you really want your tomatoes, you buy from one of the vendors. if you think they're all to expensive, you haggle until you agree on a price, or walk away. |
| 04:24 | Diablo-D3 | ArtForz: so we're back in business? |
| 04:24 | sgornick | > #-otc? Assuming the person has registered their nickname weeks ago, which is the prerequisite to place an order. |
| 04:24 | ArtForz | kinda |
| 04:24 | ArtForz | should get 5970 about == 5870 for mining |
| 04:24 | Diablo-D3 | you mean 6970 |
| 04:24 | ArtForz | yea |
| 04:24 | ArtForz | 6970 |
| 04:25 | nanotube | sgornick: well, to place it officially on the book, yes. but you can also just shout out to channel. willing sellers may verify your identity, and sell to you (possibly at a bit of a risk premium) |
| 04:25 | EvanR | nanotube: so there might be several 'prices' where trades are occuring, and if the people paying higher know about all of them, they will all go to the lowest? |
| 04:25 | ArtForz | and VLIW4 should get us a bit better VLIW usage |
| 04:25 | ArtForz | = might end up 5% faster than 5870 |
| 04:26 | nanotube | EvanR: yes, there may be some price differences between the various markets, depending on costs of entry, counterparty risk, etc. |
| 04:26 | Diablo-D3 | and uses about 5% more watts than a 5870 |
| 04:26 | nanotube | EvanR: but say in case of "the mtgox market"... trading only occurs inside the bid/ask spread. |
| 04:26 | ArtForz | yep |
| 04:26 | Diablo-D3 | ArtForz: and costs about 5% more |
| 04:26 | EvanR | nanotube: so at a single market, there is only one price? |
| 04:27 | ArtForz | but it means 6990 should end up roughly == 5970 |
| 04:27 | Diablo-D3 | I wonder where they were going with this |
| 04:27 | ArtForz | at least we wont have to go hunting for remaining stock of 58xxs |
| 04:27 | Diablo-D3 | except for the vastly improved tess |
| 04:27 | Diablo-D3 | whats the up side? |
| 04:27 | nanotube | EvanR: assuming uniform goods (like bitcoins), yes. in case of tomatoes, no, because someone might sell better quality tomatoes than someone else. :) |
| 04:27 | EvanR | currency in general is uniform? |
| 04:27 | ArtForz | I have to check detailed specs, but I suspect they also beefed up ROP and TEX units |
| 04:28 | EvanR | im going to the bitcoin-otc channel to see how this works |
| 04:28 | Keefe | but how likely is it that the price of 6990 will be <$500? |
| 04:28 | ArtForz | = 58xx was ALU heavy |
| 04:28 | Diablo-D3 | ArtForz: yeah, but it doesnt seem to be improving fps |
| 04:28 | Diablo-D3 | Keefe: it'll cost as much as the 5970 |
| 04:28 | Keefe | about as much as the 5970 did a few months ago? or now? |
| 04:28 | Diablo-D3 | Keefe: the 6970 costs as much as a 5870, the 6950 costs as much as a 5850 |
| 04:29 | Diablo-D3 | and the 68xx is really pushing the price down |
| 04:29 | ArtForz | because 68xx is the new 57xx |
| 04:29 | Keefe | i'd say a 5970 can be bought for $400 if one is patient |
| 04:29 | Diablo-D3 | well no |
| 04:29 | Diablo-D3 | 68xx is the new 58xx |
| 04:29 | Diablo-D3 | same speed, much lower price |
| 04:30 | Diablo-D3 | and less watts, and smaller physical die size |
| 04:30 | ArtForz | 6970 is maybe 10% faster than 5870 |
| 04:30 | ArtForz | 6870 is a LOT slower |
| 04:30 | Diablo-D3 | not on 3D it isnt |
| 04:30 | Diablo-D3 | the question is |
| 04:30 | da2ce7 | well for generations it counts. |
| 04:30 | Keefe | 5870 is their top single chip in the 5xxx series. 6970 is the top single chip in the 6xxx series, right? |
| 04:30 | ArtForz | yes it is, 6870 is ~5850 |
| 04:30 | ArtForz | yep |
| 04:30 | nanotube | EvanR: yes, currency, stocks, bonds... all uniform, and all usually have "one price" at any given moment in time. |
| 04:30 | Diablo-D3 | if the 69xx doesnt significantly beat 58xx |